CN116092642A - Medical image quality control method and system - Google Patents
Medical image quality control method and system Download PDFInfo
- Publication number
- CN116092642A CN116092642A CN202310209400.2A CN202310209400A CN116092642A CN 116092642 A CN116092642 A CN 116092642A CN 202310209400 A CN202310209400 A CN 202310209400A CN 116092642 A CN116092642 A CN 116092642A
- Authority
- CN
- China
- Prior art keywords
- evaluation
- medical image
- focus
- image
- quality
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- 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
- G06T7/0014—Biomedical image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
-
- 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
- G06T2207/30168—Image quality inspection
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Strategic Management (AREA)
- Quality & Reliability (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Development Economics (AREA)
- Primary Health Care (AREA)
- Game Theory and Decision Science (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Epidemiology (AREA)
- Public Health (AREA)
- Operations Research (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a medical image quality control method and a system, wherein the method comprises the following steps: according to the medical image and the checking position corresponding to the medical image, each identification unit in the medical image is identified and obtained; judging whether the relative azimuth relation of the positions of the identification units on the medical image is matched with a preset relative azimuth; dividing the identification unit on the medical image into a focus to-be-identified unit and a non-focus identification unit according to the past examination diagnosis report of the tested person; evaluating the focus unit to be identified according to a first evaluation standard to obtain a first evaluation result; evaluating the non-focus identification unit according to a second evaluation standard to obtain a second evaluation result; and outputting a quality disqualification prompt or a quality qualification prompt of the medical image according to the first evaluation result and the second evaluation result. The invention can ensure the real-time performance of quality control and simultaneously ensure the high-quality image requirement of the focus area.
Description
Technical Field
The invention relates to the field of medical images, in particular to a medical image quality control method and a medical image quality control system.
Background
Medical imaging mainly includes X-rays, computerized tomography (CT Computed Tomography) and magnetic resonance, and performs physical examination by capturing images of the interior of the human body, playing an important role in the field of disease detection. The medical image can help doctors to quickly lock the etiology and make medical treatment schemes by displaying focus through images, so that the medical image is required to have high quality and be clear and correct.
In the actual radiological examination process, due to factors such as the proficiency of the shooting technique of the detecting personnel, deviation occurs in the quality of the medical image. The existing quality control method for medical images generally carries out post-sampling inspection evaluation and then carries out shooting guidance on detection personnel. The quality control method has no real-time property on the medical image which is subjected to the spot inspection, and when the quality of the medical image which is subjected to the spot inspection is found to be unqualified, a detected person corresponding to the image is already away, and the detected person is notified to take the image again, so that the time and the labor are wasted. In the prior art, there is few methods for directly evaluating the quality of medical images at a shooting site, so that a detector can timely find problems and adjust and re-shoot the problems.
Disclosure of Invention
The research of the applicant shows that: in the shooting field medical image quality control technology, the quality requirements on areas needing to be detected in medical images are basically consistent. Therefore, if the evaluation standard is too high, the medical image is difficult to pass, the requirement on shooting of the detection personnel is high, and resources are wasted. If the evaluation criteria are too low, the quality of the areas (such as the focus of the testee) to be inspected in some medical images is insufficient, and detailed diagnosis cannot be performed.
In view of the above-mentioned drawbacks of the prior art, the present invention is to provide a method and a system for controlling quality of medical images, which can ensure real-time quality control and high quality image requirements of a lesion area.
To achieve the above object, a first aspect of the present invention discloses a medical image quality control method, the method comprising:
step S101: obtaining a medical image of a tested person, and identifying and obtaining each identification unit in the medical image according to the medical image and an inspection position corresponding to the medical image; wherein the medical image comprises an X-ray image, a CT image and a magnetic resonance image, and the identification unit comprises an organ or a tissue of the tested person;
step S102: judging whether the relative azimuth relation of the positions of the identification units on the medical image is matched with a preset relative azimuth, if so, performing step S103; if not, outputting the medical image quality disqualification prompt and entering the step S103 after being confirmed by a detection personnel; the relative azimuth relation comprises the relative position and the relative rotation angle of the identification unit on the medical image;
Step S103: obtaining a past focus of the subject and the medical image according to a past examination and diagnosis report of the subject; dividing the identification unit on the medical image into a focus to-be-identified unit and a non-focus identification unit according to the past focus; the quality evaluation standard of the focus to-be-identified unit corresponds to a first evaluation standard, and the quality evaluation standard of the non-focus identification unit corresponds to a second evaluation standard; the evaluation criteria comprise definition, distortion degree, background fog density value, diagnosis area density value and space exposure area density value, and the image quality requirement corresponding to the first evaluation criteria is higher than the second evaluation criteria;
step S104: evaluating the focus unit to be identified according to the first evaluation standard to obtain a first evaluation result; evaluating the non-focus identification unit according to the second evaluation standard to obtain a second evaluation result;
step S105: and outputting a quality disqualification prompt or a quality qualification prompt of the medical image according to the first evaluation result and the second evaluation result.
Optionally, after the step S103, the method further includes:
Classifying the focus to-be-identified units corresponding to the past focus according to the past inspection diagnosis report of the tested person;
setting a first evaluation standard according to the risk level of the focus unit to be identified; the higher the risk level of the unit to be identified, the more serious the corresponding focus is, and the higher the image quality requirement corresponding to the first evaluation criterion is.
Optionally, the model training step corresponding to the first evaluation criterion and the second evaluation criterion includes:
acquiring a historical medical image dataset, and respectively carrying out first artificial labeling and second artificial labeling on medical images in the historical medical image dataset according to a first evaluation standard and a second evaluation standard; the first manually noted category comprises a first qualified image and a first unqualified image, wherein the first unqualified image is also noted with a first unqualified item to be selected, and the first unqualified item is the unqualified reason of the first unqualified image; the second manually marked category comprises a second qualified image and a second unqualified image, wherein the second unqualified image is also marked with a second to-be-selected unqualified item, and the second to-be-selected unqualified item is the unqualified reason of the second unqualified image;
Inputting the historical medical image dataset into a first training model for training to obtain a first evaluation model corresponding to the first evaluation standard; inputting the historical medical image data set into a second training model for training to obtain a second evaluation model corresponding to the second evaluation standard;
wherein, the step S104 further includes:
and inputting the focus to-be-identified unit into the first evaluation model to obtain a first evaluation result, and inputting the non-focus identification unit into the second evaluation model to obtain a second evaluation result.
Optionally, outputting the quality failure alert of the medical image in step S105 includes:
outputting the first disqualified item or the second disqualified item of the quality disqualification reminding of the medical image.
