CN111369525B - Image analysis method, apparatus and storage medium - Google Patents

Image analysis method, apparatus and storage medium Download PDF

Info

Publication number
CN111369525B
CN111369525B CN202010137683.0A CN202010137683A CN111369525B CN 111369525 B CN111369525 B CN 111369525B CN 202010137683 A CN202010137683 A CN 202010137683A CN 111369525 B CN111369525 B CN 111369525B
Authority
CN
China
Prior art keywords
coronary
image
detected
distance
threshold range
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010137683.0A
Other languages
Chinese (zh)
Other versions
CN111369525A (en
Inventor
董昢
吴迪嘉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Lianying Intelligent Medical Technology Beijing Co ltd
Original Assignee
Lianying Intelligent Medical Technology Beijing Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Lianying Intelligent Medical Technology Beijing Co ltd filed Critical Lianying Intelligent Medical Technology Beijing Co ltd
Priority to CN202010137683.0A priority Critical patent/CN111369525B/en
Publication of CN111369525A publication Critical patent/CN111369525A/en
Application granted granted Critical
Publication of CN111369525B publication Critical patent/CN111369525B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/68—Analysis of geometric attributes of symmetry
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30101—Blood vessel; Artery; Vein; Vascular
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Quality & Reliability (AREA)
  • Radiology & Medical Imaging (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Geometry (AREA)
  • Image Analysis (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

The application relates to an image analysis method, an image analysis device and a storage medium. The method comprises the following steps: detecting the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected; the reference information is related to the position information of an atrium and a ventricle surrounded by coronary artery in the coronary artery image to be detected; dividing the coronary image to be detected to obtain divided images of left and right coronary arteries corresponding to the coronary image to be detected, and extracting central lines of the left and right coronary arteries of the divided images of the left and right coronary arteries; analyzing the coronary image to be detected according to the reference information of the coronary and the central lines of the left and right coronary to obtain an analysis result; the analysis result is used for representing dominant categories of coronary arteries in the coronary artery image to be detected. By adopting the method, the manpower can be saved.

