CN113643223A

CN113643223A - Image detection method, image detection device, computer equipment and storage medium

Info

Publication number: CN113643223A
Application number: CN202010326782.3A
Authority: CN
Inventors: 陈磊
Original assignee: Shanghai United Imaging Intelligent Healthcare Co Ltd
Current assignee: Shanghai United Imaging Intelligent Healthcare Co Ltd
Priority date: 2020-04-23
Filing date: 2020-04-23
Publication date: 2021-11-12

Abstract

The application relates to an image detection method, an image detection device, a computer device and a storage medium. The method comprises the following steps: acquiring a medical image of an object to be detected and related information of the object to be detected; carrying out segmentation processing on the medical image of the object to be detected to obtain a segmentation result; the segmentation result comprises a segmentation result of a vertebra and a segmentation result of soft tissue around the vertebra; carrying out quantitative analysis on the segmentation result of the vertebra and the segmentation result of the soft tissue around the vertebra to obtain a quantitative result of the vertebra and a quantitative result of the soft tissue around the vertebra; and generating a detection report of the object to be detected based on the related information of the object to be detected, the quantitative result of the vertebra and the quantitative result of the soft tissues around the vertebra. The method can save labor and time.

Description

Image detection method, image detection device, computer equipment and storage medium

Technical Field

The present application relates to the field of image processing technologies, and in particular, to an image detection method, an image detection apparatus, a computer device, and a storage medium.

Background

The spine consists of vertebrae, the vertebral bodies of which comprise the largest part of the spine, and intervertebral discs, which comprise a small part of the spine. Generally, the spine includes vertebrae and sacrum composed of cervical vertebrae, thoracic vertebrae, lumbar vertebrae, etc., and coccyx, and it is seen that the spine is very important for the human body. Nowadays, as the sitting time of people increases, more and more people have spine pathological changes, which seriously affect the normal life of people, so the spine detection is very necessary.

In the related art, when a doctor examines a subject, generally, a medical image taken by the subject is transmitted to a reading system of a hospital, then the medical image of the subject can be observed on the reading system, the spine of the subject can be examined according to experience, and after the examination is completed, the doctor can feed back an examination result to the subject in a form of a report.

However, the above technique has a problem that it takes a lot of labor and time to examine the subject.

Disclosure of Invention

In view of the above, it is necessary to provide an image detection method, an apparatus, a computer device, and a storage medium capable of saving labor and time in view of the above technical problems.

An image detection method, the method comprising:

acquiring a medical image of an object to be detected and related information of the object to be detected;

carrying out segmentation processing on a medical image of an object to be detected to obtain a segmentation result; the segmentation result comprises a segmentation result of the vertebra and a segmentation result of soft tissues around the vertebra;

carrying out quantitative analysis on the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra to obtain a quantization result of the vertebra and a quantization result of the soft tissues around the vertebra;

and generating a detection report of the object to be detected based on the related information of the object to be detected, the quantitative result of the vertebra and the quantitative result of the soft tissue around the vertebra.

In one embodiment, the method further includes:

determining identity identification information of the object to be detected based on the related information of the object to be detected;

associating the identity identification information of the object to be detected with the detection result of the object to be detected and storing the association in a database; the detection result of the object to be detected includes a segmentation result of the spine, a segmentation result of soft tissue around the spine, a quantization result of the spine, and a quantization result of soft tissue around the spine.

In one embodiment, the method further includes:

acquiring and displaying a segmentation result of the spine of the object to be detected, a segmentation result of soft tissue around the spine, a quantization result of the spine and a quantization result of the soft tissue around the spine from a database based on the identification information of the object to be detected;

obtaining a target quantization result determined from the quantization result of the spine and the quantization result of soft tissues around the spine;

and determining a segmentation result corresponding to the target quantization result from the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra according to the target quantization result, and displaying the segmentation result corresponding to the target quantization result.

In one embodiment, the method further includes:

carrying out quantization analysis again on the segmentation result corresponding to the target quantization result to obtain a quantization result again;

judging whether the re-quantization result is the same as the target quantization result;

and if not, updating the detection report of the object to be detected based on the re-quantization result.

In one embodiment, if the medical image of the object to be detected is a plurality of medical images of different modalities, the segmenting the medical image of the object to be detected to obtain a segmentation result includes:

carrying out image registration on a plurality of medical images of different modalities to obtain a plurality of registered medical images;

performing fusion processing on the registered medical images to obtain a fused medical image;

and carrying out segmentation processing on the fused medical image to obtain a segmentation result.

In one embodiment, the image registration of the medical images of the plurality of different modalities to obtain a plurality of registered medical images includes:

registering the spines in the medical images of different modalities by adopting a rigid registration method to obtain a plurality of first registered medical images;

and registering the vertebra in the plurality of first registered medical images and soft tissues around the vertebra by adopting a non-rigid registration method to obtain a plurality of second registered medical images, and determining the plurality of second registered medical images as a plurality of registered medical images.

In one embodiment, before performing quantitative analysis on the segmentation result of the spine and the segmentation result of the soft tissue around the spine to obtain the quantization result of the spine and the quantization result of the soft tissue around the spine, the method further includes:

determining a normalized characteristic value of the object to be detected based on the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra;

according to the normalization characteristic values, respectively carrying out normalization processing on the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra to obtain a normalization segmentation result of the vertebra and a normalization segmentation result of the soft tissues around the vertebra;

accordingly, the above-mentioned performing quantitative analysis on the segmentation result of the spine and the segmentation result of the soft tissue around the spine to obtain the quantization result of the spine and the quantization result of the soft tissue around the spine includes:

and carrying out quantitative analysis on the normalization segmentation result of the vertebra and the normalization segmentation result of soft tissues around the vertebra to obtain a quantization result of the vertebra and a quantization result of the soft tissues around the vertebra.