Optionally, the identifying unit includes at least two evaluation items, where the evaluation items include a normal evaluation item and a ticket overrule evaluation item, and step S104 includes:
in response to one of the evaluation items being failed, determining that the identification unit is failed;
in response to the number of failed common evaluation items in the evaluation items reaching a preset number, judging that the identification unit is failed; wherein the preset number is greater than or equal to 2.
The identification unit comprises a focus to-be-identified unit and a non-focus identification unit, and the focus to-be-identified unit and the non-focus identification unit can be respectively and optionally provided with a common evaluation item and a ticket overrule evaluation item according to practical application, and corresponding evaluation indexes or evaluation standards are set. Furthermore, a global evaluation item may also be provided for the medical image, i.e. each identification unit comprises the evaluation item, for example a gray value; in practical applications, the evaluation item may be evaluated for each identification unit, or the evaluation item may be evaluated for the whole medical image.
Optionally, the step S105 includes:
responding to the quantity that the first evaluation results are qualified and the second evaluation results are qualified to meet the preset quantity, and outputting quality qualification reminding to the medical image;
and outputting a quality failure prompt of the medical image in response to the fact that the number of the first evaluation results being failed and/or the second evaluation results being qualified does not meet the preset number.
Optionally, the step S105 includes:
the second evaluation result is qualified and regarded as one, and the second evaluation result is unqualified and regarded as zero;
weighting each second evaluation result; wherein the more important the non-focus recognition unit corresponding to the second evaluation result is, the greater the weight given to the second evaluation result is;
Summing the weighted second evaluation results, and outputting a quality qualification prompt to the medical image when the calculation result is greater than or equal to a third preset value and the first evaluation results are all qualified; and outputting a quality failure prompt of the medical image when the calculated result is smaller than a third preset value and/or the first evaluation result is failed.
Optionally, after outputting the quality failure reminder of the medical image in step S105, the method further includes:
and generating shooting adjustment comments according to the first evaluation result and the second evaluation result, so that the detection personnel can re-shoot the detected person according to the shooting adjustment comments.
Optionally, after the outputting of the qualified alert for the medical image in step S105, the method further includes:
obtaining an abnormal region according to the medical image; generating a corresponding diagnosis instruction according to the abnormal region;
and generating a radiographic inspection report of the tested person according to the personal information of the tested person and the diagnosis description.
In a second aspect, the invention discloses a medical image quality control system, the system comprising: the device comprises an image recognition module, an azimuth matching judgment module, a recognition unit dividing module, an evaluation module and an output module;
The image recognition module is used for obtaining a medical image of a tested person, and recognizing and obtaining each recognition unit in the medical image according to the medical image and the checking position corresponding to the medical image; wherein the medical image comprises an X-ray image, a CT image and a magnetic resonance image, and the identification unit comprises an organ or a tissue of the tested person;
the azimuth matching judging module is used for judging whether the relative azimuth relation of the positions of the identification units on the medical image is matched with a preset relative azimuth or not, and if so, the identification unit dividing module works; if not, outputting the medical image quality disqualification prompt and enabling the identification unit dividing module to work after confirmation by a detector; the relative azimuth relation comprises the relative position and the relative rotation angle of the identification unit on the medical image;
the identification unit dividing module is used for obtaining the past focus of the tested person and the medical image according to the past examination and diagnosis report of the tested person; dividing the identification unit on the medical image into a focus to-be-identified unit and a non-focus identification unit according to the past focus; the quality evaluation standard of the focus to-be-identified unit corresponds to a first evaluation standard, and the quality evaluation standard of the non-focus identification unit corresponds to a second evaluation standard; the evaluation criteria comprise definition, distortion degree, background fog density value, diagnosis area density value and space exposure area density value, and the image quality requirement corresponding to the first evaluation criteria is higher than the second evaluation criteria;
The evaluation module is used for evaluating the focus unit to be identified according to the first evaluation standard to obtain a first evaluation result; evaluating the non-focus identification unit according to the second evaluation standard to obtain a second evaluation result;
the output module is used for outputting quality disqualification reminding or quality qualification reminding of the medical image according to the first evaluation result and the second evaluation result.
Optionally, the system further comprises: the risk level dividing module and the evaluation standard setting module; the risk level dividing module and the evaluation standard setting module work after the identification unit dividing module works;
the risk level dividing module is used for classifying the focus to-be-identified units corresponding to the past focus according to the past inspection diagnosis report of the tested person;
the evaluation standard setting module is used for setting a first evaluation standard according to the risk level of the focus unit to be identified; the higher the risk level of the unit to be identified, the more serious the corresponding focus is, and the higher the image quality requirement corresponding to the first evaluation criterion is.
Optionally, the system further comprises: the marking module and the training module;
the marking module is used for acquiring a historical medical image data set, and performing first manual marking and second manual marking on medical images in the historical medical image data set according to a first evaluation standard and a second evaluation standard respectively; the first manually noted category comprises a first qualified image and a first unqualified image, wherein the first unqualified image is also noted with a first unqualified item to be selected, and the first unqualified item is the unqualified reason of the first unqualified image; the second manually marked category comprises a second qualified image and a second unqualified image, wherein the second unqualified image is also marked with a second to-be-selected unqualified item, and the second to-be-selected unqualified item is the unqualified reason of the second unqualified image;
the training module is used for inputting the historical medical image data set into a first training model for training to obtain a first evaluation model corresponding to the first evaluation standard; inputting the historical medical image data set into a second training model for training to obtain a second evaluation model corresponding to the second evaluation standard;
Wherein, the evaluation module is specifically used for:
and inputting the focus to-be-identified unit into the first evaluation model to obtain a first evaluation result, and inputting the non-focus identification unit into the second evaluation model to obtain a second evaluation result.
Optionally, the output module outputs a quality failure reminder of the medical image, specifically:
outputting the first disqualified item or the second disqualified item of the quality disqualification reminding of the medical image.
Optionally, the identification unit includes at least two evaluation items, the evaluation items include a normal evaluation item and a ticket overrule evaluation item, and the evaluation module is specifically configured to:
in response to one of the evaluation items being failed, determining that the identification unit is failed;
in response to the number of failed common evaluation items in the evaluation items reaching a preset number, judging that the identification unit is failed; wherein the preset number is greater than or equal to 2.
Optionally, the output module includes: a first pass output sub-module and a first fail output sub-module;
the first qualified output sub-module is used for outputting qualified reminding of the quality of the medical image in response to the fact that the first evaluation results are all qualified and the number of the qualified second evaluation results meets the preset number;
The first disqualification output sub-module is used for outputting quality disqualification reminding of the medical image in response to the fact that the quantity of the unqualified first evaluation result and/or the qualified second evaluation result does not meet the preset quantity.