Description

Image analysis method, apparatus and storage medium
Technical Field
The present disclosure relates to the field of image processing technologies, and in particular, to an image analysis method, apparatus, and storage medium.
Background
The heart is required to continuously beat to ensure blood circulation, and the heart is taken as a myogenic organ for pumping blood, and the heart also needs enough nutrition and energy, and the vascular system for supplying nutrition to the heart, namely coronary artery and vein, namely coronary circulation, is the artery for supplying blood to the heart, and is divided into left branch and right branch and runs on the surface of the heart, and the coronary artery can be divided into three types of dominant types according to the source of blood supply vessels (posterior descending branches) of one third of the heart after the interval, namely right dominant type, left dominant type and balanced type. The dominant type is simply that which of the two coronary arteries of the heart bears more blood supply to the myocardium. The treatment means adopted by doctors are different for the coronary distribution of different dominant types, so that the determination of the dominant type of the coronary is particularly important.
In the related art, when judging the dominant type of coronary artery, most of clinicians judge the acquired coronary artery image according to experience so as to obtain the judging result of the dominant type of coronary artery.
However, the above-described technique has a problem of consuming labor.
Disclosure of Invention
In view of the above, it is necessary to provide an image analysis method, apparatus, and storage medium that can save labor.
An image analysis method, the method comprising:
detecting the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected; the reference information of the coronary artery is related to the position information of the atrium and the ventricle surrounded by the coronary artery in the coronary artery image to be detected;
dividing the coronary image to be detected to obtain divided images of left and right coronary arteries corresponding to the coronary image to be detected, and extracting central lines of the left and right coronary arteries of the divided images of the left and right coronary arteries;
analyzing the coronary image to be detected according to the reference information of the coronary and the central lines of the left and right coronary to obtain an analysis result; the analysis result is used for representing dominant categories of coronary arteries in the coronary image to be detected.
In one embodiment, the detecting the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected includes:
Detecting and processing the atrioventricular intersection point in the coronary image to be detected by adopting a preset detection model to obtain the position information of the atrioventricular intersection point in the coronary image to be detected, and taking the position information of the atrioventricular intersection point as the reference information of the coronary;
the detection model is obtained by training based on a first sample coronary image and a gold standard image corresponding to the first sample coronary image, wherein the gold standard image corresponding to the first sample coronary image comprises a position mark of an atrioventricular intersection point corresponding to the first sample coronary image, and the atrioventricular intersection point is a point at the junction of a central vein between ventricles and a coronary sinus of a heart room.
In one embodiment, the midlines of the left and right coronary arteries include a midline of the left coronary artery and a midline of the right coronary artery, and the analyzing the coronary image to be detected according to the reference information of the coronary arteries and the midlines of the left and right coronary arteries to obtain an analysis result includes:
calculating the shortest distance between the position information of the atrioventricular node and the midline of the left coronary artery to obtain a first distance; calculating the shortest distance between the position information of the atrioventricular node and the midline of the right coronary artery to obtain a second distance;
and analyzing the coronary image to be detected according to the first distance, the second distance and the preset distance threshold range to obtain an analysis result.
In one embodiment, the preset distance threshold range includes a first distance threshold range, a second distance threshold range, and a third distance threshold range, and the analyzing the coronary image to be detected according to the first distance, the second distance, and the preset distance threshold range to obtain an analysis result includes:
matching the first distance, the second distance and the first distance threshold range; if the first distance and the second distance are not beyond the first distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is balanced;
or, matching the first distance with the second distance threshold range; if the first distance does not exceed the second distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is left dominant type;
or, matching the second distance with a third distance threshold range; if the second distance does not exceed the third distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is right dominant type.
In one embodiment, the first distance threshold range includes a left Zhi Di distance threshold range and a right branch first distance threshold range, and the matching is performed between the first distance and the second distance and the first distance threshold range; if the first distance and the second distance are not beyond the first distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is an equilibrium type, including:
Matching the first distance with a left Zhi Di distance threshold range, and matching the second distance with a right branch first distance threshold range;
if the first distance does not exceed the left Zhi Di distance threshold range and the second distance does not exceed the right branch first distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is balanced.
In one embodiment, the detecting the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected includes:
carrying out segmentation processing on the coronary image to be detected to obtain a segmentation result of the heart chamber; the segmentation result of the heart chamber comprises left and right atria and left and right ventricles of the heart;
establishing an atrioventricular coordinate system according to a segmentation result of the heart chamber, and taking the atrioventricular coordinate system as reference information of coronary artery; the lateral axis direction of the atrioventricular coordinate system is the direction of the boundary line of the atrioventricular groove of the heart, and the longitudinal axis direction of the atrioventricular coordinate system is the direction of the boundary line of the ventricular groove of the heart.
In one embodiment, the midlines of the left and right coronary arteries include a midline of the left coronary artery and a midline of the right coronary artery, and the analyzing the coronary image to be detected according to the reference information of the coronary arteries and the midlines of the left and right coronary arteries to obtain an analysis result includes:
Calculating the coordinates of each point on the midline of the left coronary artery under the atrioventricular groove coordinate system to obtain the coordinates of at least one left supporting point, and calculating the coordinates of each point on the midline of the right coronary artery under the atrioventricular groove coordinate system to obtain the coordinates of at least one right supporting point;
and analyzing the coronary image to be detected according to the coordinates of at least one left fulcrum, the coordinates of at least one right fulcrum and a preset coordinate threshold range to obtain an analysis result.
In one embodiment, the analyzing the coronary image to be detected according to the coordinates of the at least one left fulcrum, the coordinates of the at least one right fulcrum, and the preset coordinate threshold range to obtain an analysis result includes:
matching the coordinates of at least one left fulcrum and the coordinates of at least one right fulcrum with a coordinate threshold range respectively to obtain the number of left branch target points and the number of right branch target points, comparing the number of left branch target points, the number of right branch target points and a preset number threshold, and obtaining an analysis result according to a comparison result;
the left branch target point is a point of which the coordinates in at least one left supporting point do not exceed the coordinate threshold range, and the right branch target point is a point of which the coordinates in at least one right supporting point do not exceed the coordinate threshold range.
In one embodiment, if the coordinate threshold range includes a left Zhi Di coordinate threshold range and a right branch first coordinate threshold range, the number threshold includes a left Zhi Di number threshold and a right branch first number threshold, the matching the coordinates of at least one left fulcrum and the coordinates of at least one right fulcrum with the coordinate threshold range respectively to obtain the number of left branch target points and the number of right branch target points, comparing the number of left branch target points and the number of right branch target points with a preset number threshold, and obtaining an analysis result according to the comparison result, including:
matching the coordinates of at least one left fulcrum with a left Zhi Di coordinate threshold range to obtain the number of left Zhi Di target points, and matching the coordinates of at least one right fulcrum with a right branch first coordinate threshold range to obtain the number of right branch first target points;
comparing the number of left Zhi Di-target points with a left Zhi Di-number threshold, and comparing the number of right branch first target points with a right branch first number threshold;
if the number of the left Zhi Di first target points is greater than the left Zhi Di first number threshold and the number of the right first target points is greater than the right first number threshold, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is balanced.
In one embodiment, if the coordinate threshold range includes a left Zhi Di two-coordinate threshold range and a right-branch second-coordinate threshold range, the number threshold includes a left-branch second-number threshold and a right-branch second-number threshold, the matching the coordinates of at least one left fulcrum and the coordinates of at least one right fulcrum with the coordinate threshold range to obtain the number of left-branch target points and the number of right-branch target points, comparing the number of left-branch target points and the number of right-branch target points with a preset number threshold, and obtaining the analysis result according to the comparison result, including:
matching the coordinate of at least one left fulcrum with the threshold range of the left Zhi Di two coordinates to obtain the number of left Zhi Di two target points, comparing the number of the left Zhi Di two target points with the threshold of the left Zhi Di two target points, and if the number of the left Zhi Di two target points is larger than the threshold of the left Zhi Di two target points, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is left dominant type; or alternatively, the process may be performed,
and matching the coordinates of at least one right fulcrum with a right branch second coordinate threshold range to obtain the number of right branch second target points, comparing the number of right branch second target points with a right branch second number threshold, and if the number of right branch second target points is larger than the right branch second number threshold, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is right dominant type.
In one embodiment, the segmenting the coronary image to be detected to obtain segmented images of the left and right coronary corresponding to the coronary image to be detected includes:
inputting the coronary image to be detected into a preset segmentation model to obtain a probability map of left and right coronary arteries corresponding to the coronary image to be detected; the pixel value of each position on the probability map of the left and right coronary artery is the probability that the pixel value of the corresponding position on the coronary artery image to be detected belongs to the left and right coronary artery, the segmentation model is obtained by training based on a second sample coronary artery image and a gold standard image corresponding to the second sample coronary artery image, and the gold standard image corresponding to the second sample coronary artery image comprises left and right coronary artery marks corresponding to the second sample coronary artery image;
and carrying out binarization processing on the probability maps of the left and right coronary arteries according to a preset first probability threshold value to obtain segmented images of the left and right coronary arteries corresponding to the probability maps of the left and right coronary arteries.
In one embodiment, the detecting the intersection point of the room in the coronary image to be detected by using the preset detection model to obtain the position information of the intersection point of the room in the coronary image to be detected includes:
inputting the coronary image to be detected into a detection model to obtain a probability map of the intersection points of the rooms corresponding to the coronary image to be detected; the pixel value of each position on the probability graph of the atrioventricular intersection point is the probability that the pixel value of the corresponding position on the coronary image to be detected belongs to the atrioventricular intersection point;
Performing binarization processing on the probability map of the compartment intersection point according to a preset second probability threshold value to obtain a binarization mask image corresponding to the probability map of the compartment intersection point;
marking connected domains in the binarized mask image, and determining the largest connected domain according to the marked connected domains;
and acquiring a weighted center point of a probability value corresponding to the maximum connected domain, and determining the position information of the weighted center point as the position information of the intersection point of the rooms.
An image analysis apparatus, the apparatus comprising:
the detection module is used for carrying out detection processing on the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected; the reference information of the coronary artery is related to the position information of the atrium and the ventricle surrounded by the coronary artery in the coronary artery image to be detected;
the segmentation module is used for carrying out segmentation processing on the coronary image to be detected to obtain segmented images of left and right coronary arteries corresponding to the coronary image to be detected, and extracting central lines of the left and right coronary arteries of the segmented images of the left and right coronary arteries;
the analysis module is used for analyzing the coronary image to be detected according to the reference information of the coronary and the central lines of the left and right coronary so as to obtain an analysis result; the analysis result is used for representing dominant categories of coronary arteries in the coronary image to be detected.
A computer device comprising a memory storing a computer program and a processor which when executing the computer program performs the steps of:
detecting the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected; the reference information of the coronary artery is related to the position information of the atrium and the ventricle surrounded by the coronary artery in the coronary artery image to be detected;
dividing the coronary image to be detected to obtain divided images of left and right coronary arteries corresponding to the coronary image to be detected, and extracting central lines of the left and right coronary arteries of the divided images of the left and right coronary arteries;
analyzing the coronary image to be detected according to the reference information of the coronary and the central lines of the left and right coronary to obtain an analysis result; the analysis result is used for representing dominant categories of coronary arteries in the coronary image to be detected.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:
detecting the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected; the reference information of the coronary artery is related to the position information of the atrium and the ventricle surrounded by the coronary artery in the coronary artery image to be detected;
Dividing the coronary image to be detected to obtain divided images of left and right coronary arteries corresponding to the coronary image to be detected, and extracting central lines of the left and right coronary arteries of the divided images of the left and right coronary arteries;
analyzing the coronary image to be detected according to the reference information of the coronary and the central lines of the left and right coronary to obtain an analysis result; the analysis result is used for representing dominant categories of coronary arteries in the coronary image to be detected.
According to the image analysis method, the device, the computer equipment and the storage medium, the coronary artery image to be detected is detected to obtain the reference information of the coronary artery, the reference information of the coronary artery is related to the position information of the heart chambers and the atria surrounded by the coronary artery, the coronary artery image to be detected is divided and the midline is extracted to obtain the midline of the left and right coronary artery, and the coronary artery image to be detected is analyzed based on the reference information of the coronary artery and the midline of the left and right coronary artery to obtain the analysis result which can represent the dominant type of the coronary artery. In the method, the coronary image can be detected, and the dominant type of the coronary can be obtained based on the obtained reference information of the coronary and the central lines of the left and right coronary, and the dominant type is not needed to be judged manually, so that the method can save the labor cost to a certain extent; in addition, in the method, the coronary artery image is detected by the computer equipment to obtain the dominant type category, so that compared with the method of manually judging the dominant type category according to experience, the obtained dominant type category judging result is more accurate, and meanwhile, the dominant type category judging process is faster, so that the dominant type category judging time can be saved.
Drawings
FIG. 1 is an internal block diagram of a computer device of one embodiment;