An image sensing apparatus, the apparatus comprising:

the acquisition module is used for acquiring a medical image of an object to be detected and related information of the object to be detected;

the segmentation module is used for carrying out segmentation processing on the medical image of the object to be detected to obtain a segmentation result; the segmentation result comprises a segmentation result of the vertebra and a segmentation result of soft tissues around the vertebra;

the quantification module is used for carrying out quantification analysis on the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra to obtain a quantification result of the vertebra and a quantification result of the soft tissues around the vertebra;

and the generation module is used for generating a detection report of the object to be detected based on the relevant information of the object to be detected, the quantitative result of the vertebra and the quantitative result of soft tissues around the vertebra.

A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

According to the image detection method, the image detection device, the computer equipment and the storage medium, the segmentation and quantization processing are carried out on the medical image of the object to be detected, so that the segmentation result of the spine of the object to be detected, the segmentation result of soft tissue around the spine, the quantization result of the spine and the quantization result of the soft tissue around the spine are obtained, and the detection report of the object to be detected is generated based on the related information of the object to be detected and the quantization result of the object to be detected. In the method, because the segmentation and quantitative analysis of the spine and soft tissues around the spine of the object to be detected can be completed through an automatic process, and the detection report of the object to be detected is automatically generated according to the quantitative result and the related information of the object to be detected, the detection process does not need manual operation at all, so the method can save labor and time; in addition, because the method can directly segment and quantitatively analyze the vertebra on the medical image of the detected person to obtain the final detection result, and the vertebra of the detected person does not need to be detected manually according to experience, the detection result obtained by the method is more accurate, and the conditions of missing detection and false detection caused by detection through manual experience can be reduced.

Drawings

FIG. 1 is a diagram illustrating an internal structure of a computer device according to an embodiment;

FIG. 2 is a flow diagram illustrating an exemplary image detection method;

FIG. 3 is a flow chart illustrating an image detection method according to another embodiment;

FIG. 4 is a flow chart illustrating an image detection method according to another embodiment;

FIG. 4a is a diagram illustrating an interface for displaying a segmentation result and a quantization result according to another embodiment;

FIG. 5 is a flow chart illustrating an image detection method according to another embodiment;

FIG. 6 is a flow chart illustrating an image detection method according to another embodiment;

FIG. 7 is a block diagram showing the structure of an image detection apparatus according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The human spine is composed of vertebrae and intervertebral discs, 3/4 being the length of the spine is composed of vertebral bodies, 1/4 is composed of intervertebral discs. The spine is composed of 26 vertebrae, namely 24 vertebrae (7 cervical vertebrae, 12 thoracic vertebrae, 5 lumbar vertebrae), 1 sacrum and 1 coccyx. The anterior longitudinal ligament is arranged in the front of the vertebra, the posterior longitudinal ligament is arranged in the back of the vertebra, a plurality of vertebra bones are connected, a vertebral canal is arranged between the front and back of the vertebral column, and bone marrow is hidden. Normal human spine has certain mobility, but the mobility of each part is different, the mobility of cervical and lumbar segments is larger, the mobility of thoracic segment is extremely small, and the sacral segment has almost no mobility. According to the statistical report in recent years, many adults in China have cervical spondylosis, lumbar spondylosis and other spinal diseases. Wherein the number of patients with cervical spondylosis increases at a rate of several million people per year. The medical health service level of the whole process of diagnosis, treatment, rehabilitation and the like of the spondylopathy is improved by new generation technologies such as artificial intelligence and the like which are urgently needed by huge patient groups. Cervical spondylosis and lumbar spondylosis are common diseases in spondylopathy. The imaging examination is the main means for clinical diagnosis of spondylopathy, and the commonly used imaging examination methods include CT, MRI and X-ray plain film. An image post-processing workstation provided by an image equipment manufacturer and a general film reading workstation provided by a PACS (Picture Archiving and Communication Systems) can support an image doctor to manually read films, and perform differential diagnosis on various spinal diseases based on personal experience, wherein diagnosis results are different from person to person and difficult to objectively quantify, so that diagnosis standards are inconsistent, missed diagnosis and misdiagnosis are easy to occur, and time and labor are consumed. Based on this, the present application provides an image detection method, an image detection apparatus, a computer device, and a storage medium, which can solve the above technical problems.

The image detection method provided by the application can be applied to computer equipment, and the computer equipment can be a terminal or a server. Taking the computer device as a terminal as an example, the internal structure diagram thereof can be as 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 comprises a nonvolatile 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 an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication 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, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that the architecture shown in fig. 1 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

The execution subject of the embodiment of the present application may be a computer device, or may be an image detection apparatus, and the method of the embodiment of the present application will be described below with reference to the computer device as the execution subject.

In one embodiment, an image detection method is provided, and the embodiment relates to a specific process of automatically segmenting and quantifying a medical image of a to-be-detected object and finally generating a detection report. As shown in fig. 2, the method may include the steps of:

s202, acquiring a medical image of the object to be detected and related information of the object to be detected.

In this step, when acquiring the medical image of the object to be detected, the medical image of the object to be detected may be obtained by scanning a specific part or a whole body of the object to be detected, or the medical image of the object to be detected may be obtained in a database in which the medical image of the object to be detected is stored in advance, which of course may be other acquiring manners, and this embodiment is not limited specifically. In addition, the medical image of the object to be detected here may be a CT (Computed Tomography) image, a PET (Positron Emission Tomography) image, an MR (Magnetic Resonance) image, a PET-MR image, a PET-CT image, or the like; the medical image of the object to be examined may be a one-dimensional image, a two-dimensional image, a three-dimensional image, or the like.