Optionally, the output module includes: the system comprises a valuation sub-module, a weighting sub-module and a calculation output sub-module;
the assignment submodule is used for regarding the second evaluation result as qualified as one and regarding the second evaluation result as unqualified as zero;
the weighting sub-module is used for weighting each second evaluation result; wherein the more important the non-focus recognition unit corresponding to the second evaluation result is, the greater the weight given to the second evaluation result is;
the calculation output sub-module is used for summing the weighted second evaluation results, and outputting a quality qualification prompt to the medical image when the calculation result is larger than or equal to a third preset value and the first evaluation results are all qualified; and outputting a quality failure prompt of the medical image when the calculated result is smaller than a third preset value and/or the first evaluation result is failed.
Optionally, the system further comprises: the shooting adjustment opinion generation module works after the output module outputs the quality failure reminder of the medical image;
the shooting adjustment opinion generation module is used for generating shooting adjustment opinions according to the first evaluation result and the second evaluation result so that the detection personnel can re-shoot the detected person according to the shooting adjustment opinion.
Optionally, the system further comprises: the diagnosis instruction generation module and the report generation module work after the output module outputs a qualified prompt for the medical image output quality;
the diagnosis instruction generating module is used for obtaining an abnormal area according to the medical image; generating a corresponding diagnosis instruction according to the abnormal region;
the report generation module is used for generating a radiographic inspection report of the tested person according to the personal information of the tested person and the diagnosis description.
The invention has the beneficial effects that: 1. the invention can rapidly evaluate the medical image, so that the quality evaluation can be carried out on the medical image of the tested person obtained at the shooting site (inspection site), the instantaneity is ensured, and the shooting problem is found in time so that the tested person can be shot again by the detection personnel. The invention reduces the complexity of the follow-up spot check re-shooting and effectively ensures the quality of medical images. 2. The method comprises the steps of obtaining a medical image of a tested person, and identifying and obtaining each identification unit in the medical image according to the medical image and an inspection position corresponding to the medical image; judging whether the relative azimuth relation of the positions of the identification units on the medical image is matched with a preset relative azimuth; obtaining past focus of the tested person and the medical image according to the past examination and diagnosis report of the tested person; dividing the identification units on the medical image into units to be identified of the focus and non-focus identification units according to the past focus; evaluating the focus unit to be identified according to a first evaluation standard to obtain a first evaluation result; evaluating the non-focus identification unit according to a second evaluation standard to obtain a second evaluation result; and outputting a quality disqualification prompt or a quality qualification prompt of the medical image according to the first evaluation result and the second evaluation result. According to the invention, the identification units in the medical image of the tested person are divided into the focus to-be-identified unit and the non-focus identification unit in the mode, so that the focus to-be-identified unit can be subjected to higher image quality requirements, and the focus to-be-identified unit can be diagnosed in detail. The invention adopts different image quality requirements to treat the focus to-be-identified unit and the non-focus identification unit, which not only can ensure that the non-focus identification unit can not cause resource waste because the medical image is wholly judged to be unqualified due to the too high image quality requirement, but also can ensure that the focus to-be-identified unit can not be diagnosed in detail because the image quality requirement is too low. 3. The invention further classifies the focus to-be-identified units corresponding to the past focus and sets a first evaluation standard according to the risk level of the focus to-be-identified units. The method can ensure that the more serious the focus is, the higher the image quality requirement is, so that the more serious the disease can be diagnosed in detail, and the life safety of a tested person is ensured. 4. According to the invention, the medical images are evaluated as a whole according to the number of qualified second evaluation results meeting the preset number, so that the method is simple and quick. 5. According to the invention, the medical image is evaluated through the weighted summation of the second evaluation results, and the medical image can be evaluated according to the importance degree of the non-focus identification unit, so that the method is more targeted and scientific. 6. The invention can also generate shooting adjustment comments according to the first evaluation result and the second evaluation result. The camera shooting device is convenient for a detector to adjust in a targeted manner to shoot again, and blindness of adjustment is avoided. In conclusion, the invention can monitor the quality of the medical image at the shooting site and can realize timely remedy of shooting; meanwhile, the invention can realize the high-quality requirement of the image of the focus area according to the actual situation of the testee so as to ensure that the focus area can be diagnosed in detail.
Drawings
FIG. 1 is a flowchart of a medical image quality control method according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a medical image quality control system according to an embodiment of the present invention.
Description of the embodiments
The invention discloses a medical image quality control method, which can be realized by appropriately improving technical details by a person skilled in the art by referring to the content of the text. It is expressly noted that all such similar substitutions and modifications will be apparent to those skilled in the art, and are deemed to be included in the present invention. While the methods and applications of this invention have been described in terms of preferred embodiments, it will be apparent to those skilled in the relevant art that variations and modifications can be made in the methods and applications described herein, and in the practice and application of the techniques of this invention, without departing from the spirit or scope of the invention.
The research of the applicant shows that: in the shooting field medical image quality control technology, the quality requirements on areas needing to be detected in medical images are basically consistent. Therefore, if the evaluation standard is too high, the medical image is difficult to pass, the requirement on shooting of the detection personnel is high, and resources are wasted. If the evaluation criteria are too low, the quality of the areas (such as the focus of the testee) to be inspected in some medical images is insufficient, and detailed diagnosis cannot be performed. The images are not clear enough and the doctor can only get the approximate illness state. The clearer the image, the more information the physician obtains, the more detailed the diagnosis, the more likely the follow-up treatment will be assisted.
Accordingly, an embodiment of the present invention provides a medical image quality control method, as shown in fig. 1, including:
step S101: and obtaining a medical image of the tested person, and identifying and obtaining each identification unit in the medical image according to the medical image and the checking part corresponding to the medical image.
The medical image comprises an X-ray image, a CT image and a magnetic resonance image, and the identification unit comprises organs or tissues of a tested person.
In one embodiment, each identification element is an organ, bone, tissue, etc., identified by image recognition features.
It should be noted that, the medical image is generally used for examining diseases in the human body, and is used for guaranteeing the life health of the tested person.
Step S102: judging whether the relative azimuth relation of the positions of the identification units on the medical image is matched with a preset relative azimuth, if so, performing step S103; if not, outputting a medical image quality disqualification prompt, and entering step S103 after confirmation by a detector.
The relative positional relationship includes a relative position of the identification unit on the medical image and a relative rotation angle.
It should be noted that, step S102 outputs a medical image quality failure indicating that each recognition unit is not at a preset position, such as that the vertebra is not in the middle, left and right lung lobe is asymmetric, etc. These may occur for two reasons: 1. the detection personnel do not shoot well when shooting; 2. the physical causes of the tested person (such as rickets, human deformity, etc.). If it is due to the physical cause of the subject himself, the next step S103 may be performed normally. This is also the reason why the step S103 is entered after the confirmation by the inspector when the medical image quality reject reminder is output, and the inspector confirms whether the individual recognition units caused by the physical cause of the inspector are not at the preset positions. The relative positional relationship includes a position area and a direction (orientation).