FIG. 2 is a flow chart of an image analysis method according to an embodiment;
FIG. 3a is a schematic illustration of a labeled atrioventricular intersection on a first sample coronary image;
FIG. 3b is a flow chart of an image analysis method according to another embodiment;
FIG. 4 is a flow chart of an image analysis method according to another embodiment;
FIG. 5a is a flow chart of an image analysis method according to another embodiment;
FIG. 5b is a schematic illustration of an atrioventricular coordinate system established over a heart chamber;
FIG. 6 is a flow chart of an image analysis method according to another embodiment;
FIG. 7a is a flow chart of an image analysis method according to another embodiment;
FIG. 7b is a schematic illustration of left and right coronary vessels marked on a second sample coronary image;
fig. 8 is a block diagram showing the structure of an image analysis apparatus in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
Coronary arteries can be classified into three types according to the sources of blood supply vessels (posterior descending branches) of the rear third of the ventricular septum, right dominant type, left dominant type, and balanced type, namely right dominant type if the posterior descending branch blood supply vessels come from Right Coronary Artery (RCA); if the posterior descending branch blood supply vessel comes from the left circumflex coronary artery (LCX), it is left dominant; a posteriorly descending branch of blood vessel is balanced if it comes from both the Right Coronary Artery (RCA) and the left circumflex coronary artery (LCX). Currently, in clinical work, doctors determine the left and right dominance of coronary arteries based on the origin of posterior descending branches and the distribution of left and right coronary arteries at the septal plane. The right coronary artery of right side dominance crosses the crisscross, sends out the posterior descending branch on the septum surface, and sends out the right posterior lateral branch; the left dominant posterior descending branch is emitted by the left crown circumflex branch and the left crown circumflex across the crisscross; the balanced left and right coronary arteries are distributed on the heart septum surface in an balanced way and do not cross each other across the intersection of the atrioventricular node, and the posterior descending branch can be sent out by the right coronary artery or come from the two side coronary arteries. The judgment method has higher requirements on the anatomical experience of doctors, particularly for the situation of great variation of coronary artery blood vessels, the manual experience is very important, and primary doctors often have the situation of misjudgment due to insufficient experience of reading; in addition, manual judgment, especially when a doctor processes reports in a large amount, requires great effort of the doctor, and increases the risk of increasing human misjudgment. The application provides an image detection method, an image detection device, image detection equipment and a storage medium, which can solve the technical problems.
The image detection method provided by the embodiment can be applied to computer equipment, wherein the computer equipment can be a terminal or a server, and the computer equipment can be in wired or wireless communication with medical scanning equipment. Taking a computer device as an example, an internal structure diagram of the computer device may be shown in fig. 1. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement an image detection method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the structure shown in fig. 1 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
The execution subject of the embodiments of the present application may be a computer device or an image detection device, and the following description will refer to the computer device as the execution subject.
In one embodiment, an image detection method is provided, and this embodiment relates to a specific process of how to detect and analyze a coronary image, and obtain a coronary dominant type category based on a coronary image analysis result. As shown in fig. 2, the method may include the steps of:
s202, detecting and processing the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected; the reference information of the coronary is related to the position information of the atrium and ventricle surrounded by the coronary in the image of the coronary to be detected.
The atrium and ventricle surrounded by the coronary artery generally has a left atrium, a left ventricle, a right atrium and a right ventricle, and the reference information may be information related to the position information of the left atrium and the right atrium and the position information of the left ventricle and the right ventricle, for example, may be the intersection position between the ventricle and the atrium, or the coordinate system of the ventricle and the atrium established according to the position of the ventricle and the position of the atrium, or may be other information related to the position of the atrium of the ventricle, which is not limited in this embodiment. In addition, the coronary image to be detected may be a one-dimensional image, a two-dimensional image, a three-dimensional image, a four-dimensional image, or the like, and the three-dimensional image is mainly used in this embodiment, and the mode of the coronary image to be detected may be a CT image, a PET image, an MR image, or the like, for example, a CTA image when the coronary image to be detected is a CT image; the coronary in the image to be detected generally includes both left and right coronary.
Specifically, before the object to be detected is detected, a scanning device (not limited to a scanning device such as CT, PET, MR) can be used for scanning the chest or the whole body of the object to be detected to obtain an image of the chest of the image to be detected, wherein the chest image comprises coronary arteries and hearts and can be recorded as a coronary artery image to be detected; then, the computer equipment can detect the coronary artery and the heart in the coronary artery image to be detected, obtain the position information related to the coronary artery and the ventricular atrium, and record the position information as the reference information of the coronary artery.
S204, segmenting the coronary image to be detected to obtain segmented images of left and right coronary arteries corresponding to the coronary image to be detected, and extracting central lines of the left and right coronary arteries of the segmented images of the left and right coronary arteries.
When the coronary artery to be detected is segmented, a segmentation model, an image segmentation algorithm and the like can be adopted to segment the coronary artery in the coronary artery to be detected, so that left and right coronary arteries are obtained. After obtaining the coronary image to be detected, the coronary image to be detected may be subjected to segmentation processing first, and then the detection processing of S202 described above may be performed; of course, the detection processing of S202 may be performed first, and then the segmentation processing of this step may be performed; of course, the detection process of S202 and the segmentation process of the present step may be performed simultaneously, and the sequence of the steps of S202 and the present step is not particularly limited in this embodiment. In addition, when the midlines of the left and right coronary arteries are extracted, the extraction may be performed by a skeletonizing method, or may be performed by a neural network, or may be performed by another method, which is not particularly limited in this embodiment. In addition, the midline of the left and right coronary arteries may be the centerline of the left and right coronary arteries, or may be a line that may characterize the left and right coronary arteries, such as a line in the left and right coronary arteries that is near the outer wall.
Specifically, after the coronary image to be detected is obtained in S202, the computer device may divide the coronary in the coronary to be detected by using a division model, an image division algorithm, or the like, to obtain a divided image of the left coronary and a divided image of the right coronary, and then extract the midline of the divided image of the left coronary and the midline of the divided image of the right coronary, respectively, to obtain the midline of the left coronary corresponding to the divided image of the left coronary and the midline of the right coronary corresponding to the divided image of the right coronary.
S206, analyzing the coronary image to be detected according to the reference information of the coronary and the central lines of the left and right coronary, so as to obtain an analysis result; the analysis result is used for representing dominant categories of coronary arteries in the coronary image to be detected.
The dominant types of coronary artery include equilibrium type, left dominant type and right dominant type.
Specifically, after obtaining the reference information of the coronary artery and the midline of the left coronary artery and the midline of the right coronary artery, the computer device may obtain an analysis result by analyzing the relative position between the position of the atrium of the reference information center of the coronary artery and the midline of the left and right coronary arteries, and obtain the dominant type according to the analysis result. For example, if the position of the reference information center atrium of the coronary artery and the center line of the left coronary artery are relatively close, the left dominant type may be considered, if the position of the reference information center atrium of the coronary artery and the center line of the right coronary artery are relatively close, the right dominant type may be considered, and if the position of the reference information center atrium of the coronary artery and the center line of the left and right coronary arteries are relatively close, the balanced type may be considered. After the dominant type category of the coronary artery in the coronary artery image to be detected is obtained, the corresponding dominant type category can be marked on the reading sheet of the object to be detected, and meanwhile, a doctor can process the object to be detected according to the dominant type category by adopting a targeted processing means so as to obtain a better processing result.
In the image analysis method, the coronary artery image to be detected is detected to obtain the reference information of the coronary artery, the reference information of the coronary artery is related to the position information of the ventricles and the atria surrounded by the coronary artery, the coronary artery image to be detected is segmented, the midline is extracted, the midline of the left and right coronary artery is obtained, and the coronary artery image to be detected is analyzed based on the reference information of the coronary artery and the midline of the left and right coronary artery to obtain the analysis result which can represent the dominant type of the coronary artery. In the method, the coronary image can be detected, and the dominant type of the coronary can be obtained based on the obtained reference information of the coronary and the central lines of the left and right coronary, and the dominant type is not needed to be judged manually, so that the method can save the labor cost to a certain extent; in addition, in the method, the coronary artery image is detected by the computer equipment to obtain the dominant type category, so that compared with the method of manually judging the dominant type category according to experience, the obtained dominant type category judging result is more accurate, and meanwhile, the dominant type category judging process is faster, so that the dominant type category judging time can be saved.
It should be noted that, two types of reference information may be used to obtain the coronary artery, one is the position information of the intersection of the atrioventricular node, the other is the atrioventricular coordinate system, and the dominant type of the coronary artery is analyzed according to the reference information in the two cases. The case where the reference information is the positional information of the intersection of the atrioventricular node will be described first.
In another embodiment, another image detection method is provided, and this embodiment relates to a specific process of detecting a coronary image to be detected by using a detection model to obtain position information of an atrioventricular intersection in the coronary image. On the basis of the above embodiment, the step S202 may include the following step a:
step A, detecting and processing an atrioventricular intersection point in a coronary image to be detected by adopting a preset detection model to obtain the position information of the atrioventricular intersection point in the coronary image to be detected, and taking the position information of the atrioventricular intersection point as the reference information of the coronary; the detection model is obtained by training based on a first sample coronary image and a gold standard image corresponding to the first sample coronary image, wherein the gold standard image corresponding to the first sample coronary image comprises a position mark of an atrioventricular intersection point corresponding to the first sample coronary image, and the atrioventricular intersection point is a point at the junction of a central vein between ventricles and a coronary sinus of a heart room.
In step a, the central vein refers to a longitudinal vein between the left ventricle and the right ventricle of the cardiac septum, the coronary sinus refers to a longitudinal vein sinus between the left atrium and the right atrium of the cardiac septum, the cardiac septum refers to the back of the heart, and there may be multiple points at the junction of the central vein and the coronary sinus, one may be selected from the points as an atrioventricular intersection, and as to which point at the junction is selected, it may be determined according to the actual situation.
The position information of the atrioventricular intersection point may be coordinates of the atrioventricular intersection point, and the coordinates of the atrioventricular intersection point may be one-dimensional coordinates, two-dimensional coordinates, three-dimensional coordinates, or the like.
In addition, before using the detection model, the detection model needs to be trained, and the training process of the detection model can include the following three steps:
1) A training data set is generated. Marking N atrioventricular intersections P in N first sample coronary images using labeling software i And preserving its coordinates (where i refers to the atrioventricular intersection on each first sample coronary imageMarking an atrioventricular intersection point on each first sample coronary image in the range of 1-N, then generating N spherical binarization mask images with the coordinates of each atrioventricular intersection point as the circle center and the radius r in N first sample coronary images, and marking the spherical binarization mask images as gold standard images corresponding to each first sample coronary image (namely, each first sample coronary image corresponds to one gold standard image), wherein the gold standard images and the first sample coronary images are paired to form a training data set, and the number of the training data sets is N; in addition, the mode of the first sample coronary image is the same as the mode of the coronary image to be detected, the r and N sizes can be determined according to practical situations, for example, r can be 6 pixels, and the N number can be 1000, 2000, etc. For example, referring to fig. 3a, there is shown a schematic view of the atrioventricular node marked on the first sample coronary image, and the four figures are the atrioventricular node marked on the sagittal plane, the coronal plane, the horizontal plane, and the three-dimensional solid, respectively (it should be noted that these four figures are only examples and do not affect the essential content of the embodiments of the present application).
2) A detection model is built, which may be a convolutional neural network model. Constructing a convolutional neural network model, and setting super parameters of the convolutional neural network model, wherein an input channel of the convolutional neural network is 1, is a first sample coronary image, and an output channel of the convolutional neural network is 2, and is a detection probability map of N atrioventricular intersection points corresponding to the first sample coronary image and a detection probability map of a background respectively; in addition, the training data set can be divided into a training set X1, a verification set X2 and a test set X3, wherein the training set, the verification set and the test set are mutually independent, the number of the training sets, the verification set and the test set is N1, N2 and N3 respectively, the training data set, the verification set and the test set are natural numbers, and n1+n2+n3=N, and n1 is more than or equal to 1/2N; for example, assuming that the number of first sample coronary images is 1000, n1=500, n2=200, n3=300 may be possible.
3) And training a detection model. Training the detection model established in the step 2) by using the training data set generated in the step 1), wherein a training set X1 is used for training the detection model, a verification set X2 is used for evaluating the current performance of the model, and a test set X3 is used for checking the generalization performance of the model; in the training process, the training set trains the detection model for multiple rounds in multiple batches (for example, 100 rounds of training in 100 batches) and repeatedly inputs the detection model, meanwhile, the difference between the detection probability image of the output atrioventricular intersection and the golden standard image is calculated by using a cost function, the difference is used as a training error to be fed back to the detection model, and model parameters are updated through a learning algorithm; and after the training of each batch is finished, performing performance test on the detection model by using a verification set, and after the performance test index training tends to be stable, considering that the training of the detection model is finished, and storing the trained network model. In addition, the network structure of the detection model may include an input module, two downsampling modules, two upsampling modules and an output module, where, except the output module, all the other modules use a batch normalization layer and a nonlinear activation function Relu, and the nonlinear activation function in the output module uses a softmax instead, whose output value is in the (0, 1) interval, and it should be noted that the final softmax is made between the output channels, so that the sum of corresponding position elements in each finally output probability map is 1, which respectively represents the probability that the pixel at the current position in the original first sample coronary image belongs to each background or foreground.