It should be noted that the medical image of the object to be detected acquired in the present embodiment generally includes a vertebra, soft tissue around the vertebra, and the vertebra may be a cervical vertebra, a lumbar vertebra, a thoracic vertebra, and the like.

In addition, the related information of the object to be detected may include identity information of the object to be detected, such as age, sex, height, weight, identification number, mobile phone number, driver's license number, and the like of the object to be detected, and may be obtained by querying the object to be detected and inputting the queried object into the computer device, so as to obtain the related information of the object to be detected.

S204, segmenting the medical image of the object to be detected to obtain a segmentation result; the segmentation result includes a segmentation result of the spine and a segmentation result of soft tissue around the spine.

When the medical image of the object to be detected is segmented, a pre-trained segmentation model can be used for segmentation, the segmentation model can include a graph cut algorithm model, a watershed algorithm model, a GrabCT algorithm model, a machine learning model and the like, and the machine learning model can be an FCN (full Convolutional network), an FPN (flat field network), a SegNet (SegNet), a DeepLab (DeepLab), a Mask-RCNN (Mask-RCNN) model, a U-Net (U-Net), a V-Net (V-Net) model and the like.

In addition, when the segmentation model is trained, the training medical image comprises the spine and soft tissues around the spine, and the spine and the soft tissues around the spine are marked in the training medical image.

That is, after the medical image of the object to be detected is segmented by using the segmentation model, a segmentation result is obtained, the segmentation result in this step may include a segmentation image of a spine region and a segmentation image of a soft tissue region around the spine, may also include a segmentation image of a lesion region on the spine and a segmentation image of a lesion region on a soft tissue around the spine, and may also include other segmentation images related to the spine, which is not specifically limited herein. It is noted that here the soft tissue surrounding the spine may include fat, muscle, ligaments, etc. surrounding the spine.

After the spine and the soft tissues around the spine of the object to be detected are segmented, simple analysis can be carried out on the segmentation result, and the obtained preliminary analysis result is used as the segmentation result of the step. Illustratively, the preliminary analysis results may include whether the object to be detected is cervical spondylosis, lumbar spondylosis, vertebral body hyperosteogeny, lumbar physiological curvature straightening or lateral bending, vertebral facet joint hyperplasia sclerosis, intervertebral disc lesion, ligament calcification, intervertebral foramen stenosis, dural sac compression, vertebral body endplate inflammation, lesion of spinal cord, spondylolisthesis, other lesions, and the like.

And S206, carrying out quantitative analysis on the segmentation result of the vertebra and the segmentation result of the soft tissue around the vertebra to obtain the quantization result of the vertebra and the quantization result of the soft tissue around the vertebra.

The quantitative analysis here may be quantitative calculation of quantitative parameters such as the shape, volume, length, width, and the like of the spine based on the segmentation result of the spine region, the segmentation result of the soft tissue region around the spine, the segmentation result of the lesion region on the soft tissue around the spine, analysis processes such as comparison with corresponding parameter thresholds, and the like, of the quantitative parameters, and then, quantitative results such as the lesion type, the lesion degree, and the like of the spine and the soft tissue around the spine are obtained from the analysis results.

Exemplarily, some quantitative parameters of the lesion region of the object to be detected can be obtained by performing quantitative analysis on the segmentation result, and then the quantitative result is obtained through the quantitative parameters; illustratively, the quantified results of the spine and the soft tissue around the spine of the object to be detected may include spinal canal stenosis, cervical curvature abnormality, intervertebral space stenosis, uncinate process hyperplasia, cartilage degeneration, hyperosteogeny, protrusion to the periphery, calcification of the posterior longitudinal ligament, yellow ligament thickening, cartilage degenerative lesion, anterior spondylolisthesis, lateral spondylolisthesis, spinal cord compression, and the like.

Note that, before performing the quantitative analysis on the segmented results of the spines and the soft tissues around the spines of different detection objects, the segmented results of the spines and the soft tissues around the spines of the detection objects may be directly subjected to the quantitative analysis, or the segmented results of the spines and the soft tissues around the spines may be subjected to self-normalization processing (that is, each detection object corresponds to one normalization value) in advance, and then the normalization processing is performed, followed by the quantitative analysis processing.

And S208, generating a detection report of the object to be detected based on the related information of the object to be detected, the quantitative result of the vertebra and the quantitative result of soft tissues around the vertebra.

In the present step, when a detection report is generated, the above-described division result and the quantization result may be added to the detection report, or only the quantization result may be included.

Specifically, the computer device may preset a blank detection report, on which contents of the detection report, such as related information of the object to be detected, a quantization result, a segmentation image, and the like, may be listed; and then after the related information of the object to be detected, the segmentation result and the quantification result of the vertebra, the segmentation result and the quantification result of soft tissue around the vertebra are obtained, the quantification result and the segmentation result can be correspondingly output to the corresponding position of a blank detection report, so that the detection report of the object to be detected is obtained, the report does not need to be manually filled in the generation process of the detection report, and the detection report can be completely realized by computer equipment, so that the vertebra and the soft tissue around the vertebra can be automatically detected.

In the image detection method, the segmentation result of the spine of the object to be detected, the segmentation result of soft tissue around the spine, the quantization result of the spine and the quantization result of the soft tissue around the spine are obtained by performing segmentation processing and quantization processing on the medical image of the object to be detected, and a detection report of the object to be detected is generated based on the related information of the object to be detected and the quantization result of the object to be detected. In the method, because the segmentation and quantitative analysis of the spine and soft tissues around the spine of the object to be detected can be completed through an automatic process, and the detection report of the object to be detected is automatically generated according to the quantitative result and the related information of the object to be detected, the detection process does not need manual operation at all, so the method can save labor and time; in addition, because the method can directly segment and quantitatively analyze the vertebra on the medical image of the detected person to obtain the final detection result, and the vertebra of the detected person does not need to be detected manually according to experience, the detection result obtained by the method is more accurate, and the conditions of missing detection and false detection caused by detection through manual experience can be reduced.