Step S103: obtaining past focus of the tested person and the medical image according to the past examination and diagnosis report of the tested person; dividing the identification units on the medical image into units to be identified of the focus and non-focus identification units according to the past focus; the quality evaluation criterion of the lesion to be identified unit corresponds to the first evaluation criterion and the quality evaluation criterion of the non-lesion identification unit corresponds to the second evaluation criterion.
The evaluation criteria comprise definition, distortion degree, background fog density value, diagnosis area density value and space exposure area density value, and the image quality requirement corresponding to the first evaluation criteria is higher than that of the second evaluation criteria.
For example, if the lung of the subject is malignant according to the past examination and diagnosis report of the subject, the lung area where the tumor is located is set as the unit to be identified of the focus.
The definition is the definition of each fine shadow and the boundary thereof on the image; the distortion degree is the deviation between the medical image and the structure of the actual examination part; the background haze density value (minimum density) is the density produced by the partially reduced silver after development processing without exposing the photosensitive material to light; the diagnostic area density value and the blank exposure area density value are respectively densities generated by partially reduced silver of two areas in the medical image after development processing. (typically, the diagnostic region is a region to be detected, and the empty exposure region is a region not to be detected). The two requirements are to distinguish between them, and the inability to come too close together results in the inability to distinguish between diagnostic and empty exposure areas.
Optionally, after step S103, the method further includes:
classifying the focus to-be-identified units corresponding to the past focus according to the past inspection diagnosis report of the tested person;
setting a first evaluation standard according to the risk level of a focus unit to be identified; the higher the risk level of the unit to be identified, the more serious the corresponding focus, and the higher the image quality requirement corresponding to the first evaluation standard. It should be noted that, the more serious the focus, the higher the requirement on image quality, the clearer the doctor can observe the focus of the medical image, and the more detailed the diagnosis report is obtained.
Step S104: evaluating the focus unit to be identified according to a first evaluation standard to obtain a first evaluation result; and evaluating the non-focus recognition unit according to a second evaluation standard to obtain a second evaluation result.
In one embodiment, step S104 includes:
collecting first evaluation parameters of a focus unit to be identified, judging whether the first evaluation parameters meet a first evaluation standard, and if so, judging that a first evaluation result is qualified; if not, the first evaluation result is unqualified;
collecting second evaluation parameters of the non-focus identification unit, judging whether the second evaluation parameters meet the first evaluation standard, and if so, judging that the first evaluation result is qualified; if not, the first evaluation result is disqualified.
In a specific embodiment, the identification unit includes at least two evaluation items, the evaluation items include a normal evaluation item and a ticket overrule evaluation item, and step S104 includes:
in response to one of the evaluation items overruling the evaluation item as failed, determining that the identification unit is failed;
in response to the number of unqualified common evaluation items in the evaluation items reaching a preset number, judging that the identification unit is unqualified; wherein the preset number is greater than or equal to 2.
The identification unit comprises a focus to-be-identified unit and a non-focus identification unit, and the focus to-be-identified unit and the non-focus identification unit can be respectively and optionally provided with a common evaluation item and a ticket overrule evaluation item according to practical application, and corresponding evaluation indexes or evaluation standards are set. Furthermore, a global evaluation item may also be provided for the medical image, i.e. each identification unit comprises the evaluation item, for example a gray value; in practical applications, the evaluation item may be evaluated for each identification unit, or the evaluation item may be evaluated for the entire medical image.
Step S105: and outputting a quality disqualification prompt or a quality qualification prompt of the medical image according to the first evaluation result and the second evaluation result.
In a specific embodiment, the model training step corresponding to the first evaluation criterion and the second evaluation criterion includes:
Acquiring a historical medical image data set, and respectively carrying out first artificial labeling and second artificial labeling on medical images in the historical medical image data set according to a first evaluation standard and a second evaluation standard; the first manually marked category comprises a first qualified image and a first unqualified image, wherein the first unqualified image is also marked with a first unqualified item to be selected, and the first unqualified item is the unqualified reason of the first unqualified image; the second manually marked category comprises a second qualified image and a second unqualified image, the second unqualified image is also marked with a second to-be-selected unqualified item, and the second to-be-selected unqualified item is the unqualified reason of the second unqualified image;
inputting the historical medical image dataset into a first training model for training to obtain a first evaluation model corresponding to a first evaluation standard; inputting the historical medical image data set into a second training model for training to obtain a second evaluation model corresponding to a second evaluation standard;
in this embodiment, step S104 further includes:
and inputting the focus to-be-identified unit into a first evaluation model to obtain a first evaluation result, and inputting the non-focus identification unit into a second evaluation model to obtain a second evaluation result.
In this embodiment, outputting a quality failure alert of the medical image in step S105 includes:
outputting a first disqualification item or a second disqualification item of the quality disqualification reminding of the medical image.
Note that, the present invention is not limited to the above-described embodiments. The unqualified item is output to know the problem in the medical image shooting process and is used for guiding the re-shooting.
In one embodiment, step S105 includes:
responding to the quantity that each first evaluation result is qualified and the second evaluation result is qualified to meet the preset quantity, and outputting a quality qualification prompt to the medical image;
and outputting a quality failure prompt of the medical image in response to the fact that the number of the unqualified first evaluation results and/or the number of the unqualified second evaluation results does not meet the preset number.
It should be noted that the second evaluation result corresponds to a non-lesion recognition unit and thus does not need to each very strictly require image quality, but mostly satisfies the image quality requirement. Therefore, when the medical image is qualified, the first evaluation results are required to be qualified, and the number of the qualified second evaluation results meets the preset number. The image quality is measured by the preset number, so that the method is simple and quick.
In another embodiment, step S105 includes:
the second evaluation result is qualified and regarded as one, and the second evaluation result is unqualified and regarded as zero;
weighting each second evaluation result; wherein, the more important the non-focus identification unit corresponding to the second evaluation result is, the greater the weight given to the second evaluation result is;
summing the weighted second evaluation results, and outputting a quality qualification prompt to the medical image when the calculation result is greater than or equal to a third preset value and the first evaluation results are all qualified; and outputting a quality failure prompt of the medical image when the calculated result is smaller than a third preset value and/or the first evaluation result is failed.
It should be noted that this embodiment does not use a number scheme for the second evaluation result, but uses weighted summation. This makes it possible to discriminate the image quality according to the importance degree of each non-lesion recognition unit.