In addition, the detection model can be a deep convolutional neural network CNN, a generated countermeasure network GAN, convolutional neural networks U-Net and V-Net or a recurrent neural network RNN, and the like; super parameters may include network layer number, convolution kernel, learning rate, parameter initialization, training round number, and lot size. The cost function may be a set similarity measure function (Dice) or a focus loss function (Focal loss), and then the training error may also be minimized using one of a random gradient descent (Stochastic gradient descent, SGD), an adaptive moment estimation optimization algorithm (Adaptive Moment Estimation, adam), and a Momentum algorithm (Momentum) to train the detection model.
When the trained detection model is used to detect the coronary image to be detected to obtain the position information of the intersection of the atrioventricular node, as shown in fig. 3b, the step a may optionally include the following steps S302-S308:
s302, inputting the coronary image to be detected into a detection model to obtain a probability map of the intersection points of the rooms corresponding to the coronary image to be detected; the pixel value of each position on the probability map of the atrioventricular intersection point is the probability that the pixel value of the corresponding position on the coronary image to be detected belongs to the atrioventricular intersection point.
S304, binarizing the probability map of the compartment intersection point according to a preset second probability threshold value to obtain a binarization mask image corresponding to the probability map of the compartment intersection point.
S306, marking the connected domain in the binarized mask image, and determining the largest connected domain according to the marked connected domain.
S308, acquiring a weighted center point of a probability value corresponding to the maximum connected domain, and determining the position information of the weighted center point as the position information of the intersection point of the rooms.
In S302-S308, the second probability threshold may be, for example, 0.2, 0.3, 0.35, etc., according to the actual situation.
Specifically, after obtaining the coronary image to be detected, the computer device may input the coronary image to be detected into the detection model to obtain a probability map of an intersection point of the room corresponding to the coronary image to be detected, then compare the probability values of each position on the probability map of the intersection point of the room with the second probability threshold value, set the probability value on the probability map at a position where the probability value is smaller than the second probability threshold value to be 0, set the probability value on the probability map at a position where the probability value is greater than or equal to the second probability threshold value to be 1, obtain a binary probability map, and record the binary probability map as a binary mask image; then, the connected domain with each probability value of 1 on the mask image can be marked (for example, the connected domains with each probability value of 1 can be marked into different colors), the connected domain with the largest area in each connected domain with the probability value of 1 is found out from the connected domain, the connected domain with the largest area is used as the largest connected domain, and then the largest connected domain is corresponding to the probability diagram of the intersection point of the original room and room output by the detection model, so that the probability value and the coordinate of the corresponding position of the largest connected domain are obtained; or, each connected domain with the probability value of 1 is corresponding to the probability map of the original room intersection point according to the corresponding position, the connected domain with the highest probability density on the probability map of the original room intersection point is found, the connected domain with the highest probability density is used as the largest connected domain, then the probability value and the coordinate of the corresponding position of the largest connected domain are obtained, then the probability value is used as the weight, the coordinates corresponding to the weights are weighted and summed, the weighted and summed coordinates are obtained, and the weighted and summed coordinates are marked as the position information of the room intersection point; the probability value on the probability map of the original atrioventricular node obtained by the detection model is not a binary probability value, and is an actually calculated probability value, for example, may be 0.95, 0.88, 0.02, 0.23, 0.45, and the like.
According to the image analysis method provided by the embodiment, a preset detection model can be adopted to detect and process the atrioventricular intersection point in the coronary image to be detected, so that the position information of the atrioventricular intersection point in the coronary image to be detected is obtained, and the position information of the atrioventricular intersection point is used as the reference information of the coronary; the detection model is obtained by training based on the first sample coronary image and the position mark of the intersection point of the room and the room corresponding to the first sample coronary image. In this embodiment, the position information of the intersection point of the atrium and the ventricle on the coronary artery image to be detected can be obtained by using a trained detection model, and the trained detection model is trained based on the position mark of the intersection point of the atrium and the ventricle corresponding to the sample coronary artery image, so that the trained detection model is accurate, and the obtained analysis result is more accurate when the dominant type of the coronary artery is analyzed by using the position information of the intersection point of the atrium and the ventricle subsequently.
In another embodiment, another image detection method is provided, and the embodiment relates to a specific process of analyzing a coronary image to be detected based on the position information of the atrioventricular node and the central line of the coronary artery to obtain a dominant analysis result. On the basis of the above embodiment, as shown in fig. 4, the step S206 may include the following steps:
S402, calculating the shortest distance between the position information of the atrioventricular node and the midline of the left coronary artery to obtain a first distance; and calculating the shortest distance between the position information of the atrioventricular node and the midline of the right coronary artery to obtain a second distance.
Specifically, when calculating the distance, the position information of the intersection point of the atrioventricular node on the coronary image to be detected and the distance between each point on the central line of the left coronary are calculated to obtain a plurality of distances, and the shortest distance is found out from the distances and recorded as a first distance; similarly, the position information of the intersection point of the atrioventricular node on the coronary image to be detected and the distance between each point on the midline of the right coronary artery can be calculated to obtain a plurality of distances, and the shortest distance is found out from the distances and recorded as the second distance. The formula for calculating the distance can be calculated by adopting the existing distance calculation formula between the points, and the first distance and the second distance can be directly calculated by adopting the existing distance formula between the points and the line.
S404, analyzing the coronary image to be detected according to the first distance, the second distance and the preset distance threshold range to obtain an analysis result.
The preset distance threshold range can be calculated according to position information of a marked atrioventricular intersection point on the sample coronary image and central lines of left and right coronary arteries marked on the sample coronary image, and dominant categories can be marked on the sample coronary image. In addition, the preset distance threshold range includes a first distance threshold range, a second distance threshold range, and a third distance threshold range, where the first distance threshold range may be a distance threshold range of an equilibrium type coronary artery calculated according to an equilibrium type sample coronary artery image, the second distance threshold range may be a distance threshold range of a left dominant type coronary artery calculated according to a left dominant type sample coronary artery image, the third distance threshold range may be a distance threshold range of a right dominant type coronary artery calculated according to a right dominant type sample coronary artery image, and the corresponding method for calculating the preset distance threshold range may be as follows:
Calculating the shortest distance from the marked atrioventricular intersection point on the balanced sample coronary image to the central line of the left and right coronary artery to obtain the shortest distance corresponding to each balanced sample coronary image, and then calculating the average value mu of the shortest distances 1 And standard deviation sigma 1 And will [ mu ] 1 -2σ 1 ，μ 1 +2σ 1 ]The range is taken as a first distance threshold range; similarly, calculating the shortest distance from the intersection of the atria marked on the left dominant sample coronary image to the midline of the left coronary, to obtain eachThe shortest distance corresponding to the coronary images of the left dominant sample is calculated, and then the average value mu of the shortest distances is calculated 2 And standard deviation sigma 2 And will [ mu ] 2 -2σ 2 ，μ 2 +2σ 2 ]The range is taken as a second distance threshold range; similarly, calculating the shortest distance from the intersection point of the atria marked on the right dominant sample coronary image to the midline of the right coronary, obtaining the shortest distance corresponding to each right dominant sample coronary image, and then calculating the average mu of the shortest distances 3 And standard deviation sigma 3 And will [ mu ] 3 -2σ 3 ，μ 3 +2σ 3 ]The range serves as a third distance threshold range.
Specifically, after the first distance, the second distance and the preset three distance threshold ranges are obtained, optionally, the first distance, the second distance and the first distance threshold range may be matched; if the first distance and the second distance are not beyond the first distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is balanced; or, matching the first distance with the second distance threshold range; if the first distance does not exceed the second distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is left dominant type; or, matching the second distance with a third distance threshold range; if the second distance does not exceed the third distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is right dominant type.
In other words, in the case of comparing the first distance and the second distance with the predetermined distance threshold value range, the first distance and the second distance may be first respectively compared with the first distance threshold value range [ mu ] 1 -2σ 1 ，μ 1 +2σ 1 ]Comparing, if the first distance and the second distance are not beyond the first distance threshold range, considering that the dominant type of the coronary artery in the coronary artery image to be detected is an equilibrium type; if either one of the first distance and the second distance exceeds the first distance threshold range, continuing to set the first distance and the second distance threshold range [ mu ] 2 -2σ 2 ，μ 2 +2σ 2 ]Comparing, if the first distance does not exceed the second distance threshold range, considering the dominant type of the coronary artery in the coronary artery image to be detected as the left dominant type; if the first distance exceeds the second distance threshold range, then continuing to apply the second distance and the third distance threshold range [ mu ] 3 -2σ 3 ，μ 3 +2σ 3 ]Comparing, if the second distance does not exceed the third distance threshold range, considering the dominant type of the coronary artery in the coronary artery image to be detected as the right dominant type; if the second distance exceeds the third distance threshold range, outputting that the dominant type of the coronary artery in the coronary artery image to be detected is uncertain, and requiring a doctor to further determine.
When the first distance, the second distance and the preset distance threshold range are compared, there may be 6 matching sequences, 1 st: according to the method, firstly matching the balanced type, if the balanced type is not matched, matching the dominant type on the left side, and if the dominant type on the left side is not matched, matching the dominant type on the right side; 2 nd: firstly matching the balanced type, if the balanced type is not matched, matching the dominant type on the right side, and if the dominant type on the right side is not matched, matching the dominant type on the left side; 3 rd: firstly matching the dominant left type, if the dominant left type is not matched, matching the dominant right type, and if the dominant right type is not matched, matching the balanced type; 4 th: firstly matching the dominant left type, matching the balanced type if the dominant left type is not matched, and matching the dominant right type if the balanced type is not matched; 5 th: firstly matching the right dominant type, matching the balanced type if the right dominant type is not matched, and matching the left dominant type if the balanced type is not matched; 6 th: the dominant right type is matched first, if the dominant right type is not matched, the dominant left type is matched, and if the dominant left type is not matched, the balanced type is matched.
Based on the foregoing, if the first distance threshold range may include a left Zhi Di distance threshold range and a right branch first distance threshold range, optionally, the step of matching according to the first distance, the second distance, and the first distance threshold range to obtain the balanced analysis result may include: threshold the first distance and left Zhi Di distance Matching the value range, and matching the second distance with the right branch first distance threshold range; if the first distance does not exceed the left Zhi Di distance threshold range and the second distance does not exceed the right branch first distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is balanced. That is, when the first distance threshold range is determined, the shortest distance from the intersection of the atria and the ventricles marked on the balanced sample coronary image to the central line of the left coronary artery can be calculated to obtain the shortest distance of the left branch corresponding to each balanced sample coronary image, and then the average mu of the shortest distances of the left branches is calculated 1 And standard deviation sigma 1 And left branch [ mu ] 1 -2σ 1 ，μ 1 +2σ 1 ]The range is taken as a left Zhi Di-distance threshold range; meanwhile, the shortest distance from the intersection point of the marked atria and the rooms on the balanced sample coronary images to the central line of the right coronary can be calculated to obtain the right branch shortest distance corresponding to each balanced sample coronary image, and then the average value mu of the shortest distances of the right branches is calculated 1 And standard deviation sigma 1 And right branch [ mu ] 1 -2σ 1 ，μ 1 +2σ 1 ]The range is used as a right branch first distance threshold range; after the first distance and the second distance are obtained by calculating the coronary image to be detected, the first distance and the second distance can be matched with a corresponding left Zhi Di distance threshold range and a corresponding right branch first distance threshold range, and if the first distance and the second distance do not exceed the corresponding distance threshold range, the dominant type of the coronary is determined to be balanced; if any one of the first distance and the second distance exceeds the corresponding distance threshold range, the operation can be performed according to the 6 comparison sequences, so as to obtain the final analysis result of the dominant type of the coronary artery.
According to the image analysis method provided by the embodiment, the shortest distance between the position information of the atrioventricular node and the central line of the left coronary artery can be calculated to obtain the first distance, the shortest distance between the position information of the atrioventricular node and the central line of the right coronary artery can be calculated to obtain the second distance, and the coronary artery image to be detected is analyzed according to the first distance, the second distance and the preset distance threshold range to obtain an analysis result. In this embodiment, the final dominant type can be obtained by comparing the threshold value with the shortest distance between the intersection of the atrioventricular node and the midlines of the left and right coronary arteries, and the method is simple and direct, and does not need manual analysis, so that the manpower can be saved to a certain extent, the analysis time of the dominant type of the coronary arteries can be saved, and the analysis speed of the dominant type of the coronary arteries can be improved.
Next, the case where the reference information is the atrioventricular coordinate system will be described.
In another embodiment, another image detection method is provided, and this embodiment relates to a specific process of detecting a coronary image to be detected by using a detection model to obtain an atrioventricular coordinate system in the coronary image. On the basis of the above embodiment, as shown in fig. 5a, the step S202 may include the following steps:
S502, segmenting the coronary image to be detected to obtain a segmentation result of the heart chamber; the segmentation result of the heart chamber includes left and right atria and left and right ventricles of the heart.
The heart chamber comprises a temple ditch chamber which is respectively a left atrium, a left ventricle, a right atrium and a right ventricle. When the coronary image to be detected is segmented, a cavity segmentation model or an image segmentation algorithm and the like are adopted to segment the cavity in the coronary image to be detected; if the cavity segmentation model is adopted for segmentation, the cavity segmentation model can be trained in advance, and the cavity segmentation model can be obtained by training according to the sample coronary image and gold standard images of four cavities corresponding to the sample coronary image.
Specifically, when the coronary image to be detected is segmented, the coronary image to be detected can be input into a pre-trained cavity segmentation model to obtain segmented images of four cavities, wherein the segmented images of the four cavities can be one image or each cavity corresponds to one segmented image respectively; in addition, the divided images of the four chambers may include positional information of the respective chambers.