In another embodiment, another image detection method is provided, and the embodiment relates to a specific process of how to save the detection result of the object to be detected. As shown in fig. 3, the method may further include the steps of:

s302, based on the relevant information of the object to be detected, the identity identification information of the object to be detected is determined.

In this step, after the relevant information of the object to be detected is obtained in S202, because the relevant information of the object to be detected is more or longer, and the like, a larger storage space is consumed for storage, and storage is not convenient, therefore, a part of the relevant information of the object to be detected can be selected as the identification information of the object to be detected by a certain rule, so that the object to be detected can be identified, and the storage space can also be reduced; for example, the rule may be that an initial letter of all relevant information of the object to be detected is taken, and then the identification information of the object to be detected is formed, or a part of the identification number of the object to be detected (or a part of the driver license number, a part of the telephone number, etc.) may also be used as the identification information of the object to be detected, which of course may also be according to other rules, which is not specifically limited in this embodiment.

S304, associating the identity identification information of the object to be detected with the detection result of the object to be detected and storing the association in a database; the detection result of the object to be detected includes a segmentation result of the spine, a segmentation result of soft tissue around the spine, a quantization result of the spine, and a quantization result of soft tissue around the spine.

Specifically, after obtaining the identification information of the object to be detected, and obtaining the segmentation result of the spine of the object to be detected, the segmentation result of the soft tissue around the spine, the quantization result of the spine, and the quantization result of the soft tissue around the spine, the identification information of the object to be detected, the segmentation result and the quantization result corresponding to the identification information of the object to be detected may be bound together, that is, a correspondence relationship is established, and the established correspondence relationship is stored in the database. Here, when the correspondence relationship is established, the correspondence relationship between each of the division results and the quantization result may be established, and the correspondence relationship between each of the division results and the quantization result may be stored in the database.

After the identity identification information of the object to be detected, the corresponding segmentation result and the quantization result are saved, the identity identification information of the object to be detected can be output to a user or a doctor, so that the user or the doctor can find the segmentation result and the quantization result corresponding to the index in a database by taking the identity identification of the object to be detected as the index in the subsequent process, and recheck can be conveniently carried out.

The image detection method provided by this embodiment can obtain the identification information of the object to be detected based on the related information of the object to be detected, and store the identification information, the segmentation result and the quantization result of the spine and the soft tissue around the spine in a database in an associated manner. In this embodiment, since the identity information, the segmentation result and the quantization result of the object to be detected can be associated and stored, a data basis can be provided for the follow-up review of the segmentation result, and a doctor or a user can conveniently search the segmentation result and the quantization result of the object to be detected in the later stage.

In another embodiment, another image detection method is provided, and the embodiment relates to a specific process of how to obtain a segmentation result and a quantization result of an object to be detected from a database and recheck the segmentation result of the object to be detected. As shown in fig. 4, the method may further include the steps of:

s402, acquiring the segmentation result of the vertebra of the object to be detected, the segmentation result of soft tissues around the vertebra, the quantization result of the vertebra and the quantization result of the soft tissues around the vertebra from the database based on the identification information of the object to be detected, and displaying the segmentation result, the segmentation result of the soft tissues around the vertebra, the quantization result of the vertebra and the quantization result of the soft tissues around the vertebra.

S404, obtaining a target quantization result determined from the quantization result of the vertebra and the quantization result of soft tissues around the vertebra.

And S406, determining a segmentation result corresponding to the target quantization result from the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra according to the target quantization result, and displaying the segmentation result corresponding to the target quantization result.

In this embodiment, after the identification information of the object to be detected is stored in the database, the identification information of the object to be detected can be output, so that a subsequent doctor can find the segmentation result of the spine, the segmentation result of the soft tissue around the spine, the quantization result of the spine and the quantization result of the soft tissue around the spine corresponding to the identification information of the object to be detected in the database, and after finding the segmentation result and the quantization result, the segmentation result and the quantization result can also be displayed on a display interface of a reading system or a terminal for the doctor to view. Generally, the quantified result of the spine and the quantified result of the soft tissue around the spine may include a plurality of quantified analysis results, such as spinal canal stenosis, cervical curvature abnormality, intervertebral stenosis, uncinate process hyperplasia, cartilage degeneration, hyperosteogeny, protrusion to the periphery, calcification of the posterior longitudinal ligament, thickening of the ligamentum flavum, cartilage degenerative lesion, anterior spondylolisthesis, lateral spondylolisthesis, and spinal cord compression, which may all be listed on a display interface, and then a doctor may click and select one quantified result on the displayed plurality of quantified results if needing to check which quantified result, and the selected quantified result is recorded as a target quantified result.

After the target quantization result is obtained by selection, the target segmentation result corresponding to the target quantization result may be found in the database according to the stored correspondence between the segmentation result and the quantization result, and then the target segmentation result is displayed on the display interface, where the display position may be determined according to the distribution of the actual display interface, for example, as shown in fig. 4a, a detection report (also referred to as an imaging report) of the object to be detected may be displayed on the left side of the display interface, the target segmentation result (the target segmentation result may be a segmentation image and may be displayed in an image display area) may be displayed in the middle, and the quantization result (also referred to as an image symptom) may be displayed on the right side of the display interface, which of course may be other manners, and this embodiment is not specifically limited thereto. It should be noted that fig. 4a is only an example, and the contents of images, reports, and the like do not affect the essential contents of the embodiments of the present application. In addition, when the target segmentation result is displayed, a lesion area or the like may be marked on the target segmentation result.