In one case, the third preset value of qualified medical image output quality is 4, and the medical image is qualified when the quality evaluation weighted sum value of the item identification unit is greater than 4. For example, the second result for the heart is acceptable and the weight is 3, the second result for the lung lobes is acceptable and the weight is 2, and the second result for the vertebrae is unacceptable and the weight is 1.3+2>4, so that the medical image is qualified.
Optionally, after outputting the quality failure reminder of the medical image in step S105, the method further includes:
and generating shooting adjustment comments according to the first evaluation result and the second evaluation result so that the detection personnel can re-shoot the detected person according to the shooting adjustment comments.
It should be noted that the shooting adjustment opinion can better help the inspector to correctly re-shoot, and ensure the re-shooting quality.
Optionally, after the medical image output quality qualification reminding in step S105, the method further includes:
obtaining an abnormal region according to the medical image; generating a corresponding diagnosis instruction according to the abnormal region;
a radiological image examination report of the subject is generated based on the personal information and the diagnostic instructions of the subject.
It should be noted that, the embodiment can automatically generate the corresponding report, thereby reducing the workload of the detection personnel.
In a specific application process, if the medical image is a chest orthographic image;
step S301: acquiring a chest orthographic image, and identifying each identification unit in the chest orthographic image;
step S302: judging whether the chest orthotopic image is in the middle position of the display medium, and comprises the outer edges of two side ribs, diaphragm angles of two side ribs and soft tissues on the tip of the lung; whether the scapula is projected outside the lung field, whether the thoracic vertebra is positioned in the center of the image, whether bilateral sternoclavicular joints are symmetrically displayed, whether the bilateral clavicle positions are equal in height and tend to be horizontal;
Step S303: obtaining past focus of the positive images of the tested person and the chest according to the past examination and diagnosis report of the tested person (assuming that the left lung of the tested person has malignant tumor); and determining the identification unit corresponding to the left lung as a focus to-be-identified unit, and determining other units as non-focus identification units.
Step S304: evaluating the focus unit to be identified according to a first evaluation standard to obtain a first evaluation result; and evaluating the non-focus recognition unit according to a second evaluation standard to obtain a second evaluation result.
For example, the diagnostic region density value in the first evaluation criterion: d=0.24 to 2.1, diagnostic region density value in the second evaluation criterion: d=0.25 to 2.0, and the image quality requirement corresponding to the first evaluation criterion is higher than that of the second evaluation criterion.
The front radiographic image evaluation criteria for the chest are listed below, wherein the left column is the first evaluation criteria and the right column is the second evaluation criteria.
TABLE 1 evaluation of the qualified Standard Range values for the front radiographic images of the chest
Step S305: and outputting a quality unqualified prompt or a quality qualified prompt to the chest orthotopic image according to the first evaluation result and the second evaluation result.
The embodiment of the invention can acquire the medical image of the tested person to carry out quality evaluation at the shooting site (inspection site), can ensure real-time performance, and can find the shooting problem in time so as to enable the testing personnel to shoot the tested person again. The embodiment of the invention reduces the complexity of the follow-up spot check re-shooting and effectively ensures the quality of medical images.
According to the embodiment of the invention, the medical image of the tested person is obtained, and each identification unit in the medical image is identified and obtained according to the medical image and the checking part corresponding to the medical image; judging whether the relative azimuth relation of the positions of the identification units on the medical image is matched with a preset relative azimuth; obtaining past focus of the tested person and the medical image according to the past examination and diagnosis report of the tested person; dividing the identification units on the medical image into units to be identified of the focus and non-focus identification units according to the past focus; evaluating the focus unit to be identified according to a first evaluation standard to obtain a first evaluation result; evaluating the non-focus identification unit according to a second evaluation standard to obtain a second evaluation result; and outputting a quality disqualification prompt or a quality qualification prompt of the medical image according to the first evaluation result and the second evaluation result. According to the embodiment of the invention, the identification units in the medical image of the tested person are divided into the focus to-be-identified unit and the non-focus identification unit in the mode, so that the focus to-be-identified unit can be subjected to higher image quality requirements, and the focus to-be-identified unit can be diagnosed in detail. The embodiment of the invention adopts different image quality requirements to treat the focus to-be-identified unit and the non-focus identification unit, thereby not only ensuring that the non-focus identification unit can not cause resource waste because the whole medical image is judged to be unqualified due to the too high image quality requirement, but also ensuring that the focus to-be-identified unit can not be diagnosed in detail due to the too low image quality requirement.
The embodiment of the invention further classifies the focus to-be-identified units corresponding to the past focus and sets the first evaluation standard according to the risk level of the focus to-be-identified units. The method can ensure that the more serious the focus is, the higher the image quality requirement is, so that the more serious the disease can be diagnosed in detail, and the life safety of a tested person is ensured.
According to the embodiment of the invention, the medical images are evaluated as a whole according to the number of qualified second evaluation results meeting the preset number, so that the method is simple and quick.
According to the embodiment of the invention, the medical image is evaluated through the weighted summation of the second evaluation results, and the medical image can be evaluated according to the importance degree of the non-focus identification unit, so that the method is more targeted and scientific.
The embodiment of the invention can also generate shooting adjustment comments according to the first evaluation result and the second evaluation result. The camera shooting device is convenient for a detector to adjust in a targeted manner to shoot again, and blindness of adjustment is avoided.
In summary, the embodiment of the invention can monitor the quality of the medical image at the shooting site, and can realize timely remedy of shooting; meanwhile, the embodiment of the invention can realize the high-quality requirement of the image of the focus area according to the actual situation of the testee so as to ensure that the focus area can be diagnosed in detail.
Based on the above-mentioned provided medical image quality control method, the embodiment of the present invention further provides a medical image quality control system, as shown in fig. 2, where the system includes: an image recognition module 201, an azimuth matching judgment module 202, a recognition unit division module 203, an evaluation module 204 and an output module 205;
the image recognition module 201 is configured to obtain a medical image of a subject, and recognize and obtain each recognition unit in the medical image according to the medical image and an inspection position corresponding to the medical image; the medical image comprises an X-ray image, a CT image and a magnetic resonance image, and the identification unit comprises organs or tissues of a tested person;
the azimuth matching judging module 202 is configured to judge whether a relative azimuth relation of the position of each identification unit on the medical image is matched with a preset relative azimuth, and if yes, enable the identification unit dividing module 203 to work; if not, outputting a medical image quality disqualification prompt and enabling the identification unit dividing module 203 to work after confirmation by a detector; the relative azimuth relation comprises the relative position and the relative rotation angle of the identification unit on the medical image;
the identification unit dividing module 203 is configured to obtain a past focus of the subject and the medical image according to a past inspection diagnosis report of the subject; dividing the identification units on the medical image into units to be identified of the focus and non-focus identification units according to the past focus; the quality evaluation standard of the focus to-be-identified unit corresponds to the first evaluation standard, and the quality evaluation standard of the non-focus identification unit corresponds to the second evaluation standard; the evaluation criteria comprise definition, distortion degree, background fog density value, diagnosis area density value and space exposure area density value, and the image quality requirement corresponding to the first evaluation criteria is higher than that of the second evaluation criteria;
The evaluation module 204 is configured to evaluate the unit to be identified of the lesion according to a first evaluation criterion, so as to obtain a first evaluation result; evaluating the non-focus identification unit according to a second evaluation standard to obtain a second evaluation result;
and the output module 205 is configured to output a quality failure reminder or a quality failure reminder for the medical image according to the first evaluation result and the second evaluation result.