S504, establishing an atrioventricular coordinate system according to a segmentation result of the heart chamber, and taking the atrioventricular coordinate system as reference information of coronary artery; the lateral axis direction of the atrioventricular coordinate system is the direction of the boundary line of the atrioventricular groove of the heart, and the longitudinal axis direction of the atrioventricular coordinate system is the direction of the boundary line of the ventricular groove of the heart.
Wherein, the atrioventricular groove boundary line refers to the boundary line between the atrium and the ventricle, and the ventricular groove boundary line refers to the boundary line between the left ventricle and the right ventricle of the heart septum.
Specifically, after obtaining the segmented image of each chamber, the computer device may also obtain the position information of each chamber, where the position information also includes the position information of the boundary point of each chamber, and then may perform fitting processing (such as least square fitting) on the position of each point on the boundary line of the atrioventricular groove to obtain an X axis, or may record the X axis as a transverse axis, where the positive direction of the transverse axis may be the direction of the right atrium of the right ventricle; the positions of the points on the boundary line of the ventricular groove can be fitted to obtain a Y-axis, which can be also marked as a longitudinal axis, the positive direction of the longitudinal axis can be the direction of the left ventricle and the right ventricle, the X-axis and the Y-axis are mutually perpendicular, wherein the origin can be the intersection point of the X-axis and the Y-axis, or the intersection point of the above-mentioned atrioventricular chamber, the X-axis and the Y-axis can obtain an atrioventricular chamber coordinate system through the obtained origin, and the established atrioventricular chamber coordinate system can be shown in fig. 5 b.
The image analysis method provided by the embodiment can segment the coronary image to be detected to obtain the segmentation result of the heart chamber; the segmentation result of the heart chamber comprises left and right atria and left and right ventricles of the heart; and establishing an atrioventricular coordinate system according to the segmentation result of the heart chamber, and taking the atrioventricular coordinate system as the reference information of the coronary artery. In the embodiment, as the atrioventricular coordinate system can be established according to the segmentation result of the heart chamber, a calculation basis can be provided for the subsequent analysis of the dominant type of the coronary artery; in addition, in the embodiment, the coordinate system is fitted by using the segmentation result of the heart chamber instead of directly establishing the coordinate system according to the segmentation result, so that the coordinate system fitted by the embodiment is more accurate, and the subsequent analysis result of the dominant coronary artery can be more accurate.
In another embodiment, another image detection method is provided, and the embodiment relates to a specific process of analyzing a coronary image to be detected based on an atrioventricular coordinate system and midlines of left and right coronary arteries to obtain a dominant analysis result. On the basis of the above embodiment, as shown in fig. 6, the step S206 may include the following steps:
s602, calculating the coordinates of each point on the central line of the left coronary artery under the atrioventricular groove coordinate system to obtain the coordinates of at least one left supporting point, and calculating the coordinates of each point on the central line of the right coronary artery under the atrioventricular groove coordinate system to obtain the coordinates of at least one right supporting point.
Specifically, after the atrioventricular coordinate system is established, the coordinates of each point on the midline of the left coronary artery and the coordinates of each point on the midline of the right coronary artery can be calculated under the standard of the atrioventricular coordinate system, wherein the coordinates of each point on the midline of the left coronary artery are recorded as the coordinates of the left supporting point, and the sitting marks of each point on the midline of the right coronary artery are recorded as the coordinates of the right supporting point. The coordinates herein may be one-dimensional coordinates, two-dimensional coordinates, three-dimensional coordinates, etc., and mainly two-dimensional coordinates of each point are used herein, which are coordinates in the X-axis direction and coordinates in the Y-axis direction, respectively.
S604, analyzing the coronary image to be detected according to the coordinates of at least one left fulcrum, the coordinates of at least one right fulcrum and a preset coordinate threshold range to obtain an analysis result.
In this step, optionally, the following step B may be used to analyze the coronary image to be detected:
step B, matching the coordinates of at least one left fulcrum and the coordinates of at least one right fulcrum with a coordinate threshold range respectively to obtain the number of left branch target points and the number of right branch target points, comparing the number of left branch target points and the number of right branch target points with a preset number threshold, and obtaining an analysis result according to a comparison result; the left branch target point is a point of which the coordinates in at least one left supporting point do not exceed the coordinate threshold range, and the right branch target point is a point of which the coordinates in at least one right supporting point do not exceed the coordinate threshold range.
The preset coordinate threshold range may be an X-axis coordinate threshold range, which may be calculated according to a midline of a posterior descending branch marked on a sample coronary image, and a dominant class may also be marked on the sample coronary image. Specifically, first, marking a left posterior descending branch and a right posterior descending branch on an equilibrium sample coronary artery image in advance, then calculating to obtain the X-axis coordinate of the left posterior descending branch on each equilibrium sample coronary artery image under an atrioventricular coordinate system, taking the maximum coordinate and the minimum coordinate of the X-axis coordinate as a left Zhi Di coordinate threshold range, and simultaneously calculating to obtain the X-axis coordinate of the right posterior descending branch on each equilibrium sample coronary artery image under the atrioventricular coordinate system, and taking the maximum coordinate and the minimum coordinate of the X-axis coordinate as a right branch first coordinate threshold range; secondly, marking a left posterior descending branch on the left dominant sample coronary image, calculating to obtain the X-axis coordinate of the left posterior descending branch on each left dominant sample coronary image under an atrioventricular groove coordinate system, and taking the maximum coordinate and the minimum coordinate of the X-axis coordinate as a left branch second coordinate threshold range; and thirdly, marking a right posterior descending branch on the right dominant sample coronary image, calculating to obtain the X-axis coordinate of the right posterior descending branch on each right dominant sample coronary image under the atrioventricular groove coordinate system, and taking the maximum coordinate and the minimum coordinate of the X-axis coordinate as a right branch second coordinate threshold range.
Specifically, after the coordinates of at least one left fulcrum and the coordinates of at least one right fulcrum are obtained, the coordinates of the left fulcrum and the corresponding left branch coordinate threshold range can be compared to obtain the number of left branch target points in the left branch coordinate threshold range, meanwhile, the coordinates of the right fulcrum and the corresponding right branch coordinate threshold range can be compared to obtain the number of right branch target points in the right branch coordinate threshold range, then the number of left branch target points and the number of right branch target points are compared with the number threshold, and the dominant type analysis result is obtained according to the comparison result.
When the number of left branch target points and the number of right branch target points are obtained and compared with the number threshold value, the number of left branch target points and the number of right branch target points can be divided into three cases for comparison, namely, the number of left branch target points and the number threshold value are compared with the balanced coordinate threshold value range and the number threshold value, the number of left dominant coordinate threshold value range and the number threshold value are compared with the right dominant coordinate threshold value range and the number threshold value.
In the specific comparison, the balanced type condition has two coordinate threshold ranges, so that two quantity thresholds can be correspondingly provided, namely a first quantity threshold of a left branch and a first quantity threshold of a right branch, the comparison process can be to match the coordinate of at least one left fulcrum with a left Zhi Di coordinate threshold range to obtain the quantity of a left Zhi Di target point, and match the coordinate of at least one right fulcrum with a first coordinate threshold range of the right branch to obtain the quantity of a first target point of the right branch; comparing the number of left Zhi Di-target points with a left Zhi Di-number threshold, and comparing the number of right branch first target points with a right branch first number threshold; if the number of the left Zhi Di first target points is greater than the left Zhi Di first number threshold and the number of the right first target points is greater than the right first number threshold, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is balanced.
In addition, the left Zhi Di two-coordinate threshold range in the case of the left dominant type may correspond to a number threshold, and be recorded as a left branch second number threshold, and the specific comparison process may be that the coordinates of at least one left fulcrum and the left Zhi Di two-coordinate threshold range are matched to obtain the number of left Zhi Di two target points, and the number of left Zhi Di two target points and the left Zhi Di number threshold are compared, and if the number of left Zhi Di two target points is greater than the left Zhi Di number threshold, it is determined that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is the left dominant type.
In addition, the range of the right branch second coordinate threshold under the right dominance condition may correspond to a number threshold, and be recorded as a right branch second number threshold, and the specific comparison process may be that the coordinates of at least one right fulcrum and the range of the right branch second coordinate threshold are matched to obtain the number of the right branch second target points, and the number of the right branch second target points and the right branch second number threshold are compared, if the number of the right branch second target points is greater than the right branch second number threshold, it is determined that the analysis result is that the dominance type category of the coronary artery in the coronary artery image to be detected is the right dominance type.
The magnitude of the left Zhi Di first number threshold, the magnitude of the right branch first number threshold, the magnitude of the left Zhi Di second number threshold, and the magnitude of the right branch second number threshold may be determined according to practical situations, and for example, assuming that the number of left fulcrums is 1000 and the number of left branch target points is 800, the left Zhi Di first number threshold may be 700, 750, or the like.
In addition, when the coordinates of the left fulcrum and the coordinates of the right fulcrum are compared with the corresponding coordinate threshold ranges, and when the coordinates of the left fulcrum and/or the coordinates of the right fulcrum do not exceed the corresponding coordinate threshold ranges, the number of points in the coordinate threshold ranges may be compared with the corresponding number threshold.
For example, assume that left Zhi Di-coordinate threshold ranges are denoted as [ CL (μ -2σ), CL (μ+2σ) ], corresponding left Zhi Di-number thresholds are denoted as cl_min, right branch first-coordinate threshold ranges are denoted as [ CR (μ -2σ), CR (μ+2σ) ], corresponding right branch first-number thresholds are denoted as cr_min, left Zhi Di-number threshold ranges are denoted as [ L (μ -2σ), L (μ+2σ) ], corresponding left Zhi Di-number thresholds are denoted as l_min, right branch second-coordinate threshold ranges are denoted as [ R (μ -2σ), R (μ+2σ) ], and corresponding right branch second-number thresholds are denoted as r_min.
Then when judging the dominant type of the coronary image to be detected,
if 1) counting is satisfied at the same time to obtain the number of points with the coordinate X absolute value of the central line point of the left branch of the coronary artery in the value range of [ CL (mu-2sigma), CL (mu+2sigma) ] and the number of the points is larger than a counted threshold CL_min, 2) counting is performed to obtain the number of points with the coordinate X absolute value of the central line point of the right branch of the coronary artery in the value range of [ CR (mu-2sigma), CR (mu+2sigma) ] and the number of the points is larger than a counted threshold CR_min, judging as balanced;
if the statistics shows that the absolute value of the coordinate X of the central line point of the left branch of the coronary artery is within the value range of [ L (mu-2 sigma), L (mu+2 sigma) ] and the number of the points is larger than the statistical threshold L_min, judging that the coronary artery is left dominant;
if the statistics shows that the absolute value of the coordinate X of the central line point of the right branch of the coronary artery is within the value range of [ R (mu-2 sigma), R (mu+2 sigma) ] and the number of the points is larger than the statistical threshold value R_min, the right dominant type is judged.
It should be noted that, in the above-mentioned comparing the coordinates of the left fulcrum and the coordinates of the right fulcrum with the corresponding coordinate threshold ranges, and when the coordinates of the left fulcrum and/or the coordinates of the right fulcrum do not exceed the corresponding coordinate threshold ranges, there may be 6 comparison sequences, where the 6 comparison sequences are the same as the 6 matching sequences corresponding to the above-mentioned intersection points, but the matching parameters during matching are different, which is not described herein again.
According to the image analysis method provided by the embodiment, the coordinates of each point on the central line of the left coronary artery under the atrioventricular groove coordinate system can be calculated to obtain the coordinates of at least one left supporting point, the coordinates of each point on the central line of the right coronary artery under the atrioventricular groove coordinate system can be calculated to obtain the coordinates of at least one right supporting point, and the image of the coronary artery to be detected is analyzed according to the coordinates of at least one left supporting point, the coordinates of at least one right supporting point and a preset coordinate threshold range to obtain an analysis result. In this embodiment, the number of target points of the coordinates of the points on the central lines of the left and right coronary arteries in the coordinate threshold range can be counted, and the number of target points and the threshold value are compared to obtain the dominant type category.
In another embodiment, another image detection method is provided, and this embodiment relates to a specific process of how to divide the coronary artery in the coronary image to be detected. On the basis of the above embodiment, as shown in fig. 7a, the step S204 may include the following steps:
S702, inputting the coronary image to be detected into a preset segmentation model to obtain a probability map of left and right coronary corresponding to the coronary image to be detected; the pixel value of each position on the probability map of the left and right coronary artery is the probability that the pixel value of the corresponding position on the coronary artery image to be detected belongs to the left and right coronary artery, the segmentation model is obtained by training based on a second sample coronary artery image and a gold standard image corresponding to the second sample coronary artery image, and the gold standard image corresponding to the second sample coronary artery image comprises left and right coronary artery marks corresponding to the second sample coronary artery image.
S704, binarizing the probability maps of the left and right coronary arteries according to a preset first probability threshold value to obtain segmented images of the left and right coronary arteries corresponding to the probability maps of the left and right coronary arteries.
In this embodiment, when the segmentation model is used to segment the left and right coronary arteries in the coronary artery image to be detected, the segmentation model needs to be trained in advance, and when the segmentation model is trained, the training process may include the following three steps:
1) A training data set is generated. Drawing (or marking) N coronary left branch coronary vessel regions of interest (the pixel values in the regions are marked as V1, and V1 belongs to natural numbers) and N coronary right branch coronary vessel regions of interest (the pixel values in the regions are marked as V2, and V2 belongs to natural numbers which are not equal to V1) in N second sample coronary vessel images respectively by using marking software, setting the pixel values outside the left and right branch coronary vessel regions of interest as 0, generating N left and right branch coronary vessel drawing images by such processing, and matching the N left and right branch coronary vessel drawing images with the N second sample coronary vessel images to form a training data set, wherein the number of the N is N; in addition, the mode of the second sample coronary image is the same as the mode of the coronary image to be detected, and the number of N can also be determined according to practical situations, for example, can be 1000, 2000, etc. For example, referring to fig. 7b, a schematic diagram of left and right coronary vessels marked on the second sample coronary image is shown, and four figures are left and right coronary vessels marked on a sagittal plane, a coronal plane, a horizontal plane, and a three-dimensional perspective, respectively (it should be noted that these four figures are only an example and do not affect the essential content of the embodiments of the present application).