Optionally, after the target segmentation result is displayed, performing quantization analysis again on the segmentation result corresponding to the target quantization result to obtain a quantization result again; judging whether the re-quantization result is the same as the target quantization result; and if not, updating the detection report of the object to be detected based on the re-quantization result. That is, after observing the target segmentation result, if a significant error is detected, for example, the curvature of the cervical vertebrae displayed on the target segmentation result is significantly straight, but the curvature corresponding to the target quantization result is inverse-bowed, the physician may perform quantization analysis again on the target segmentation result, where the quantization analysis again may be to recalculate the quantization parameter on the target segmentation result, obtain a quantization result again through the quantization parameter calculated again, compare whether the two quantization results corresponding to the target segmentation result are the same, if the two quantization results are the same, take the detection report generated in S208 as a final detection report, and if the two quantization results are different, modify the detection report generated in S208, and correspondingly update the second quantization result on the detection report to obtain the final detection report. For example, continuing to take the curvature detection as an example, the curvature of the cervical vertebrae in the target segmented image may be recalculated, and then the curvature class is obtained according to the calculated curvature, for example, the curvature class obtained by the second quantization result is straight, so that the curvature class needs to be changed from the initial anti-bow to straight on the detection report of S208, and a changed detection report is obtained.

The image detection method provided by this embodiment can obtain and display a segmentation result and a quantization result related to the spine of the object to be detected based on the identification information of the object to be detected, select and display a target quantization result from the displayed multiple quantization results, obtain and display a segmentation result corresponding to the target quantization result, and recheck the segmentation result corresponding to the target quantization result to obtain a final inspection report. In the embodiment, the segmentation result corresponding to the target quantization result can be obtained and displayed according to the stored database, so that a doctor can conveniently and intuitively observe the quantization result and the segmentation result of the spine; in addition, the target quantization result can be selected from the quantization results, and the segmentation result corresponding to the target quantization result is displayed, so that a doctor can avoid blindly checking all segmentation results of the spine, and the efficiency of reading the picture by the doctor is improved; in addition, the embodiment can recheck the displayed target segmentation result, so that missed diagnosis and misdiagnosis conditions in the initial detection can be avoided, and the accuracy of spine detection can be improved.

In another embodiment, another image detection method is provided, and this embodiment relates to a specific process of performing image registration on a plurality of medical images of different modalities and segmenting the registered images if the medical image of the object to be detected is a plurality of medical images of different modalities. As shown in fig. 5, the step S204 may include the following steps:

s502, image registration is carried out on the medical images in different modalities to obtain a plurality of registered medical images.

Image registration refers to a process of matching and superimposing two or more images acquired at different times, different sensors (or imaging devices), or under different conditions (weather, illuminance, camera position and angle, etc.). Firstly, extracting features of two images to obtain feature points, finding matched feature point pairs by similarity measurement, then obtaining image space coordinate transformation parameters by the matched feature point pairs, and finally carrying out image registration by the coordinate transformation parameters.

In this step, the same registration method may be used to perform image registration on a plurality of medical images of different modalities, or different registration methods may be used to perform image registration on a plurality of medical images of different modalities, optionally, this embodiment mainly uses two different registration methods to perform image registration on a plurality of medical images of different modalities, and the registration process includes the following steps a1 and a 2:

step A1, the spine in the medical images of different modalities is registered by using a rigid registration method, and a plurality of first registered medical images are obtained.

Step A2, registering the vertebra and soft tissue around the vertebra in the first registered medical images by using a non-rigid registration method to obtain a plurality of second registered medical images, and determining the second registered medical images as the registered medical images.

Wherein the medical images of different modalities may include any two or three of CT, MR, PET, and so forth; the rigid registration method is mainly used for registering organs with relatively fixed relative positions between the organs, for example, in the embodiment, the rigid registration method is adopted to register a plurality of vertebras of the vertebra on images with different modes one by one, the process is repeated, the image registration of all the vertebras of the vertebra is completed, a plurality of medical images after the registration are obtained, and taking the multi-modal images before the registration as CT and MR as examples, paired registration images (for example, CT-MR pairs) can be obtained through rigid registration; the non-rigid registration method, which may also be referred to as a flexible registration method, may register a plurality of vertebrae with motion on the medical image, for example, in this embodiment, a multi-modal medical image of the plurality of vertebrae and the soft tissue around the vertebrae is registered by using the flexible registration method.

It should be noted that, taking a multi-modality image as CT and MR as an example, in this embodiment, the advantage of imaging bone tissue by CT and the advantage of imaging soft tissue around bone by MR are utilized, image registration is completed on the vertebra bone and the soft tissue around the vertebra on the two modality images by using a registration algorithm, and the registered images have the advantages of imaging bone tissue by CT and soft tissue by MR, so that when a subsequent doctor uses the registered medical image to perform lesion detection, a lesion can be better and comprehensively observed to obtain a more accurate detection result.

And S504, performing fusion processing on the plurality of registered medical images to obtain a fused medical image.

In this step, after the registered medical images are fused, a fused medical image is generally obtained, and the fused medical image combines the imaging advantages of the medical images of the respective modalities, so that a better spine detection result can be obtained by using the fused medical image in the following step.

S506, segmentation processing is carried out on the fused medical image to obtain a segmentation result.

The complementary information expressed by the images of different modalities can be integrated and integrated through image registration and image fusion, so that when a doctor uses the registered and fused images for segmentation, the obtained segmentation result is relatively accurate, and when the focus on the spine is detected based on the segmentation result, the obtained detection result of the focus on the spine is more accurate, namely, the accuracy of the focus detection on the spine can be improved to a certain extent through image registration.