Optionally, the system further comprises: the risk level dividing module and the evaluation standard setting module; the risk level dividing module and the evaluation criterion setting module operate after the identification unit dividing module 203 operates;
the risk level dividing module is used for classifying the focus to-be-identified units corresponding to the past focus according to the past inspection diagnosis report of the tested person;
the evaluation standard setting module is used for setting a first evaluation standard according to the risk level of the unit to be identified of the focus; the higher the risk level of the unit to be identified, the more serious the corresponding focus, and the higher the image quality requirement corresponding to the first evaluation standard.
Optionally, the system further comprises: the marking module and the training module;
the marking module is used for acquiring a historical medical image data set, and respectively carrying out first manual marking and second manual marking on medical images in the historical medical image data set according to a first evaluation standard and a second evaluation standard; the first manually marked category comprises a first qualified image and a first unqualified image, wherein the first unqualified image is also marked with a first unqualified item to be selected, and the first unqualified item is the unqualified reason of the first unqualified image; the second manually marked category comprises a second qualified image and a second unqualified image, the second unqualified image is also marked with a second to-be-selected unqualified item, and the second to-be-selected unqualified item is the unqualified reason of the second unqualified image;
The training module is used for inputting the historical medical image data set into the first training model for training to obtain a first evaluation model corresponding to the first evaluation standard; inputting the historical medical image data set into a second training model for training to obtain a second evaluation model corresponding to a second evaluation standard;
wherein, the evaluation module 204 is specifically configured to:
and inputting the focus to-be-identified unit into a first evaluation model to obtain a first evaluation result, and inputting the non-focus identification unit into a second evaluation model to obtain a second evaluation result.
Optionally, the output module outputs a quality failure reminder of the medical image, specifically:
outputting a first disqualification item or a second disqualification item of the quality disqualification reminding of the medical image.
Optionally, the identifying unit includes at least two evaluation items, the evaluation items include a general evaluation item and a ticket overrule evaluation item, and the evaluation module 204 is specifically configured to:
in response to one of the evaluation items overruling the evaluation item as failed, determining that the identification unit is failed;
in response to the number of unqualified common evaluation items in the evaluation items reaching a preset number, judging that the identification unit is unqualified; wherein the preset number is greater than or equal to 2.
Optionally, the output module 205 includes: a first pass output sub-module and a first fail output sub-module;
the first qualified output sub-module is used for responding to the fact that the quantity of the qualified first evaluation results and the qualified second evaluation results meet the preset quantity and outputting quality qualified prompt to the medical image;
the first disqualification output sub-module is used for outputting a quality disqualification prompt of the medical image in response to the fact that the quantity of the unqualified first evaluation result and/or the qualified second evaluation result does not meet the preset quantity.
Optionally, the output module 205 includes: the system comprises a valuation sub-module, a weighting sub-module and a calculation output sub-module;
the assignment sub-module is used for regarding the second evaluation result as qualified as one and regarding the second evaluation result as unqualified as zero;
the weighting sub-module is used for weighting each second evaluation result; wherein, the more important the non-focus identification unit corresponding to the second evaluation result is, the greater the weight given to the second evaluation result is;
the calculation output sub-module is used for summing the weighted second evaluation results, and outputting a quality qualification prompt to the medical image when the calculation result is larger than or equal to a third preset value and the first evaluation results are all qualified; and outputting a quality failure prompt of the medical image when the calculated result is smaller than a third preset value and/or the first evaluation result is failed.
Optionally, the system further comprises: the shooting adjustment opinion generation module works after the output module 205 outputs a quality failure reminder for the medical image;
and the shooting adjustment opinion generation module is used for generating shooting adjustment opinions according to the first evaluation result and the second evaluation result so as to enable the detection personnel to re-shoot the detected person according to the shooting adjustment opinions.
Optionally, the system further comprises: a diagnostic specification generation module and a report generation module that operate after the output module 205 outputs a quality qualification reminder for the medical image;
the diagnosis instruction generation module is used for obtaining an abnormal region according to the medical image; generating a corresponding diagnosis instruction according to the abnormal region;
and the report generation module is used for generating a radiographic inspection report of the tested person according to the personal information and the diagnosis description of the tested person.
The system of the embodiment of the invention can acquire the medical image of the tested person to carry out quality evaluation on a shooting site (checking site), can ensure real-time performance, and can find the shooting problem in time so as to enable the testing personnel to shoot the tested person again. The system of the embodiment of the invention reduces the complexity of re-shooting in later spot check and effectively ensures the quality of medical images.
The system of the embodiment of the invention obtains the medical image of the tested person, and identifies and obtains each identification unit in the medical image according to the medical image and the checking part corresponding to the medical image; judging whether the relative azimuth relation of the positions of the identification units on the medical image is matched with a preset relative azimuth; obtaining past focus of the tested person and the medical image according to the past examination and diagnosis report of the tested person; dividing the identification units on the medical image into units to be identified of the focus and non-focus identification units according to the past focus; evaluating the focus unit to be identified according to a first evaluation standard to obtain a first evaluation result; evaluating the non-focus identification unit according to a second evaluation standard to obtain a second evaluation result; and outputting a quality disqualification prompt or a quality qualification prompt of the medical image according to the first evaluation result and the second evaluation result. The system of the embodiment of the invention divides the identification unit in the medical image of the tested person into the focus to-be-identified unit and the non-focus identification unit through the mode, and can carry out higher image quality requirements on the focus to-be-identified unit, thereby ensuring that the focus to-be-identified unit can be diagnosed in detail. The system of the embodiment of the invention adopts different image quality requirements to treat the focus to-be-identified unit and the non-focus identification unit, thereby not only ensuring that the non-focus identification unit can not cause resource waste because the whole medical image is judged to be unqualified due to the too high image quality requirement, but also ensuring that the focus to-be-identified unit can not be diagnosed in detail due to the too low image quality requirement.
The system of the embodiment of the invention further classifies the focus to-be-identified units corresponding to the past focus and sets a first evaluation standard according to the risk level of the focus to-be-identified units. The method can ensure that the more serious the focus is, the higher the image quality requirement is, so that the more serious the disease can be diagnosed in detail, and the life safety of a tested person is ensured.