2) A segmentation model is built, which may also be a convolutional neural network model. Constructing a convolutional neural network model, and setting super parameters of the convolutional neural network model, wherein an input channel of the network is 1, the input channel is a second sample coronary image, an output channel is 3, and the left branch probability maps of N left and right coronary blood vessel segmentation images respectively correspond to the probability maps of the right branch probability maps and the probability maps corresponding to the background; in addition, the training data set in the 1) can be divided into a training set X1 verification set X2 and a testing set X3, wherein the training set, the verification set and the testing set are mutually independent, the number of the training sets, the verification set and the testing set is N1, N2 and N3 respectively, the training data set is a natural number, n1+n2+n3=N, and N1 is more than or equal to 1/2N; for example, assuming that the number of second sample coronary images is 1000, n1=500, n2=200, n3=300 may be possible.
3) And training a segmentation model. Training the segmentation model established in step 2) using the training data set generated in step 1). The training set X1 is used for training the segmentation model, the verification set X2 is used for evaluating the current performance of the model, and the test set X3 is used for checking the generalization performance of the model; in the training process, the training set trains a plurality of batches (for example, 100 batches are used for training 100 rounds) and repeatedly inputs the segmentation model to train for a plurality of rounds, meanwhile, the difference between the output image and the golden standard image is calculated by using a cost function and is used as a training error to be fed back to the segmentation model (the difference between pixel values of the left and right branch images which are respectively calculated and output and the corresponding left and right branch golden standard images is calculated, and the sum value or the average value of the two differences is fed back to the segmentation model), and model parameters are updated through a learning algorithm; and after the training of each batch is finished, performing performance test on the segmentation model by using the verification set, and after the performance test index training tends to be stable, considering that the training of the segmentation model is finished, and storing the trained network model. In addition, the network structure of the segmentation model may be the same as that of the detection model, and will not be described herein.
In addition, the segmentation model can be a deep convolutional neural network CNN, a generative countermeasure network GAN, convolutional neural networks U-Net and V-Net or a recurrent neural network RNN, and the like; super parameters may include network layer number, convolution kernel, learning rate, parameter initialization, training round number, and lot size. The cost function may be a set similarity measure function (Dice) or a focus loss function (Focal loss), and then the training error may also be minimized using one of a random gradient descent (Stochastic gradient descent, SGD), an adaptive moment estimation optimization algorithm (Adaptive Moment Estimation, adam), and a Momentum algorithm (Momentum) to train the segmentation model.
After training the segmentation model, inputting the coronary image to be detected into the trained segmentation model to obtain a probability map of a left coronary vessel and a probability map of a right coronary vessel corresponding to the coronary image to be detected, wherein each probability value on the probability map of the left coronary vessel can represent the probability that the corresponding position belongs to the left coronary vessel, and each probability value on the probability map of the right coronary vessel can represent the probability that the corresponding position belongs to the right coronary vessel; then, each probability value on the probability map of the left coronary vessel can be compared with the first probability threshold value, the probability value which is smaller than the first probability threshold value is set as 0, the probability value which is larger than or equal to the first probability threshold value is set as 1, a binarized image is obtained, and the image is recorded as a segmented image of the left coronary vessel; similarly, each probability value on the probability map of the right coronary vessel may be compared with the first probability threshold, the probability value on which the probability value is smaller than the first probability threshold is set to 0, the probability value on which the probability value is greater than or equal to the first probability threshold is set to 1, and a binarized image is obtained and recorded as a segmented image of the right coronary vessel.
According to the image analysis method provided by the embodiment, the coronary image to be detected can be input into the preset segmentation model, the probability diagrams of the left and right coronary corresponding to the coronary image to be detected are obtained, binarization processing is carried out on the probability diagrams of the left and right coronary according to the preset first probability threshold, and the segmentation images of the left and right coronary corresponding to the probability diagrams of the left and right coronary are obtained. In this embodiment, since the segmentation model is obtained by training based on the sample coronary image and the golden standard images of the left and right coronary markers, the segmentation model obtained by training is relatively accurate, so that when the left and right coronary on the coronary image to be detected are segmented by using the trained segmentation model, the obtained left and right coronary segmentation images are relatively accurate, and then the coronary midline extracted by using the segmentation image later and the obtained dominant analysis result are also relatively accurate.
It should be understood that, although the steps in the flowcharts of fig. 2, 3b, 4, 5a, 6, 7a are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps of fig. 2, 3b, 4, 5a, 6, 7a may comprise a plurality of steps or stages, which are not necessarily performed at the same time, but may be performed at different times, nor does the order of execution of the steps or stages necessarily follow one another, but may be performed alternately or alternately with other steps or at least a portion of the steps or stages in other steps.
In one embodiment, as shown in fig. 8, there is provided an image analysis apparatus including: a detection module 10, a segmentation module 11 and an analysis module 12, wherein:
the detection module 10 is used for carrying out detection processing on the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected; the reference information of the coronary artery is related to the position information of the atrium and the ventricle surrounded by the coronary artery in the coronary artery image to be detected;
the segmentation module 11 is used for carrying out segmentation processing on the coronary image to be detected to obtain segmented images of left and right coronary arteries corresponding to the coronary image to be detected, and extracting central lines of the left and right coronary arteries of the segmented images of the left and right coronary arteries;
the analysis module 12 is used for analyzing the coronary image to be detected according to the reference information of the coronary and the central lines of the left and right coronary to obtain an analysis result; the analysis result is used for representing dominant categories of coronary arteries in the coronary image to be detected.
For specific limitations of the image analysis apparatus, reference may be made to the above limitations of the image analysis method, and no further description is given here.
In another embodiment, another image analysis device is provided, and based on the above embodiment, the detection module 10 is further configured to perform detection processing on an intersection point of a room in a coronary image to be detected by using a preset detection model, obtain location information of the intersection point of the room in the coronary image to be detected, and use the location information of the intersection point of the room as reference information of the coronary; the detection model is obtained by training based on a first sample coronary image and a gold standard image corresponding to the first sample coronary image, wherein the gold standard image corresponding to the first sample coronary image comprises a position mark of an atrioventricular intersection point corresponding to the first sample coronary image, and the atrioventricular intersection point is a point at the junction of a central vein between ventricles and a coronary sinus of a heart room.
Optionally, the detection module 10 may include a first detection unit, a first processing unit, an acquisition unit, and a determination unit, where:
the first detection unit is used for inputting the coronary image to be detected into the detection model to obtain a probability map of the intersection points of the rooms corresponding to the coronary image to be detected; the pixel value of each position on the probability graph of the atrioventricular intersection point is the probability that the pixel value of the corresponding position on the coronary image to be detected belongs to the atrioventricular intersection point;
the first processing unit is used for carrying out binarization processing on the probability map of the intersection points of the rooms according to a preset second probability threshold value to obtain a binarization mask image corresponding to the probability map of the intersection points of the rooms;
an acquisition unit for marking the connected domain in the binarized mask image and determining the largest connected domain according to the marked connected domain;
and the determining unit is used for acquiring a weighted center point of the probability value corresponding to the maximum connected domain and determining the position information of the weighted center point as the position information of the intersection point of the rooms.
In another embodiment, another image analysis apparatus is provided, and the analysis module 12 may include a calculation unit and an analysis unit, where:
The computing unit is used for computing the shortest distance between the position information of the atrioventricular intersection point and the midline of the left coronary artery to obtain a first distance; calculating the shortest distance between the position information of the atrioventricular node and the midline of the right coronary artery to obtain a second distance;
the analysis unit is used for analyzing the coronary image to be detected according to the first distance, the second distance and the preset distance threshold range to obtain an analysis result.
Optionally, the preset distance threshold range includes a first distance threshold range, a second distance threshold range, and a third distance threshold range, and the analysis unit is further configured to match the first distance, the second distance, and the first distance threshold range; if the first distance and the second distance are not beyond the first distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is balanced; or, matching the first distance with the second distance threshold range; if the first distance does not exceed the second distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is left dominant type; or, matching the second distance with a third distance threshold range; if the second distance does not exceed the third distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is right dominant type.
Optionally, the first distance threshold range includes a left Zhi Di first distance threshold range and a right first distance threshold range, and the analysis unit is further configured to match the first distance to the left Zhi Di first distance threshold range, and match the second distance to the right first distance threshold range; if the first distance does not exceed the left Zhi Di distance threshold range and the second distance does not exceed the right branch first distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is balanced.
In another embodiment, another image analysis apparatus is provided, and the detection module 10 may include a second detection unit and a setup unit, where:
the second detection unit is used for carrying out segmentation processing on the coronary image to be detected to obtain a segmentation result of the heart chamber; the segmentation result of the heart chamber comprises left and right atria and left and right ventricles of the heart;
the establishing unit is used for establishing an atrioventricular coordinate system according to the segmentation result of the heart chamber, and taking the atrioventricular coordinate system as the reference information of coronary artery; the lateral axis direction of the atrioventricular coordinate system is the direction of the boundary line of the atrioventricular groove of the heart, and the longitudinal axis direction of the atrioventricular coordinate system is the direction of the boundary line of the ventricular groove of the heart.
In another embodiment, another image analysis device is provided, where based on the foregoing embodiment, the computing unit is further configured to calculate coordinates of each point on a midline of the left coronary artery under an atrioventricular groove coordinate system to obtain coordinates of at least one left fulcrum, and calculate coordinates of each point on a midline of the right coronary artery under the atrioventricular groove coordinate system to obtain coordinates of at least one right fulcrum;
the analysis unit is further configured to analyze the coronary image to be detected according to the coordinates of the at least one left fulcrum, the coordinates of the at least one right fulcrum, and the preset coordinate threshold range, so as to obtain an analysis result.
Optionally, the analyzing unit is further configured to match the coordinates of the at least one left fulcrum and the coordinates of the at least one right fulcrum with a coordinate threshold range, obtain the number of left branch target points and the number of right branch target points, compare the number of left branch target points and the number of right branch target points with a preset number threshold, and obtain an analysis result according to the comparison result; the left branch target point is a point of which the coordinates in at least one left supporting point do not exceed the coordinate threshold range, and the right branch target point is a point of which the coordinates in at least one right supporting point do not exceed the coordinate threshold range.
Optionally, if the coordinate threshold range includes a left Zhi Di coordinate threshold range and a right branch first coordinate threshold range, the number threshold includes a left Zhi Di number threshold and a right branch first number threshold; the analysis unit is further configured to match the coordinates of at least one left fulcrum with a left Zhi Di-coordinate threshold range to obtain the number of left Zhi Di-target points, and match the coordinates of at least one right fulcrum with a right-branch first-coordinate threshold range to obtain the number of right-branch first-target points; comparing the number of left Zhi Di-target points with a left Zhi Di-number threshold, and comparing the number of right branch first target points with a right branch first number threshold; if the number of the left Zhi Di first target points is greater than the left Zhi Di first number threshold and the number of the right first target points is greater than the right first number threshold, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is balanced.
Optionally, if the coordinate threshold range includes a left Zhi Di two coordinate threshold range and a right branch second coordinate threshold range, the number threshold includes a left branch second number threshold and a right branch second number threshold; the analysis unit is further configured to match the coordinates of the at least one left fulcrum with a left Zhi Di two-coordinate threshold range to obtain the number of left Zhi Di two-target points, compare the number of left Zhi Di two-target points with a left Zhi Di two-number threshold, and determine that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is a left dominant type if the number of left Zhi Di two-target points is greater than the left Zhi Di two-number threshold; or matching the coordinate of at least one right fulcrum with the right branch second coordinate threshold range to obtain the number of right branch second target points, comparing the number of right branch second target points with the right branch second number threshold, and if the number of right branch second target points is larger than the right branch second number threshold, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is right dominant type.
In another embodiment, another image analysis apparatus is provided, and on the basis of the above embodiment, the above powder each module 11 may include a dividing unit and a second processing unit, wherein:
the segmentation unit is used for inputting the coronary image to be detected into a preset segmentation model to obtain a probability map of left and right coronary corresponding to the coronary image to be detected; the pixel value of each position on the probability map of the left and right coronary artery is the probability that the pixel value of the corresponding position on the coronary artery image to be detected belongs to the left and right coronary artery, the segmentation model is obtained by training based on a second sample coronary artery image and a gold standard image corresponding to the second sample coronary artery image, and the gold standard image corresponding to the second sample coronary artery image comprises left and right coronary artery marks corresponding to the second sample coronary artery image;
and the second processing unit is used for carrying out binarization processing on the probability maps of the left and right coronary arteries according to a preset first probability threshold value to obtain segmented images of the left and right coronary arteries corresponding to the probability maps of the left and right coronary arteries.
For specific limitations of the image analysis apparatus, reference may be made to the above limitations of the image analysis method, and no further description is given here. The respective modules in the image analysis apparatus described above may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In an embodiment, there is also provided a computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the method embodiments described above when the computer program is executed.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, carries out the steps of the method embodiments described above.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, or the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples merely represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.