The image detection method provided by the embodiment can be used for registering medical images of different modalities through an image registration method, and performing fusion processing and segmentation processing on the registered medical images to obtain a segmentation result. In this embodiment, complementary information expressed by images of different modalities can be integrated and integrated through image registration and image fusion, so that the obtained spine lesion detection result is more accurate when the registered and fused images are subsequently used for spine lesion detection, that is, the method of this embodiment can improve the accuracy of spine lesion detection to a certain extent.

In another embodiment, another image detection method is provided, and the embodiment relates to a specific process of normalizing the segmentation result of the spine and soft tissue around the spine and performing quantitative analysis based on the normalization processing result. As shown in fig. 6, before S206, the method may further include:

s602, determining a normalized characteristic value of the object to be detected based on the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra.

In this step, the normalized feature value of each object to be detected is obtained by calculating the segmentation result of the object to be detected, that is, each object to be detected has a normalized feature value of its own, and the normalized feature values of different objects to be detected may be the same or different. The normalization characteristic value of each object to be detected normalizes the segmentation result, so that the accuracy of subsequent analysis processing such as classification and quantification can be ensured on the basis.

In addition, the normalized feature value may include two parts of normalized values, namely, an image normalized value and a normalized value of the size of the spine, and the above-mentioned segmentation result of the spine and the segmentation result of the soft tissue around the spine may include a segmentation image of the spine and a segmentation image of the soft tissue around the spine, where the spine and the soft tissue around the spine are combined to form the spine. The image normalization value may be obtained by averaging the CT values of soft tissue (e.g., adipose tissue) around the spine on the segmented image and taking any one or all of the average value and the standard deviation as the image normalization value; the normalized value of the spine size may be the size of the spine as a whole in the segmentation result, where the size of the spine as a whole may be the volume of the spine, the length of the spine, and so on.

S604, according to the normalized characteristic values, normalization processing is respectively carried out on the segmentation result of the vertebra and the segmentation result of the soft tissue around the vertebra, and the normalization segmentation result of the vertebra and the normalization segmentation result of the soft tissue around the vertebra are obtained.

In this step, after obtaining the image normalization value of the object to be detected, each image value on the segmented image of the vertebra and the segmented image of the soft tissue around the vertebra may be divided by the image normalization value to obtain a normalized segmented image of the vertebra and a normalized segmented image of the soft tissue around the vertebra, and at the same time, the size of the vertebra of the object to be detected and the size of the soft tissue around the vertebra on the segmented image may be statistically obtained, and then the size of the vertebra of the object to be detected and the size of the soft tissue around the vertebra on the segmented image are divided by the normalization value of the size of the vertebra, respectively, so as to finally obtain a normalized segmentation result of the vertebra and a normalized segmentation result of the soft tissue around the vertebra. Of course, the above-mentioned procedure of normalizing the size of the spine may be performed in reverse order with the procedure of normalizing the image values, or may be performed simultaneously, which is not particularly limited in this embodiment. By the normalization processing of the image value and the size of the spine, the quantitative analysis error caused by different ages, heights, sexes, image brightness, image gray levels and the like of the detection objects can be avoided, and the detection accuracy is improved.

Accordingly, the above S206 may include the steps of:

and S606, carrying out quantitative analysis on the normalization segmentation result of the vertebra and the normalization segmentation result of soft tissues around the vertebra to obtain a quantization result of the vertebra and a quantization result of the soft tissues around the vertebra.

The image detection method provided by this embodiment may perform normalization processing on the segmentation result of the spine and the segmentation result of soft tissue around the spine by using the normalization feature value of the object to be detected, and perform quantitative analysis processing on the segmentation result after the normalization processing to obtain a quantitative analysis result. In this embodiment, since the normalization feature value corresponding to the object to be detected can be used to normalize the segmentation result related to the spine, the quantitative analysis error caused by the difference in age, height, sex, image brightness, image gray scale, etc. of the object to be detected can be avoided, and the accuracy of subsequent detection of the spine focus region can be improved.

In another embodiment, in order to facilitate a more detailed description of the technical solution of the present application, the following description is given in conjunction with a more detailed embodiment, and the method may include the following steps S1-S12:

and S1, acquiring medical images of a plurality of different modalities of the object to be detected and related information of the object to be detected.

S2, performing image registration on the medical images of different modalities by adopting a rigid registration method and a non-rigid registration method to obtain a plurality of registered medical images, and performing fusion processing on the plurality of aligned medical images to obtain a fused medical image.

S3, performing segmentation processing on the fused medical image to obtain a segmentation result; the segmentation result includes a segmentation result of the spine and a segmentation result of soft tissue around the spine.

And S4, determining the normalized characteristic value of the object to be detected based on the segmentation result of the vertebra and the segmentation result of the soft tissue around the vertebra.

And S5, respectively carrying out normalization processing on the segmentation result of the vertebra and the segmentation result of the soft tissue around the vertebra according to the normalization characteristic values to obtain the normalization segmentation result of the vertebra and the normalization segmentation result of the soft tissue around the vertebra.

And S6, carrying out quantitative analysis on the normalization segmentation result of the vertebra and the normalization segmentation result of the soft tissue around the vertebra to obtain a quantization result of the vertebra and a quantization result of the soft tissue around the vertebra.

And S7, generating a detection report of the object to be detected based on the related information of the object to be detected, the quantitative result of the vertebra and the quantitative result of the soft tissues around the vertebra.

And S8, associating and storing the identification information of the object to be detected with the segmentation result of the spine of the object to be detected, the segmentation result of soft tissue around the spine, the quantization result of the spine and the quantization result of the soft tissue around the spine into a database.

And S9, determining the identity information of the object to be detected based on the related information of the object to be detected.