The system provided by the embodiment of the invention evaluates the whole medical image simply and quickly by meeting the preset number through the qualified number of the second evaluation result.
The system of the embodiment of the invention evaluates the whole medical image through the weighted summation of the second evaluation result, and can evaluate the whole medical image according to the importance degree of the non-focus identification unit, thereby being more targeted and scientific.
The system of the embodiment of the invention can also generate shooting adjustment comments according to the first evaluation result and the second evaluation result. The camera shooting device is convenient for a detector to adjust in a targeted manner to shoot again, and blindness of adjustment is avoided.
In summary, the system of the embodiment of the invention can monitor the quality of the medical image at the shooting site, and can realize timely remedy of shooting; meanwhile, the system of the embodiment of the invention can realize the high-quality requirement of the image of the focus area according to the actual situation of the testee so as to ensure that the focus area can be diagnosed in detail.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.
In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.
The foregoing is merely illustrative of the preferred embodiments of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.
Claims (10)
1. A medical image quality control method, the method comprising:
step S101: obtaining a medical image of a tested person, and identifying and obtaining each identification unit in the medical image according to the medical image and an inspection position corresponding to the medical image; wherein the medical image comprises an X-ray image, a CT image and a magnetic resonance image, and the identification unit comprises an organ or a tissue of the tested person;
step S102: judging whether the relative azimuth relation of the positions of the identification units on the medical image is matched with a preset relative azimuth, if so, performing step S103; if not, outputting the medical image quality disqualification prompt and entering the step S103 after being confirmed by a detection personnel; the relative azimuth relation comprises the relative position and the relative rotation angle of the identification unit on the medical image;
step S103: obtaining a past focus of the subject and the medical image according to a past examination and diagnosis report of the subject; dividing the identification unit on the medical image into a focus to-be-identified unit and a non-focus identification unit according to the past focus; the quality evaluation standard of the focus to-be-identified unit corresponds to a first evaluation standard, and the quality evaluation standard of the non-focus identification unit corresponds to a second evaluation standard; the evaluation criteria comprise definition, distortion degree, background fog density value, diagnosis area density value and space exposure area density value, and the image quality requirement corresponding to the first evaluation criteria is higher than the second evaluation criteria;
Step S104: evaluating the focus unit to be identified according to the first evaluation standard to obtain a first evaluation result; evaluating the non-focus identification unit according to the second evaluation standard to obtain a second evaluation result;
step S105: and outputting a quality disqualification prompt or a quality qualification prompt of the medical image according to the first evaluation result and the second evaluation result.
2. The medical image quality control method according to claim 1, wherein after the step S103, the method further comprises:
classifying the focus to-be-identified units corresponding to the past focus according to the past inspection diagnosis report of the tested person;
setting a first evaluation standard according to the risk level of the focus unit to be identified; the higher the risk level of the unit to be identified, the more serious the corresponding focus is, and the higher the image quality requirement corresponding to the first evaluation criterion is.
3. The medical image quality control method according to claim 1, wherein the model training step corresponding to the first evaluation criterion and the second evaluation criterion includes:
acquiring a historical medical image dataset, and respectively carrying out first artificial labeling and second artificial labeling on medical images in the historical medical image dataset according to a first evaluation standard and a second evaluation standard; the first manually noted category comprises a first qualified image and a first unqualified image, wherein the first unqualified image is also noted with a first unqualified item to be selected, and the first unqualified item is the unqualified reason of the first unqualified image; the second manually marked category comprises a second qualified image and a second unqualified image, wherein the second unqualified image is also marked with a second to-be-selected unqualified item, and the second to-be-selected unqualified item is the unqualified reason of the second unqualified image;
Inputting the historical medical image dataset into a first training model for training to obtain a first evaluation model corresponding to the first evaluation standard; inputting the historical medical image data set into a second training model for training to obtain a second evaluation model corresponding to the second evaluation standard;
wherein, the step S104 further includes:
and inputting the focus to-be-identified unit into the first evaluation model to obtain a first evaluation result, and inputting the non-focus identification unit into the second evaluation model to obtain a second evaluation result.
4. The medical image quality control method according to claim 3, wherein outputting the quality reject notification of the medical image in step S105 comprises:
outputting the first disqualified item or the second disqualified item of the quality disqualification reminding of the medical image.
5. The medical image quality control method according to claim 1, wherein the identification unit includes at least two evaluation items including a normal evaluation item and a ticket overrule evaluation item, and step S104 includes:
in response to one of the evaluation items being failed, determining that the identification unit is failed;
In response to the number of failed common evaluation items in the evaluation items reaching a preset number, judging that the identification unit is failed; wherein the preset number is greater than or equal to 2.
6. The medical image quality control method according to claim 1, wherein the step S105 includes:
responding to the quantity that the first evaluation results are qualified and the second evaluation results are qualified to meet the preset quantity, and outputting quality qualification reminding to the medical image;
and outputting a quality failure prompt of the medical image in response to the fact that the number of the first evaluation results being failed and/or the second evaluation results being qualified does not meet the preset number.
7. The medical image quality control method according to claim 1, wherein the step S105 includes:
the second evaluation result is qualified and regarded as one, and the second evaluation result is unqualified and regarded as zero;
weighting each second evaluation result; wherein the more important the non-focus recognition unit corresponding to the second evaluation result is, the greater the weight given to the second evaluation result is;
summing the weighted second evaluation results, and outputting a quality qualification prompt to the medical image when the calculation result is greater than or equal to a third preset value and the first evaluation results are all qualified; and outputting a quality failure prompt of the medical image when the calculated result is smaller than a third preset value and/or the first evaluation result is failed.
8. The medical image quality control method according to claim 1, wherein after outputting the quality reject reminder of the medical image in step S105, the method further comprises:
and generating shooting adjustment comments according to the first evaluation result and the second evaluation result, so that the detection personnel can re-shoot the detected person according to the shooting adjustment comments.
9. The medical image quality control method according to claim 1, wherein after outputting a quality-eligible reminder for the medical image in the step S105, the method further comprises:
obtaining an abnormal region according to the medical image; generating a corresponding diagnosis instruction according to the abnormal region;
and generating a radiographic inspection report of the tested person according to the personal information of the tested person and the diagnosis description.
10. A medical image quality control system, the system comprising: the device comprises an image recognition module, an azimuth matching judgment module, a recognition unit dividing module, an evaluation module and an output module;
the image recognition module is used for obtaining a medical image of a tested person, and recognizing and obtaining each recognition unit in the medical image according to the medical image and the checking position corresponding to the medical image; wherein the medical image comprises an X-ray image, a CT image and a magnetic resonance image, and the identification unit comprises an organ or a tissue of the tested person;
The azimuth matching judging module is used for judging whether the relative azimuth relation of the positions of the identification units on the medical image is matched with a preset relative azimuth or not, and if so, the identification unit dividing module works; if not, outputting the medical image quality disqualification prompt and enabling the identification unit dividing module to work after confirmation by a detector; the relative azimuth relation comprises the relative position and the relative rotation angle of the identification unit on the medical image;
the identification unit dividing module is used for obtaining the past focus of the tested person and the medical image according to the past examination and diagnosis report of the tested person; dividing the identification unit on the medical image into a focus to-be-identified unit and a non-focus identification unit according to the past focus; the quality evaluation standard of the focus to-be-identified unit corresponds to a first evaluation standard, and the quality evaluation standard of the non-focus identification unit corresponds to a second evaluation standard; the evaluation criteria comprise definition, distortion degree, background fog density value, diagnosis area density value and space exposure area density value, and the image quality requirement corresponding to the first evaluation criteria is higher than the second evaluation criteria;
The evaluation module is used for evaluating the focus unit to be identified according to the first evaluation standard to obtain a first evaluation result; evaluating the non-focus identification unit according to the second evaluation standard to obtain a second evaluation result;
the output module is used for outputting quality disqualification reminding or quality qualification reminding of the medical image according to the first evaluation result and the second evaluation result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310209400.2A CN116092642B (en) | 2023-03-07 | 2023-03-07 | Medical image quality control method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310209400.2A CN116092642B (en) | 2023-03-07 | 2023-03-07 | Medical image quality control method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116092642A true CN116092642A (en) | 2023-05-09 |
| CN116092642B CN116092642B (en) | 2023-06-20 |
Family
ID=86204601
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310209400.2A Active CN116092642B (en) | 2023-03-07 | 2023-03-07 | Medical image quality control method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116092642B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010051456A (en) * | 2008-08-27 | 2010-03-11 | Fujifilm Corp | Apparatus, method and program for evaluating medical image |
| CN104615915A (en) * | 2015-01-15 | 2015-05-13 | 于飞 | Computer auxiliary image quality analysis control method and system for medical radiation images |
| CN111986182A (en) * | 2020-08-25 | 2020-11-24 | 卫宁健康科技集团股份有限公司 | Auxiliary diagnosis method, system, electronic device and storage medium |
| CN113052795A (en) * | 2018-12-28 | 2021-06-29 | 上海联影智能医疗科技有限公司 | X-ray chest radiography image quality determination method and device |
| CN114334097A (en) * | 2022-01-05 | 2022-04-12 | 深圳市步锐生物科技有限公司 | Automatic assessment method based on lesion progress on medical image and related product |
-
2023
- 2023-03-07 CN CN202310209400.2A patent/CN116092642B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010051456A (en) * | 2008-08-27 | 2010-03-11 | Fujifilm Corp | Apparatus, method and program for evaluating medical image |
| CN104615915A (en) * | 2015-01-15 | 2015-05-13 | 于飞 | Computer auxiliary image quality analysis control method and system for medical radiation images |
| CN113052795A (en) * | 2018-12-28 | 2021-06-29 | 上海联影智能医疗科技有限公司 | X-ray chest radiography image quality determination method and device |
| CN111986182A (en) * | 2020-08-25 | 2020-11-24 | 卫宁健康科技集团股份有限公司 | Auxiliary diagnosis method, system, electronic device and storage medium |
| CN114334097A (en) * | 2022-01-05 | 2022-04-12 | 深圳市步锐生物科技有限公司 | Automatic assessment method based on lesion progress on medical image and related product |
Non-Patent Citations (2)
| Title |
|---|
| PEI-SHAN HO: "Machine learning framework for automatic image quality evaluation involving a mammographic American College of Radiology phantom", 《PHYSICA MEDICA》, vol. 102, pages 1 - 8, XP087189133, DOI: 10.1016/j.ejmp.2022.08.004 * |
| 冯庆宇: "第二讲 数字化医疗影像临床质量评价", 《中国医疗设备》, vol. 25, no. 8, pages 148 - 152 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116092642B (en) | 2023-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Robinson | Radiology's Achilles' heel: error and variation in the interpretation of the Röntgen image. | |
| US20060100507A1 (en) | Method for evaluation of medical findings in three-dimensional imaging, in particular in mammography | |
| US9918691B2 (en) | Device and method for determining image quality of a radiogram image | |
| US9668712B2 (en) | Method and system for quantitative imaging | |
| EP3373814B1 (en) | X-ray image inhalation quality monitoring | |
| JP2008537691A (en) | How to expand the field of imaging software in diagnostic workups | |
| EP3644273A1 (en) | System for determining image quality parameters for medical images | |
| KR20240013724A (en) | Artificial Intelligence Training Using a Multipulse X-ray Source Moving Tomosynthesis Imaging System | |
| Halsted et al. | Diagnostic errors by radiology residents in interpreting pediatric radiographs in an emergency setting | |
| EP4027887B1 (en) | Inhalation metric for chest x-ray images | |
| CN116092642B (en) | Medical image quality control method and system | |
| Geertse et al. | Added value of prereading screening mammograms for breast cancer by radiologic technologists on early screening outcomes | |
| Chen et al. | Image quality control in lumbar spine radiography using enhanced U-Net neural networks | |
| Niu et al. | A deep-learning system for automatic detection of osteoporotic vertebral compression fractures at thoracolumbar junction using low-dose computed tomography images | |
| TWI810680B (en) | Method and system of analyzing anteroposterior pelvic radiographs | |
| Kumarihami et al. | Development of Brief Image Quality Evaluation Criteria for Digital OrthoPantomography (OPG) Images in Dental Radiography | |
| US10824892B2 (en) | System and method for imaging an object via pre-shot processing | |
| EP4134972A1 (en) | Machine learning based quality assessment of medical imagery and its use in facilitating imaging operations | |
| EP4202826A1 (en) | X-ray image analysis system, x-ray imaging system and method for analysing an x-ray image | |
| Catalina et al. | External Validation of an Artificial Intelligence Algorithm for the Analysis of Chest X-Rays in the Primary Care Setting | |
| Ween et al. | Sharpness and noise in digital chest radiographs, assessed by visual rating | |
| Gunawan et al. | Analysis of Window Width Variations on MSCT Stonegraphy Anatomic Image Information at Sanjiwani Hospital of Gianyar Regency | |
| Jessop | Task-specific observer performance and image quality: direct and indirect relationships in low-dose CT images | |
| CN115376645A (en) | Chest lateral X-ray film intelligent evaluation system and method | |
| CN115359879A (en) | Chest X-ray film technology evaluation method and system |
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 |