Claims (14)

1. A method of image analysis, the method comprising:
detecting the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected; the reference information is related to the position information of an atrium and a ventricle surrounded by coronary artery in the coronary artery image to be detected;
dividing the coronary image to be detected to obtain divided images of left and right coronary arteries corresponding to the coronary image to be detected, and extracting central lines of the left and right coronary arteries of the divided images of the left and right coronary arteries;
Analyzing the coronary image to be detected according to the reference information of the coronary and the central lines of the left and right coronary to obtain an analysis result; the analysis result is used for representing dominant categories of coronary arteries in the coronary artery image to be detected.
2. The method according to claim 1, wherein the detecting the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected includes:
detecting and processing an atrioventricular intersection point in a coronary image to be detected by adopting a preset detection model to obtain the position information of the atrioventricular intersection point in the coronary image to be detected, and taking the position information of the atrioventricular intersection point as the reference information of the coronary;
the detection model is obtained by training based on a first sample coronary image and a gold standard image corresponding to the first sample coronary image, wherein the gold standard image corresponding to the first sample coronary image comprises a position mark of an atrioventricular intersection point corresponding to the first sample coronary image, and the atrioventricular intersection point is a point at the junction of a central vein between heart septum ventricles and a coronary sinus of a heart room.
3. The method according to claim 2, wherein the midlines of the left and right coronary comprise a midline of the left coronary and a midline of the right coronary, and the analyzing the coronary image to be detected according to the reference information of the coronary and the midlines of the left and right coronary to obtain an analysis result comprises:
Calculating the shortest distance between the position information of the atrioventricular intersection point and the midline of the left coronary artery to obtain a first distance; calculating the shortest distance between the position information of the atrioventricular intersection point and the midline of the right coronary artery to obtain a second distance;
and analyzing the coronary image to be detected according to the first distance, the second distance and a preset distance threshold range to obtain the analysis result.
4. A method according to claim 3, wherein the preset distance threshold range includes a first distance threshold range, a second distance threshold range, and a third distance threshold range, and the analyzing the coronary image to be detected according to the first distance, the second distance, and the preset distance threshold range, to obtain the analysis result includes:
matching the first distance, the second distance and the first distance threshold range; if the first distance and the second distance are not beyond the first distance threshold range, determining that the analysis result is that the dominant type of the coronary in the coronary image to be detected is balanced;
or, matching the first distance with the second distance threshold range; if the first distance does not exceed the second distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is left dominant type;
Or, matching the second distance with the third distance threshold range; and if the second distance does not exceed the third distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is right dominant type.
5. The method of claim 4, wherein the first distance threshold range comprises a left Zhi Di distance threshold range and a right branch first distance threshold range, the matching the first distance and the second distance to the first distance threshold range; if the first distance and the second distance do not exceed the first distance threshold range, determining that the analysis result is that the dominant type of the coronary in the coronary image to be detected is balanced, including:
matching the first distance to the left Zhi Di distance threshold range, and matching the second distance to the right branch first distance threshold range;
if the first distance does not exceed the left Zhi Di distance threshold range and the second distance does not exceed the right branch first distance threshold range, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is an equilibrium type.
6. The method according to claim 1, wherein the detecting the coronary image to be detected to obtain the reference information of the coronary in the coronary image to be detected includes:
dividing the coronary image to be detected to obtain a heart chamber division result; the segmentation result of the heart chamber comprises left and right atria and left and right ventricles of the heart;
establishing an atrioventricular coordinate system according to a segmentation result of the heart chamber, and taking the atrioventricular coordinate system as reference information of the coronary artery; the lateral axis direction of the atrioventricular coordinate system is the direction of the boundary line of the atrioventricular groove of the heart, and the longitudinal axis direction of the atrioventricular coordinate system is the direction of the boundary line of the ventricular groove of the heart.
7. The method according to claim 6, wherein the midlines of the left and right coronary comprise a midline of the left coronary and a midline of the right coronary, and the analyzing the coronary image to be detected according to the reference information of the coronary and the midlines of the left and right coronary to obtain an analysis result comprises:
calculating the coordinates of each point on the midline of the left coronary artery under the atrioventricular groove coordinate system to obtain the coordinates of at least one left fulcrum, and calculating the coordinates of each point on the midline of the right coronary artery under the atrioventricular groove coordinate system to obtain the coordinates of at least one right fulcrum;
And analyzing the coronary image to be detected according to the coordinates of the at least one left fulcrum, the coordinates of the at least one right fulcrum and a preset coordinate threshold range to obtain an analysis result.
8. The method according to claim 7, wherein the analyzing the coronary image to be detected according to the coordinates of the at least one left fulcrum and the coordinates of the at least one right fulcrum and the preset coordinate threshold range, to obtain an analysis result, includes:
matching the coordinates of the at least one left fulcrum and the coordinates of the at least one right fulcrum with the coordinate threshold range respectively to obtain the number of left branch target points and the number of right branch target points, comparing the number of left branch target points, the number of right branch target points and a preset number threshold, and obtaining the analysis result according to the comparison result;
the left branch target point is a point of which the coordinates in the at least one left supporting point do not exceed the coordinate threshold range, and the right branch target point is a point of which the coordinates in the at least one right supporting point do not exceed the coordinate threshold range.
9. The method of claim 8, wherein if the coordinate threshold range comprises a left Zhi Di coordinate threshold range and a right first coordinate threshold range, the number threshold comprises a left Zhi Di number threshold and a right first number threshold,
The matching the coordinates of the at least one left fulcrum and the coordinates of the at least one right fulcrum with the coordinate threshold range respectively to obtain the number of left branch target points and the number of right branch target points, comparing the number of left branch target points, the number of right branch target points and a preset number threshold, and obtaining the analysis result according to the comparison result, including:
matching the coordinates of the at least one left fulcrum with the left Zhi Di-coordinate threshold range to obtain the number of left Zhi Di-target points, and matching the coordinates of the at least one right fulcrum with the right branch first-coordinate threshold range to obtain the number of right branch first-target points;
comparing the number of the left Zhi Di first target points with the left Zhi Di first number threshold, and comparing the number of the right branch first target points with the right branch first number threshold;
if the number of the left Zhi Di-target points is greater than the left Zhi Di-number threshold and the number of the right first-target points is greater than the right first-number threshold, determining that the analysis result is that the dominant type of the coronary artery in the coronary artery image to be detected is balanced.
10. The method of claim 8, wherein if the coordinate threshold range includes a left Zhi Di two coordinate threshold range and a right count second coordinate threshold range, the number threshold includes a left count second number threshold and a right count second number threshold,
the matching the coordinates of the at least one left fulcrum and the coordinates of the at least one right fulcrum with the coordinate threshold range respectively to obtain the number of left branch target points and the number of right branch target points, comparing the number of left branch target points, the number of right branch target points and a preset number threshold, and obtaining the analysis result according to the comparison result, including:
matching the coordinates of the at least one left fulcrum with the left Zhi Di two-coordinate threshold range to obtain the number of left Zhi Di two-target points, comparing the number of the left Zhi Di two-target points with the left Zhi Di two-number threshold, and if the number of the left Zhi Di two-target points is greater than the left Zhi Di two-number threshold, determining that the analysis result is that the dominant type of coronary artery in the coronary artery image to be detected is left dominant type; or alternatively, the process may be performed,
and matching the coordinates of the at least one right fulcrum with the right branch second coordinate threshold range to obtain the number of right branch second target points, comparing the number of right branch second target points with the right branch second number threshold, and determining that the analysis result is that the dominant type class of the coronary artery in the coronary artery image to be detected is right dominant type if the number of right branch second target points is larger than the right branch second number threshold.
11. The method according to any one of claims 1 to 10, wherein the segmenting the coronary image to be detected to obtain segmented images of the left and right coronary corresponding to the coronary image to be detected includes:
inputting the coronary image to be detected into a preset segmentation model to obtain a probability map of left and right coronary corresponding to the coronary image to be detected; the pixel value of each position on the probability map of the left and right coronary artery is the probability that the pixel value of the corresponding position on the coronary artery image to be detected belongs to the left and right coronary artery, the segmentation model is obtained by training based on a second sample coronary artery image and a gold standard image corresponding to the second sample coronary artery image, and the gold standard image corresponding to the second sample coronary artery image comprises left and right coronary artery marks corresponding to the second sample coronary artery image;
and carrying out binarization processing on the probability maps of the left and right coronary arteries according to a preset first probability threshold value to obtain segmented images of the left and right coronary arteries corresponding to the probability maps of the left and right coronary arteries.
12. The method according to any one of claims 2-5, wherein the detecting the atrioventricular intersection in the coronary image to be detected using the preset detection model to obtain the position information of the atrioventricular intersection in the coronary image to be detected includes:
Inputting the coronary image to be detected into the detection model to obtain a probability map of an atrioventricular node corresponding to the coronary image to be detected; the pixel value of each position on the probability map of the atrioventricular intersection point is the probability that the pixel value of the corresponding position on the coronary image to be detected belongs to the atrioventricular intersection point;
performing binarization processing on the probability map of the compartment intersection point according to a preset second probability threshold value to obtain a binarization mask image corresponding to the probability map of the compartment intersection point;
marking the connected domain in the binarization mask image, and determining the largest connected domain in the binarization mask image according to the marked connected domain;
and acquiring a weighted center point of a probability value corresponding to the maximum connected domain, and determining the position information of the weighted center point as the position information of the atrioventricular intersection.
13. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 12 when the computer program is executed.
14. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 12.
CN202010137683.0A 2020-03-02 2020-03-02 Image analysis method, apparatus and storage medium Active CN111369525B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010137683.0A CN111369525B (en) 2020-03-02 2020-03-02 Image analysis method, apparatus and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010137683.0A CN111369525B (en) 2020-03-02 2020-03-02 Image analysis method, apparatus and storage medium