And S10, acquiring and displaying the segmentation result of the spine, the segmentation result of soft tissues around the spine, the quantization result of the spine and the quantization result of the soft tissues around the spine of the object to be detected from the database based on the identification information of the object to be detected.

And S11, acquiring a target quantization result determined from the quantization result of the vertebra and the quantization result of the soft tissue around the vertebra, determining a segmentation result corresponding to the target quantization result from the segmentation result of the vertebra and the segmentation result of the soft tissue around the vertebra according to the target quantization result, and displaying the segmentation result corresponding to the target quantization result.

And S12, carrying out quantization analysis again on the segmentation result corresponding to the target quantization result to obtain a quantization result again, judging whether the quantization result again is the same as the target quantization result, and if not, updating the detection report of the object to be detected based on the quantization result again.

It should be understood that although the various steps in the flow charts of fig. 2-6 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-6 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.

In one embodiment, as shown in fig. 7, there is provided an image detection apparatus including: an obtaining module 10, a segmentation module 11, a quantization module 12 and a generation module 13, wherein:

the acquiring module 10 is used for acquiring a medical image of an object to be detected and related information of the object to be detected;

the segmentation module 11 is configured to perform segmentation processing on a medical image of an object to be detected to obtain a segmentation result; the segmentation result comprises a segmentation result of the vertebra and a segmentation result of soft tissues around the vertebra;

a quantization module 12, configured to perform quantization analysis on the segmentation result of the spine and the segmentation result of the soft tissue around the spine to obtain a quantization result of the spine and a quantization result of the soft tissue around the spine;

and the generating module 13 is configured to generate a detection report of the object to be detected based on the related information of the object to be detected, the quantization result of the spine and the quantization result of the soft tissue around the spine.

For specific limitations of the image detection apparatus, reference may be made to the above limitations of the image detection method, which are not described herein again.

In another embodiment, another image detection apparatus is provided, and on the basis of the above embodiment, the apparatus may further include an identification determination module and an association saving module, where:

the identification determining module is used for determining the identity identification information of the object to be detected based on the relevant information of the object to be detected;

the association storage module is used for associating the identity identification information of the object to be detected with the detection result of the object to be detected and storing the association storage information and the detection result into a database; the detection result of the object to be detected includes a segmentation result of the spine, a segmentation result of soft tissue around the spine, a quantization result of the spine, and a quantization result of soft tissue around the spine.

In another embodiment, another image detection apparatus is provided, which may further include an acquisition display module, an object acquisition module, and a result display module, on the basis of the above embodiment, wherein:

the acquisition and display module is used for acquiring and displaying the segmentation result of the spine, the segmentation result of soft tissue around the spine, the quantization result of the spine and the quantization result of the soft tissue around the spine of the object to be detected from the database based on the identification information of the object to be detected;

a target acquisition module for acquiring a target quantization result determined from a quantization result of a vertebra and a quantization result of soft tissue around the vertebra;

and the result display module is used for determining a segmentation result corresponding to the target quantization result from the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra according to the target quantization result and displaying the segmentation result corresponding to the target quantization result.

Optionally, the apparatus may further include a rechecking module, where the rechecking module is configured to perform a second quantization analysis on the segmentation result corresponding to the target quantization result to obtain a second quantization result; judging whether the re-quantization result is the same as the target quantization result; and if not, updating the detection report of the object to be detected based on the re-quantization result.

In another embodiment, another image detection apparatus is provided, and on the basis of the above embodiment, the segmentation module 11 may further include a registration unit, a fusion unit, and a segmentation unit, where:

the registration unit is used for carrying out image registration on a plurality of medical images in different modalities to obtain a plurality of registered medical images;

the fusion unit is used for carrying out fusion processing on the registered medical images to obtain a fused medical image;

and the segmentation unit is used for carrying out segmentation processing on the fused medical image to obtain a segmentation result.

Optionally, the registration unit is further configured to register the vertebrae in the medical images of the multiple different modalities by using a rigid registration method, so as to obtain multiple first registered medical images; and registering the vertebra in the plurality of first registered medical images and soft tissues around the vertebra by adopting a non-rigid registration method to obtain a plurality of second registered medical images, and determining the plurality of second registered medical images as a plurality of registered medical images.

In another embodiment, another image detection apparatus is provided, and on the basis of the above embodiment, before the quantization module 12 performs quantization analysis on the segmentation result of the spine and the segmentation result of the soft tissue around the spine to obtain the quantization result of the spine and the quantization result of the soft tissue around the spine, the apparatus may further include a normalization module, where the normalization module is configured to determine a normalized feature value of the object to be detected based on the segmentation result of the spine and the segmentation result of the soft tissue around the spine; according to the normalization characteristic values, respectively carrying out normalization processing on the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra to obtain a normalization segmentation result of the vertebra and a normalization segmentation result of the soft tissues around the vertebra;

accordingly, the quantization module 12 is further configured to perform quantization analysis on the normalized segmentation result of the vertebra and the normalized segmentation result of the soft tissue around the vertebra to obtain a quantization result of the vertebra and a quantization result of the soft tissue around the vertebra.

The modules in the image detection device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

In one embodiment, the processor, when executing the computer program, further performs the steps of:

determining identity identification information of the object to be detected based on the related information of the object to be detected; associating the identity identification information of the object to be detected with the detection result of the object to be detected and storing the association in a database; the detection result of the object to be detected includes a segmentation result of the spine, a segmentation result of soft tissue around the spine, a quantization result of the spine, and a quantization result of soft tissue around the spine.

acquiring and displaying a segmentation result of the spine of the object to be detected, a segmentation result of soft tissue around the spine, a quantization result of the spine and a quantization result of the soft tissue around the spine from a database based on the identification information of the object to be detected; obtaining a target quantization result determined from the quantization result of the spine and the quantization result of soft tissues around the spine; and determining a segmentation result corresponding to the target quantization result from the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra according to the target quantization result, and displaying the segmentation result corresponding to the target quantization result.

carrying out quantization analysis again on the segmentation result corresponding to the target quantization result to obtain a quantization result again; judging whether the re-quantization result is the same as the target quantization result; and if not, updating the detection report of the object to be detected based on the re-quantization result.

carrying out image registration on a plurality of medical images of different modalities to obtain a plurality of registered medical images; performing fusion processing on the registered medical images to obtain a fused medical image; and carrying out segmentation processing on the fused medical image to obtain a segmentation result.

registering the spines in the medical images of different modalities by adopting a rigid registration method to obtain a plurality of first registered medical images; and registering the vertebra in the plurality of first registered medical images and soft tissues around the vertebra by adopting a non-rigid registration method to obtain a plurality of second registered medical images, and determining the plurality of second registered medical images as a plurality of registered medical images.

determining a normalized characteristic value of the object to be detected based on the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra; according to the normalization characteristic values, respectively carrying out normalization processing on the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra to obtain a normalization segmentation result of the vertebra and a normalization segmentation result of the soft tissues around the vertebra; and carrying out quantitative analysis on the normalization segmentation result of the vertebra and the normalization segmentation result of soft tissues around the vertebra to obtain a quantization result of the vertebra and a quantization result of the soft tissues around the vertebra.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. An image detection method, characterized in that the method comprises:

carrying out segmentation processing on the medical image of the object to be detected to obtain a segmentation result; the segmentation result comprises a segmentation result of a vertebra and a segmentation result of soft tissue around the vertebra;

carrying out quantitative analysis on the segmentation result of the vertebra and the segmentation result of the soft tissue around the vertebra to obtain a quantitative result of the vertebra and a quantitative result of the soft tissue around the vertebra;

and generating a detection report of the object to be detected based on the related information of the object to be detected, the quantitative result of the vertebra and the quantitative result of the soft tissues around the vertebra.

2. The method of claim 1, further comprising:

determining the identity identification information of the object to be detected based on the related information of the object to be detected;

associating the identity identification information of the object to be detected with the detection result of the object to be detected and storing the association in a database; the detection result of the object to be detected comprises a segmentation result of the vertebra, a segmentation result of soft tissue around the vertebra, a quantification result of the vertebra and a quantification result of the soft tissue around the vertebra.

3. The method of claim 2, further comprising:

acquiring and displaying a segmentation result of the spine of the object to be detected, a segmentation result of soft tissues around the spine, a quantization result of the spine and a quantization result of the soft tissues around the spine from the database based on the identification information of the object to be detected;

obtaining a target quantization result determined from the quantization result of the spine and the quantization result of soft tissue around the spine;

and according to the target quantization result, determining a segmentation result corresponding to the target quantization result from the segmentation result of the vertebra and the segmentation result of soft tissues around the vertebra, and displaying the segmentation result corresponding to the target quantization result.

4. The method of claim 3, further comprising:

5. The method according to any one of claims 1 to 4, wherein if the medical image of the object to be detected is a plurality of medical images of different modalities, the segmenting the medical image of the object to be detected to obtain a segmentation result includes:

carrying out image registration on the medical images of the different modalities to obtain a plurality of registered medical images;

and carrying out segmentation processing on the fused medical image to obtain the segmentation result.

6. The method according to claim 5, wherein the image registering the plurality of medical images of different modalities to obtain a plurality of registered medical images comprises:

registering the spines in the medical images of the different modalities by adopting a rigid registration method to obtain a plurality of first registered medical images;

and registering the vertebra in the plurality of first registered medical images and soft tissues around the vertebra by adopting a non-rigid registration method to obtain a plurality of second registered medical images, and determining the plurality of second registered medical images as the plurality of registered medical images.

7. The method according to any one of claims 1 to 4, wherein before the performing the quantitative analysis on the segmentation result of the vertebra and the segmentation result of the soft tissue around the vertebra to obtain the quantification result of the vertebra and the quantification result of the soft tissue around the vertebra, the method further comprises:

determining a normalized characteristic value of the object to be detected based on the segmentation result of the spine and the segmentation result of soft tissues around the spine;

according to the normalization characteristic values, respectively carrying out normalization processing on the segmentation result of the vertebra and the segmentation result of the soft tissues around the vertebra to obtain a normalization segmentation result of the vertebra and a normalization segmentation result of the soft tissues around the vertebra;

correspondingly, the performing quantitative analysis on the segmentation result of the vertebra and the segmentation result of the soft tissue around the vertebra to obtain a quantitative result of the vertebra and a quantitative result of the soft tissue around the vertebra includes:

and carrying out quantitative analysis on the normalization segmentation result of the vertebra and the normalization segmentation result of the soft tissues around the vertebra to obtain the quantization result of the vertebra and the quantization result of the soft tissues around the vertebra.

8. An image detection apparatus, characterized in that the apparatus comprises:

the segmentation module is used for carrying out segmentation processing on the medical image of the object to be detected to obtain a segmentation result; the segmentation result comprises a segmentation result of a vertebra and a segmentation result of soft tissue around the vertebra;

the quantification module is used for carrying out quantification analysis on the segmentation result of the vertebra and the segmentation result of the soft tissues around the vertebra to obtain the quantification result of the vertebra and the quantification result of the soft tissues around the vertebra;

and the generation module is used for generating a detection report of the object to be detected based on the relevant information of the object to be detected, the quantitative result of the vertebra and the quantitative result of the soft tissues around the vertebra.

9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.