Publications (2)

Publication Number Publication Date
CN111369525A CN111369525A (en) 2020-07-03
CN111369525B true CN111369525B (en) 2023-05-30

Family

ID=71210384

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010137683.0A Active CN111369525B (en) 2020-03-02 2020-03-02 Image analysis method, apparatus and storage medium

Country Status (1)

Country Link
CN (1) CN111369525B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112164027B (en) * 2020-09-01 2023-10-27 联影智能医疗科技(北京)有限公司 Spinal dural sac compression detection method, equipment and storage medium
CN112336361B (en) * 2020-09-29 2023-10-31 查云飞 Coronary artery detection method, device and storage medium
CN112801999B (en) * 2021-02-05 2023-06-23 深圳睿心智能医疗科技有限公司 Method and device for determining heart coronary artery dominance
CN112907563B (en) * 2021-03-17 2023-12-01 复旦大学附属华山医院 CTA full-automatic collateral circulation scoring method and system based on deep learning
CN113139961B (en) * 2021-05-28 2024-04-23 东软医疗***股份有限公司 Method and device for determining dominant type of coronary artery
CN113409323B (en) * 2021-06-18 2022-03-08 推想医疗科技股份有限公司 Coronary artery segmentation method and device, computer-readable storage medium and electronic equipment
CN113450322B (en) * 2021-06-22 2024-05-24 东软医疗***股份有限公司 Method and device for judging provenance abnormality of coronary artery
CN114078139B (en) * 2021-11-25 2024-04-16 四川长虹电器股份有限公司 Image post-processing method based on human image segmentation model generation result
CN114418947B (en) * 2021-12-16 2023-11-03 深圳市阅影科技有限公司 Naming method and device for heart coronary blood vessel
CN116030326B (en) * 2023-01-18 2023-09-08 山东省人工智能研究院 Multi-view fused coronary plaque identification method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102012215035A1 (en) * 2012-08-23 2014-02-27 Siemens Aktiengesellschaft Method for simulation of virtual or actual stress of coronary stent for open-holding of e.g. trachea of human stent-angioplasty patient after dilation during cancer treatment, involves representing segment superimposed with visualization
CN104978725A (en) * 2014-04-03 2015-10-14 上海联影医疗科技有限公司 Method and device for dividing coronary artery
CN105741310A (en) * 2016-03-21 2016-07-06 东北大学 Heart's left ventricle image segmentation system and method
CN110443808A (en) * 2019-07-04 2019-11-12 杭州深睿博联科技有限公司 Medical image processing method and device, equipment, storage medium for the detection of brain middle line

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010117573A2 (en) * 2009-04-07 2010-10-14 Virginia Commonwealth University Automated measurement of brain injury indices using brain ct images, injury data, and machine learning
WO2015139977A1 (en) * 2014-03-19 2015-09-24 Koninklijke Philips N.V. Method of scan geometry planning for determining wall thickness of an anatomic detail using magnetic resonance imaging

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102012215035A1 (en) * 2012-08-23 2014-02-27 Siemens Aktiengesellschaft Method for simulation of virtual or actual stress of coronary stent for open-holding of e.g. trachea of human stent-angioplasty patient after dilation during cancer treatment, involves representing segment superimposed with visualization
CN104978725A (en) * 2014-04-03 2015-10-14 上海联影医疗科技有限公司 Method and device for dividing coronary artery
CN105741310A (en) * 2016-03-21 2016-07-06 东北大学 Heart's left ventricle image segmentation system and method
CN110443808A (en) * 2019-07-04 2019-11-12 杭州深睿博联科技有限公司 Medical image processing method and device, equipment, storage medium for the detection of brain middle line

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
64排螺旋CT对冠状动脉解剖形态显示的影像评价;钱家新;;现代医用影像学(01);全文 *
Cardiac substructure segmentation with deep learning for improved cardiac sparing;Eric D.Morris等;PMC;全文 *
Coronary Artery Segmentation in Cardiac CT Angiography Using 3D Multi-Channel U-net;Yo-Chuan Chen等;Image and Video Processing;全文 *
Graph Convolutional Networks for Coronary Artery Segmentation in Cardiac CT Angiography;Jelmer m.Wolterink等;Electrical Engineering and Systems Science;全文 *
形状约束下活动轮廓模型冠脉血管图像多尺度分割;郭笑妍等;中国图象图形学报(07);全文 *
心脏核磁共振图像左心室底层组织分割方法;徐礼胜等;东北大学学报(自然科学版)(10);全文 *

Also Published As

Publication number Publication date
CN111369525A (en) 2020-07-03

Similar Documents

Publication Publication Date Title
CN111369525B (en) Image analysis method, apparatus and storage medium
CN111292314B (en) Coronary artery segmentation method, device, image processing system and storage medium
Zabihollahy et al. Convolutional neural network‐based approach for segmentation of left ventricle myocardial scar from 3D late gadolinium enhancement MR images
CN111383259B (en) Image analysis method, computer device, and storage medium
US20150278727A1 (en) Systems and methods for using geometry sensitivity information for guiding workflow
KR102219890B1 (en) Apparatus for spatial normalization of medical image using deep learning and method thereof
CN109934812A (en) Image processing method, device, server and storage medium
CN111709485B (en) Medical image processing method, device and computer equipment
CN108846829B (en) Lesion site recognition device, computer device, and readable storage medium
CN109712163B (en) Coronary artery extraction method, device, image processing workstation and readable storage medium
CN113538471B (en) Plaque segmentation method, plaque segmentation device, computer equipment and storage medium
CN111932552B (en) Aorta modeling method and device
CN111145160B (en) Method, device, server and medium for determining coronary artery branches where calcified regions are located
CN111488872B (en) Image detection method, image detection device, computer equipment and storage medium
CN111951276A (en) Image segmentation method and device, computer equipment and storage medium
CN111652300A (en) Spine curvature classification method, computer device and storage medium
CN111568451A (en) Exposure dose adjusting method and system
CN110738702B (en) Three-dimensional ultrasonic image processing method, device, equipment and storage medium
CN108597589B (en) Model generation method, target detection method and medical imaging system
US10354349B2 (en) Systems and methods for using geometry sensitivity information for guiding workflow
CN116091466A (en) Image analysis method, computer device, and storage medium
CN109087357B (en) Scanning positioning method and device, computer equipment and computer readable storage medium
CN113160199B (en) Image recognition method and device, computer equipment and storage medium
CN113129418B (en) Target surface reconstruction method, device, equipment and medium based on three-dimensional image
CN114298999A (en) Method for detecting vascular structure variation, readable storage medium, and program product

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant