Method for establishing a prenatal diagnosis model and diagnostic system for aortic coarctation

Through the aortic constriction prenatal diagnosis system, the aortic constriction prenatal diagnosis model is analyzed using the fetal number of weeks and the measured values ​​of multiple ultrasound indicators, which solves the problem of misdiagnosis and missed diagnosis in the existing technology and improves the accuracy of diagnosis Sex and efficiency reduce medical burden.

CN116189905BActive Publication Date: 2024-11-12FUWAI HOSPITAL CHINESE ACAD OF MEDICAL SCI & PEKING UNION MEDICAL COLLEGE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310211479.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-02-28
Publication Date
2024-11-12
Estimated Expiration
2043-02-28

AI Technical Summary

Technical Problem

The existing prenatal diagnosis methods for aortic constriction have a high false positive and false negative rates, resulting in misdiagnosis and misdiagnosis, affecting prenatal consultation and postnatal management, and overdiagnosis increases medical burden.

Method used

A prenatal diagnosis system for aortic constriction is adopted. By obtaining the measured values ​​of fetal weeks and multiple ultrasound indicators, aortic constriction prenatal diagnosis model is used for analysis, comprehensively evaluate the status of the aorta, output diagnostic results, and reduce The rate of misdiagnosis and missed diagnosis improves diagnosis accuracy.

Benefits of technology

Simplify the diagnostic process, improve diagnostic accuracy, reduce the rate of misdiagnosis and missed diagnosis, and provide more accurate diagnostic results through the comprehensive analysis of multiple ultrasound index data, and reduce unnecessary medical burden.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116189905B_ABST
    Figure CN116189905B_ABST
Patent Text Reader

Abstract

The present application relates to a method for establishing a prenatal diagnosis model for aortic coarctation and a diagnosis system, belonging to the technical field of diagnosis systems, wherein the diagnosis system comprises: a fetal week acquisition module for acquiring the fetal week; an ultrasound index data acquisition module for acquiring the measured values ​​of multiple ultrasound indexes of the fetal aorta; and a prenatal diagnosis model for aortic coarctation, for analyzing the measured values ​​of the fetal week and the multiple ultrasound indexes to generate a diagnosis result. The present application has the effect of reducing the incidence of misdiagnosis and missed diagnosis and improving the accuracy of prenatal diagnosis of aortic coarctation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of diagnostic systems, and in particular to a method for establishing a prenatal diagnostic model for aortic coarctation and a diagnostic system. Background Art

[0002] Coarctation of the aorta is the most common congenital disease in the fetus that is missed or misdiagnosed. The high number of false positive and false negative cases seriously affects prenatal consultation and postnatal management. If severe coarctation of the aorta is not discovered in time before or even after birth, not only will the child have a very high risk of death, but it will also place a heavy burden on the medical insurance system. Vice versa, if the fetus is overdiagnosed, it will also cause family anxiety and increase unnecessary medical expenses, which will also increase the burden on the family and social medical system.

[0003] Existing diagnostic clues include small aortic diameter, aortic diameter smaller than superior vena cava diameter, imbalance in the aortic / main pulmonary artery diameter ratio, asymmetry of the left and right ventricles (right heart dominance), and a relatively small left heart system. However, the fetal heart is relatively small, the anatomical structure is complex, and the hemodynamics are special. The diagnostic specificity or sensitivity of the above parameters is not high, which can easily lead to misdiagnosis or missed diagnosis, resulting in low accuracy in prenatal diagnosis of aortic coarctation. Summary of the invention

[0004] In order to improve the accuracy of prenatal diagnosis of aortic coarctation, the present application provides a method for establishing a prenatal diagnosis model for aortic coarctation and a diagnosis system.

[0005] In a first aspect, the present application provides a prenatal diagnosis system for aortic coarctation, which adopts the following technical solution:

[0006] A prenatal diagnosis system for aortic coarctation, comprising:

[0007] A fetal week acquisition module is used to obtain the fetal week;

[0008] An ultrasound index data acquisition module is used to obtain actual measured values ​​of multiple ultrasound indexes of the fetal aortic arch;

[0009] The prenatal diagnosis model for aortic coarctation is used to generate a diagnosis result by analyzing the fetal week and the measured values ​​of the multiple ultrasound indicators.

[0010] By adopting the above technical solution, the system obtains the fetal week and the actual measured values ​​of multiple ultrasound indicators, and inputs the fetal week and the actual measured values ​​of the ultrasound indicators into the prenatal diagnosis model of aortic coarctation. The prenatal diagnosis model of aortic coarctation outputs the diagnosis result after analysis. Therefore, it can simplify the diagnosis process, automatically give the diagnosis result, reduce the misdiagnosis and missed diagnosis rate, and improve the accuracy of diagnosis.

[0011] Specifically, the ultrasound index data include: the angle between the transverse arch and the isthmus, the inner diameter of the transverse arch, the Z value of the isthmus and additional index data, and the additional index data at least include: the Z value of the transverse arch, the inner diameter of the isthmus, the distance between the left subclavian artery and the left common carotid artery, the ratio of the distance between the branch arteries and the ratio of the inner diameter of the isthmus to the left subclavian artery.

[0012] By adopting the above technical solution, a single diagnostic indicator cannot accurately evaluate the true state of the aortic arch. The system uses multiple ultrasound indicator data to comprehensively evaluate whether the aorta is normal, thereby improving diagnostic accuracy.

[0013] Furthermore, the aortic coarctation prenatal diagnosis model comprises:

[0014] A diagnostic critical value acquisition submodule, used to acquire the diagnostic critical values ​​of various ultrasound indicators corresponding to the fetal week;

[0015] The first judgment submodule is used to compare the measured value of the angle between the transverse arch and the isthmus with the corresponding diagnostic critical value to determine whether the angle between the transverse arch and the isthmus is normal;

[0016] A second judgment submodule is used for comparing the measured value of the inner diameter of the transverse arch and the measured value of the isthmus Z value with the corresponding diagnostic critical values ​​one by one to judge whether the inner diameter of the transverse arch and the isthmus Z value are normal when the first judgment module judges that the angle between the transverse arch and the isthmus is normal;

[0017] A first output submodule, configured to output a diagnosis result of a normal arch when the second judgment submodule determines that the inner diameter of the transverse arch portion and the Z value of the isthmus portion are both normal;

[0018] A third judgment submodule is used for comparing the measured value of the inner diameter of the transverse arch and the measured value of the isthmus Z value with the corresponding diagnostic critical values ​​one by one when the first judgment module judges that the angle between the transverse arch and the isthmus is abnormal, to judge whether the inner diameter of the transverse arch and the isthmus Z value are normal;

[0019] A second output submodule, configured to output a diagnosis result of aortic coarctation when the third judgment module determines that both the inner diameter of the transverse arch and the Z value of the isthmus are abnormal;

[0020] A detailed analysis submodule, for obtaining the measured value of the additional index for analysis to obtain a diagnosis result when the second judgment module determines that there are abnormal items in the inner diameter of the transverse arch and the Z value of the isthmus;

[0021] A detailed analysis submodule, used for obtaining the measured value of the additional index for analysis to obtain a diagnosis result when the third judgment module determines that there is a normal item in the inner diameter of the transverse arch part and the Z value of the isthmus;

[0022] The detailed analysis submodule includes:

[0023] The fourth judgment submodule is used to compare the measured values ​​of the additional indicators with the corresponding diagnostic critical values ​​one by one to determine whether the additional indicators are abnormal;

[0024] The score calculation submodule is used to calculate the score according to the weight ratio corresponding to each normal additional indicator;

[0025] The third output submodule is used to perform any of the following:

[0026] When the score reaches a preset score, outputting a diagnosis result of aortic coarctation;

[0027] When the score does not reach the preset score, the diagnosis result of normal bow is output.

[0028] By adopting the above technical scheme, the diagnostic critical value acquisition submodule obtains the diagnostic critical value of each ultrasound index according to the fetal week, the first judgment submodule judges whether the angle between the transverse arch and the isthmus is normal, and when it is normal, the second judgment submodule judges whether the inner diameter of the transverse arch and the Z value of the isthmus are normal, the first output submodule outputs the diagnosis result of normal arch when the above three ultrasound index data are normal, the second output submodule outputs the diagnosis result of aortic coarctation when the above three ultrasound index data are abnormal, and when there is a diagnostic inconsistency between the inner diameter of the transverse arch and the Z value of the isthmus and the angle between the transverse arch and the isthmus, the detailed analysis submodule performs detailed analysis, so that multiple ultrasound index data can be integrated to obtain a more accurate diagnosis result.

[0029] Furthermore, the ultrasound index data acquisition module includes:

[0030] An image information acquisition submodule, used for acquiring image information with aortic arch sections;

[0031] A first line generation submodule, for generating a first line from the opening of the left common carotid artery to the opening of the left subclavian artery in response to first operation information of the user;

[0032] A second line generation submodule, for generating a second line parallel to the isthmus lumen from the proximal end to the distal end of the isthmus in response to second operation information of the user;

[0033] The angle measurement module is used to measure the angle between the first line and the second line to obtain the angle between the transverse arch part and the isthmus part.

[0034] By adopting the above technical solution, when the ultrasound index data acquisition module obtains the angle between the transverse arch and the isthmus, the image information acquisition submodule obtains image information with the aortic section, the first line generation submodule generates a first line in response to the user's first operation information, and the second line generation submodule generates a second line in response to the user's second operation information. Then, the angle measurement module can measure the angle between the transverse arch and the isthmus, so that the angle can be automatically measured according to the user's operation, thereby simplifying the diagnosis process and improving work efficiency.

[0035] In another possible implementation, the system further includes:

[0036] A storage module, used to store the measured values ​​and diagnosis results of ultrasound indicators of historical cases;

[0037] An aorta actual information storage module is used to store the actual aorta information of the fetus after birth;

[0038] A historical information screening module is used to compare the diagnosis results of historical cases with the actual information of the aorta after the fetus is born, and screen out historical cases with false positive diagnosis results;

[0039] A standby database generation module, used to generate a standby database using the measured values ​​of various ultrasound indicators of historical cases with false positive diagnosis results and the corresponding fetal weeks;

[0040] A review module is used to compare the actual measured values ​​of the abnormal ultrasound indicators and / or the actual measured values ​​of the additional indicators with the corresponding actual measured values ​​of the ultrasound indicators and / or the actual measured values ​​of the additional indicators in the backup database when the analysis module determines that there is aortic coarctation, and when there are actual measured values ​​of the ultrasound indicators and / or the actual measured values ​​of the additional indicators in the backup database that are consistent with the comparison, generate a prompt message of possible false positive.

[0041] By adopting the above technical solution, the storage module saves the relevant data of historical cases, the aorta actual information storage module saves the actual information of the aorta after the fetus is born, the historical information screening module screens false positive historical cases, and the backup database generation module generates a backup database based on the relevant data of false positive historical cases, so that the backup database is used as a reference. When the review module diagnoses aortic coarctation, it can once again compare and check the data in the backup database to give a more accurate diagnostic reference opinion.

[0042] The system further comprises a guidance module; the guidance module comprises:

[0043] The guidance submodule is used to guide the user to check the corresponding examination items of the fetus in sequence according to the preset examination order;

[0044] The storage submodule is used to store the ultrasound index data and additional index data measured by the ultrasound device when receiving the user's inspection completion operation.

[0045] By adopting the above technical solution, during the ultrasound examination of the fetus, the guidance submodule guides the user to perform the examination items according to the preset examination order, and after the user completes the examination, the ultrasound index data and additional index data measured by the ultrasound equipment are stored to achieve automatic storage, thereby improving the convenience of user diagnosis.

[0046] The system also includes a report generation module; the report generation module includes:

[0047] The difference calculation submodule is used to calculate the difference between the measured value of each abnormal ultrasound index and the diagnostic critical value of the corresponding ultrasound index;

[0048] An abnormal proportion calculation submodule, used to calculate the abnormal proportion of each abnormal ultrasound index based on the difference and the diagnostic critical value;

[0049] A key indicator selection submodule is used to select a preset number of ultrasound indicators with high abnormal proportions as key indicators;

[0050] A recommendation submodule, used for recommending the ultrasound image information corresponding to the acquisition of the key indicators to the user for selection;

[0051] The report generation submodule is used to generate a report based on the ultrasound image and the diagnosis result selected by the user.

[0052] By adopting the above technical scheme, the difference calculation submodule calculates the difference between the actual measured value of each abnormal ultrasound index and the diagnostic critical value, the abnormal proportion calculation submodule can calculate the abnormal proportion based on the difference and the diagnostic critical value, the key index selection submodule selects the key index according to the level of the abnormal proportion, the recommendation submodule recommends the ultrasound image corresponding to the key index to the user, and then the report generation submodule can efficiently generate a report based on the ultrasound image and diagnostic results selected by the user.

[0053] In another possible implementation, the system further includes:

[0054] An original ultrasonic image information acquisition module is used to acquire original ultrasonic image information;

[0055] an aortic arch recognition module, used for inputting the original ultrasound image information into a trained aortic arch recognition model to recognize the aortic arch;

[0056] a fourth determination module, for determining whether the original ultrasound image information includes anatomical landmarks necessary for diagnosing the aorta, wherein the anatomical landmarks necessary for diagnosing the aorta include the transverse arch, isthmus, brachiocephalic artery, left common carotid artery, left subclavian artery and ductus arteriosus;

[0057] an ultrasound image information generating module, configured to, when the fourth determining module determines that the result is yes, crop, rotate or scale the original ultrasound image information to generate ultrasound image information of a preset size, and mark the anatomical landmarks necessary for diagnosing the aorta included in the ultrasound image information;

[0058] The prompt information generating module is used to generate prompt information to prompt the user to re-acquire the original ultrasound image information when the fourth judging module judges that the result is no.

[0059] By adopting the above technical solution, the original ultrasonic image information acquisition module acquires the original ultrasonic image information, the aortic arch recognition module inputs the original ultrasonic image information into the trained aortic arch recognition model, thereby identifying the aortic arch, and the fourth judgment module determines that the original ultrasonic image information includes the anatomical landmarks necessary for diagnosing aortic coarctation, then the ultrasonic image information generation module processes the original ultrasonic image information to obtain ultrasonic image information of a preset size, and when the fourth judgment module determines that the original ultrasonic image information does not include the anatomical landmarks required for diagnosing aortic coarctation, the prompt information generation module generates prompt information to prompt the user to re-acquire the original ultrasonic image information, and then automatically process to obtain standardized ultrasonic image information, obtain the anatomical landmarks necessary for diagnosing aortic coarctation, accurately locate the lesion site, ensure data accuracy, and facilitate efficient diagnosis.

[0060] Furthermore, the aortic arch identification module includes:

[0061] A contour recognition submodule, used to recognize the black and white parts in the original ultrasonic image information, mark the intersection of black and white, and obtain contour mark image information;

[0062] A side length determination module, used to determine the area of ​​a closed figure surrounded by the marker in the contour marker image information, calculate the ratio of the area of ​​the closed figure to the area of ​​the original ultrasound image information, and determine the side length of the sub-image information obtained after segmentation according to the ratio;

[0063] An image information segmentation submodule, used to segment the contour mark image information according to the side length to obtain a plurality of non-overlapping sub-image information;

[0064] A fifth determination submodule, used to determine whether the sub-image information includes a logo;

[0065] a first image information determination submodule, configured to determine the sub-image information as first image information after the fifth determination submodule determines that the sub-image information includes an identifier;

[0066] A sixth judgment submodule is used to determine whether there is any adjacent first image information of the current first image information, taking the first image information located in the center of the contour mark image information as a starting point;

[0067] The second image information determination submodule determines the current first image information as the second image information after the sixth determination submodule determines that it is yes, and determines the adjacent first image information as the current image information, and the sixth determination submodule is executed again; after the sixth determination submodule determines that it is no, the current first image information is determined as the second image information, and the sixth determination submodule stops executing;

[0068] An image information combination submodule, used to determine the ultrasonic sub-image information corresponding to the second image information, and obtain regular core ultrasonic image information according to a combination of a plurality of corresponding ultrasonic sub-image information;

[0069] The recognition module is used to input the core ultrasound image information into a trained aortic arch recognition model to recognize the aortic arch.

[0070] By adopting the above technical solution, the contour recognition submodule identifies the position where the dark and white colors intersect in the original ultrasonic image information to obtain the contour identification image information. The side length determination module obtains the side length of the sub-image information according to the ratio of the image area enclosed by the identification in the contour identification image information to the area of ​​the original ultrasonic image. Then, the image information segmentation submodule can segment the contour identification information according to the side length to obtain the sub-image information. Then, the fifth judgment submodule distinguishes the sub-image information with the identification, and the first image information determination submodule determines the sub-image information with the identification as the first image information; the sixth judgment submodule and the second image information determination submodule determine the multiple first image information clustered together as the second image information, and then the image information combination submodule combines the ultrasonic sub-image information corresponding to the second image information to obtain the core ultrasonic image information. The core ultrasonic image information includes the aortic arch and contains fewer anatomical landmarks that are useless for diagnosis, which enables the recognition module to quickly identify and obtain the result.

[0071] In a second aspect, the present application provides a method for establishing a prenatal diagnosis model of aortic coarctation applied to the system described in the first aspect, using the following technical solution:

[0072] Obtain diagnostic accuracy information for multiple candidate ultrasound index data;

[0073] Determining a plurality of ultrasound indices for establishing a model from the plurality of candidate ultrasound indices according to the diagnostic accuracy information of the candidate ultrasound indices;

[0074] Based on the diagnostic critical values ​​and diagnostic procedures corresponding to the ultrasound indicators at different fetal weeks, a prenatal diagnosis model for aortic coarctation is established;

[0075] By adopting the above technical solution, the electronic device screens multiple candidate ultrasound indicator data according to their diagnostic accuracy, selects multiple ultrasound indicators with high accuracy, and uses the selected ultrasound indicator data to establish a model, thereby obtaining a diagnostic model with high diagnostic accuracy. In addition, the electronic device can automatically perform multi-layer deliberation and judgment on abnormal conditions of multiple ultrasound indicator data based on the input ultrasound indicator data and fetal week data, thereby improving diagnostic accuracy.

[0076] In summary, the present application includes at least one of the following beneficial technical effects:

[0077] 1. The electronic device screens the diagnostic accuracy of multiple candidate ultrasound indicators, selects ultrasound indicators with high accuracy, and then establishes a model to obtain a diagnostic model with high diagnostic accuracy. In addition, the electronic device can automatically perform multi-layer deliberation and judgment on the abnormal conditions of multiple ultrasound indicators based on the input measured values ​​of ultrasound indicators and fetal weeks and other data, thereby improving the diagnostic accuracy.

[0078] 2. The system outputs the diagnosis results after analysis, thus simplifying the diagnosis process, automatically giving the diagnosis results, reducing the missed diagnosis and misdiagnosis rates, and improving the accuracy of diagnosis;

[0079] 3. The system can integrate multiple ultrasound index data to obtain more accurate diagnosis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 It is a flow chart of the method for establishing a prenatal diagnosis model of aortic coarctation in an embodiment of the present application.

[0081] Figure 2 It is a schematic diagram of establishing a ROC curve based on ultrasound indicators in an embodiment of the present application.

[0082] Figure 3 It is a structural block diagram of the prenatal diagnosis model of aortic coarctation in the embodiment of the present application.

[0083] Figure 4 It is a schematic diagram of the anatomy of the aortic arch in an embodiment of the present application.

[0084] Figure 5 It is a schematic diagram of determining the angle between the transverse arch part and the isthmus part according to ultrasound image information in an embodiment of the present application.

[0085] Figure 6 It is a structural block diagram of the prenatal diagnosis system of aortic coarctation in an embodiment of the present application.

[0086] Figure 7 7-1 is the contour identification image information, and 7-2 is the core ultrasound image information.

[0087] Figure numerals: 1, ascending aorta; 2, transverse arch; 3, isthmus; 4, descending aorta; 5, brachiocephalic artery; 6, left common carotid artery; 7, left subclavian artery; 8, ductus arteriosus; 9, pulmonary artery. Implementation

[0088] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0089] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0090] The embodiment of the present application discloses a method for establishing a prenatal diagnosis model for aortic coarctation, which is performed by an electronic device. The electronic device can be a server or a terminal device, wherein the server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a desktop computer, etc., but is not limited thereto.

[0091] The method for establishing a prenatal diagnosis model for aortic coarctation is based on Figure 1 , including the following process (step S101 to step S103):

[0092] Step S101: obtaining diagnostic accuracy information of a plurality of candidate ultrasound indicators.

[0093] Specifically, the electronic device lists all existing indicators for diagnosing aortic coarctation as alternative ultrasound indicators, and compares the accuracy of diagnosis by establishing ROC curves (receiver operating characteristic curves) for various alternative ultrasound indicators.

[0094] Step S102: determining a plurality of ultrasound indices for establishing a model from a plurality of candidate ultrasound indices according to the diagnostic accuracy information of the candidate ultrasound indices.

[0095] Specifically, in the ROC curve, the closer to the upper left corner, the higher the accuracy of the corresponding test. Therefore, the electronic device can place the ROC curves corresponding to multiple candidate ultrasound index data in the same coordinate system for intuitive comparison.

[0096] For example, refer to Figure 2 The electronic device uses the angle between the transverse arch and the isthmus, the inner diameter of the transverse arch and the Z value of the isthmus with higher accuracy as ultrasonic indicators for establishing the model.

[0097] Step S103: establishing a prenatal diagnosis model for aortic coarctation based on the diagnostic critical values ​​and diagnostic procedures of ultrasound indicators corresponding to different fetal weeks.

[0098] Specifically, the diagnostic process is to make judgments step by step based on the ultrasound index data. The electronic equipment establishes a prenatal diagnosis model for aortic coarctation, inputs the actual measured values ​​of the ultrasound index, and the model automatically analyzes to obtain the diagnosis result, and then outputs the diagnosis result.

[0099] Aortic coarctation prenatal diagnosis model 200, reference Figure 3 ,include:

[0100] The diagnostic critical value acquisition submodule 201 is used to acquire the diagnostic critical values ​​of various ultrasound indicators corresponding to the fetal week.

[0101] Specifically, affected by fetal growth, the reference values ​​for diagnosing fetuses of different weeks are different. The diagnostic critical value in the diagnostic model is the diagnostic critical value for judging whether each ultrasound index data is abnormal. Therefore, the electronic device sets multiple groups of diagnostic critical values ​​of ultrasound indexes for different fetal weeks and saves them. The diagnostic critical value acquisition submodule 201 can acquire a corresponding group of diagnostic critical values ​​according to the fetal week.

[0102] The first judgment submodule 202 is used to compare the measured value of the angle between the transverse arch and the isthmus with the diagnostic critical value of the corresponding ultrasound index to determine whether the angle between the transverse arch and the isthmus is normal.

[0103] Reference Figure 4 The figure shows the anatomical diagram of the aortic arch segments. Aortic isthmus coarctation and transverse arch hypoplasia are the most intuitive signs for prenatal diagnosis of aortic coarctation. According to research, when aortic coarctation exists, the angle between the transverse arch and the isthmus is smaller, and the aortic arch is more curved.

[0104] In order to obtain the angle between the transverse arch and the isthmus, firstly, the first line is connected from the opening of the left common carotid artery 6 to the opening of the left subclavian artery 7 to obtain the first line, and then the second line is obtained from the proximal end to the distal end of the isthmus 3 parallel to the lumen of the isthmus 3. The angle between the two lines is the angle between the transverse arch and the isthmus.

[0105] And it was observed from the ROC curve of the angle between the transverse arch and the isthmus that the prediction accuracy based on the angle between the transverse arch and the isthmus was the highest. Therefore, the angle between the transverse arch and the isthmus was determined first to improve the diagnostic accuracy.

[0106] Furthermore, the model compares the measured value of the angle between the transverse arch and the isthmus with the corresponding diagnostic critical value. For example, the diagnostic critical value of the angle between the transverse arch and the isthmus is 117°. When the measured value of the angle between the transverse arch and the isthmus is less than 117°, it is judged to be abnormal; if the measured value of the angle between the transverse arch and the isthmus is greater than or equal to 117°, it is judged to be normal. Figure 5 In Figure A, the measured value of the angle between the transverse arch and the isthmus is 120.5°, which means that the angle between the transverse arch and the isthmus is normal.

[0107] The second judgment submodule 203 is used to compare the measured value of the inner diameter of the transverse arch and the measured value of the isthmus Z value with the corresponding diagnostic critical values ​​one by one when the first judgment module 202 judges that the angle between the transverse arch and the isthmus is normal, so as to judge whether the inner diameter of the transverse arch and the isthmus Z value are normal.

[0108] Specifically, according to the ROC curve, it was observed that the two indicators of the inner diameter of the transverse arch and the Z value of the isthmus were more accurate. Therefore, after the preliminary judgment of the angle between the transverse arch and the isthmus, the two indicators of the inner diameter of the transverse arch and the Z value of the isthmus were further judged.

[0109] The aortic arch can be divided into the ascending aorta 1, the transverse arch 2, the isthmus 3 and the descending aorta 4. Aortic stenosis generally occurs from the isthmus 3 to the opening of the ductus arteriosus 8. When the inner diameter of the fetal transverse arch is smaller than the corresponding diagnostic threshold, it can be judged as abnormal.

[0110] The first output submodule 204 is used to output the diagnosis result of the normal arch when the second judgment submodule 203 judges that the inner diameter of the transverse arch and the Z value of the isthmus are both normal.

[0111] Specifically, when the inner diameter of the transverse arch and the Z value of the isthmus are normal, the angle between the transverse arch and the isthmus is also normal. If multiple ultrasound index data with high detection accuracy are judged to be normal, it can be determined that the aortic arch is normal.

[0112] The third judgment submodule 205 is used to compare the measured value of the inner diameter of the transverse arch and the measured value of the Z value of the isthmus with the corresponding diagnostic critical values ​​one by one when the first judgment module 202 judges that the angle between the transverse arch and the isthmus is abnormal, so as to judge whether the inner diameter of the transverse arch and the Z value of the isthmus are normal.

[0113] The second output submodule 206 is used to output the diagnosis result of aortic coarctation when the third judgment submodule 205 judges that both the inner diameter of the transverse arch and the Z value of the isthmus are abnormal.

[0114] Furthermore, the model 200 also includes:

[0115] The detailed analysis submodule 207 is used to obtain the measured values ​​of the additional indicators for analysis to obtain the diagnosis result when the second judgment submodule 203 judges that there are abnormal items in the inner diameter of the transverse arch and the Z value of the isthmus.

[0116] Specifically, in order to give a more cautious diagnosis, when there are abnormal items in the inner diameter of the transverse arch and the Z value of the isthmus, further scrutiny is required to obtain the actual measured values ​​of additional indicators for analysis, and a comprehensive judgment should be made based on multiple data to reduce the misdiagnosis rate.

[0117] Additional indicators include at least the transverse arch Z value, the isthmus internal diameter, and the distance between the left subclavian artery and the left common carotid artery (i.e. Figure 4 B in the figure), the ratio of the distance between branch arteries ( Figure 4 (B / C) and the ratio of the inner diameter of the isthmus to the left subclavian artery.

[0118] The additional indicators are all existing data commonly used in diagnosing aortic coarctation. The diagnostic model allows users to edit and add additional indicators. The more abnormal items in the additional indicators, the higher the possibility of aortic coarctation. In order to further analyze, the detailed analysis submodule 207 includes:

[0119] The fourth judgment submodule 2071 is used to compare the measured values ​​of the additional indicators with the corresponding diagnostic critical values ​​one by one to determine whether the additional indicators are abnormal.

[0120] Specifically, each additional indicator has a corresponding diagnostic critical value and a corresponding rule for judging it as abnormal.

[0121] For example:

[0122] A transverse arch Z value <-4 is abnormal;

[0123] The inner diameter of the isthmus is <-1.5mm, which is considered abnormal.

[0124] The score calculation submodule 2072 is used to calculate the score according to the weight ratio corresponding to each normal additional indicator.

[0125] Specifically, based on the analysis of the ROC curve, it can be obtained that the accuracy of each additional indicator is different. In order to improve the accuracy of the final diagnosis, the diagnostic model can preset the weight ratio corresponding to each additional indicator. The higher the accuracy of the additional indicator, the higher the corresponding weight ratio. The sum of the weight ratios of each additional indicator is 1.

[0126] Furthermore, the diagnostic model is preset with preset values, which are used to calculate the score. After determining the abnormal items in the additional indicators, the score calculation submodule multiplies the preset values ​​with the weight ratios corresponding to the abnormal items, calculates the intermediate values, and then adds the multiple intermediate values ​​to calculate the score.

[0127] The third output submodule 2073 is configured to perform any of the following:

[0128] When the score reaches the preset score, the diagnosis result of aortic coarctation is output;

[0129] When the score does not reach the preset score, the diagnosis result of normal bow is output.

[0130] Specifically, the preset score is set according to actual conditions. When the score reaches the preset score, the more abnormal items there are in the additional indicators, the higher the possibility of aortic coarctation.

[0131] Furthermore, the detailed analysis submodule 207 is also used to obtain the measured value of the additional index for analysis to obtain the diagnosis result when the third judgment submodule 206 judges that there are normal items in the inner diameter of the transverse arch part and the Z value of the isthmus.

[0132] Specifically, after the angle between the transverse arch and the isthmus is abnormal, the third judgment submodule 205 and the second output submodule 206 further judge whether the inner diameter of the transverse arch and the Z value of the isthmus are abnormal. If both are abnormal, the possibility of aortic coarctation is high, and the diagnosis result of aortic coarctation is output; if there are normal items in the inner diameter of the transverse arch and the Z value of the isthmus when the angle between the transverse arch and the isthmus is abnormal, further refinement is performed, and the detailed analysis submodule 207 performs analysis based on the additional indicator data, and the analysis method is the same as above.

[0133] Therefore, the electronic device can establish a prenatal diagnosis model 200 for aortic coarctation according to the method for establishing a prenatal diagnosis model for aortic coarctation, so as to improve the accuracy of diagnosing aortic coarctation.

[0134] The embodiment of the present application also provides a prenatal diagnosis system for aortic coarctation.

[0135] Reference Figure 6 The aortic coarctation prenatal diagnosis system 300 comprises:

[0136] The fetal week acquisition module 301 is used to acquire the fetal week.

[0137] The ultrasound index data acquisition module 302 is used to acquire the actual measured values ​​of multiple ultrasound indexes of the fetal aortic arch.

[0138] Specifically, the system is connected to each ultrasound instrument used to detect the fetus, and receives the fetal week and actual measured values ​​of multiple ultrasound indicators sent by each ultrasound instrument.

[0139] Ultrasonic indices include: the angle between the transverse arch and the isthmus, the inner diameter of the transverse arch, the Z value of the isthmus and additional index data. The additional indicators at least include: the Z value of the transverse arch, the inner diameter of the isthmus, the distance between the left subclavian artery and the left common carotid artery, the ratio of the distance between branch arteries and the ratio of the inner diameter of the isthmus to the left subclavian artery.

[0140] The measured value of each ultrasonic index needs to be obtained by the user through operation according to the image. Taking the angle between the transverse arch and the isthmus as an example, the ultrasonic index data acquisition module 302 includes:

[0141] The ultrasound image information acquisition submodule is used to acquire ultrasound image information with aortic arch section.

[0142] Reference Figure 5 , which is the ultrasonic image information obtained by the system through the ultrasonic instrument.

[0143] Specifically, the user operates the ultrasound instrument to move the probe, and obtains clear ultrasound image information with aortic sections on the instrument screen, and the ultrasound image information acquisition submodule stores the ultrasound image information.

[0144] The first line generating submodule is used to generate a first line from the opening of the left common carotid artery to the opening of the left subclavian artery in response to the first operation information of the user.

[0145] Specifically, the first operation information is an action in which the user finds the left common carotid opening and the left subclavian artery opening on the image information and clicks the two positions in sequence with the mouse, and the first line generation submodule automatically connects the two points into a first line.

[0146] The second line generation submodule is used to generate a second line parallel to the isthmus lumen from the proximal end to the distal end of the isthmus in response to the second operation information of the user.

[0147] Specifically, the second operation information is that the user finds the proximal end and the distal end of the isthmus parallel to the isthmus lumen on the image information, and clicks the above two positions with the mouse in sequence, and the second line generation submodule automatically connects the two points into a second line.

[0148] The angle measurement module is used to measure the angle between the first line and the second line to obtain the angle between the transverse arch and the isthmus.

[0149] Specifically, if the first line and the second line do not intersect, the first line and the second line are extended, and the angle measurement module measures the angle between the first line and the second line.

[0150] The prenatal diagnosis model 200 for aortic coarctation is used to generate a diagnosis result by analyzing the fetal week and the measured values ​​of multiple ultrasound indicators.

[0151] During the diagnosis process, the aortic coarctation prenatal diagnosis model 200 can automatically obtain the actual measured values ​​of the fetal week and ultrasound indicators obtained during instrument examination without manual operation input.

[0152] In another possible implementation, the aortic coarctation prenatal diagnosis system 300 further includes:

[0153] The storage module is used to save the ultrasound index data and diagnosis results of historical cases.

[0154] Specifically, the system saves historical cases and generates a database.

[0155] The actual aorta information storage module is used to obtain the actual aorta information of the fetus after birth.

[0156] The historical information screening module is used to compare the diagnosis results of historical cases with the actual information of the aorta after the fetus is born, and to screen out historical cases with false positive diagnosis results.

[0157] A standby database generation module is used to generate a standby database using the measured values ​​of various ultrasound indicators of historical cases with false positive diagnosis results and the corresponding fetal weeks;

[0158] Specifically, the system will re-examine fetuses diagnosed with aortic coarctation before birth. The user inputs the fetal case, and the system can obtain the actual information of the fetal aorta through the case, compare it with the diagnosis results of historical cases, and determine whether it is a false positive. False positive cases are of great reference significance for the subsequent diagnosis of new fetuses, so false positive historical cases are screened out to generate a backup database.

[0159] The review module is used to compare the actual measured values ​​of the abnormal ultrasound indicators and / or the actual measured values ​​of the additional indicators with the corresponding actual measured values ​​of the ultrasound indicators and / or the actual measured values ​​of the additional indicators in the spare database when the aortic coarctation prenatal diagnosis model 200 determines that there is aortic coarctation. When there are actual measured values ​​of the ultrasound indicators and / or the actual measured values ​​of the additional indicators in the spare database that are consistent with the comparison, a prompt message of possible false positive is generated.

[0160] Specifically, the backup database can be used as a reference. When aortic coarctation is diagnosed, a re-examination is conducted and the abnormal items are compared with the indicators in the backup database. When data with consistent indicators exist, historical cases can be used as a reference to provide a reference for user diagnosis, thereby reducing the possibility of misdiagnosis.

[0161] In another possible implementation, the system 300 further includes a guidance module, which includes:

[0162] The guidance submodule is used to guide the user to check the corresponding positions of the fetal examination items in sequence according to the preset examination order.

[0163] For example, the system guides the user to first obtain the angle between the transverse arch and the isthmus of the fetus, and then guides the user to obtain the inner diameter of the transverse arch and the Z value of the isthmus in turn until the diagnosis is completed, and guides the user to find the corresponding heart section, which is suitable for novice use.

[0164] The storage submodule is used to store the ultrasound index data and additional index data measured by the ultrasound device when receiving the user's inspection completion operation.

[0165] Specifically, the user check completion operation may be the last click operation when acquiring data, or may be a preset shortcut key.

[0166] When the user views the image information through the ultrasound device, after the corresponding data is obtained through operation, the ultrasound device can automatically store the data and upload it to the system. For example, the user clicks on a point on the image information to obtain the first line and the second line. When the user clicks the last point, the user operation is completed. At this time, the ultrasound device measures the angle between the transverse arch and the isthmus, and the system stores the data measured by the ultrasound device.

[0167] In another possible implementation, the system 300 further includes a report generation module, and the report generation module includes:

[0168] The difference calculation submodule is used to calculate the difference between the actual measured value of each abnormal ultrasound indicator and the corresponding diagnostic critical value.

[0169] Specifically, when the fetus is being diagnosed, each time the actual measured value of an ultrasound index is obtained, the difference between the actual measured value of the ultrasound index and the diagnostic critical value is automatically calculated and saved.

[0170] The abnormal proportion calculation submodule is used to calculate the abnormal proportion of each abnormal ultrasound index based on the difference and the diagnostic critical value.

[0171] Specifically, the system divides the difference value by the diagnostic threshold value to calculate the abnormality ratio. The size of the difference value can reflect the degree of abnormality of the current test item. The larger the difference value, the larger the abnormality ratio, and the higher the degree of abnormality.

[0172] The key indicator selection submodule is used to select a preset number of ultrasound indicators with high abnormal proportions as key indicators.

[0173] The recommendation submodule is used to recommend the corresponding ultrasound image information when obtaining key indicators to the user for selection.

[0174] Specifically, in order to select the most representative abnormal items in the diagnosis results, the system selects key indicators and recommends ultrasound image information to the user, and the user selects image information from it for generating a report based on experience. Furthermore, the system can recommend key indicators in order of high to low abnormality ratio.

[0175] The report generation submodule is used to generate a report based on the ultrasound image and diagnosis results selected by the user.

[0176] Specifically, the system provides a report template, automatically imports the diagnosis results, edits and modifies the report content according to the user's operation, and then generates a report.

[0177] In another possible implementation, in order to enable the ultrasound image information acquisition submodule to quickly and accurately obtain ultrasound image information containing the aortic arch, the system 300 further includes:

[0178] The original ultrasonic image information acquisition module is used to acquire the original ultrasonic image information.

[0179] Specifically, the original ultrasonic image information is unprocessed ultrasonic image information. After the user operates the ultrasonic instrument, the system can obtain the original ultrasonic image information through the ultrasonic instrument.

[0180] The aortic arch recognition module is used to input the original ultrasound image information into the trained aortic arch recognition model to recognize the aortic arch.

[0181] Specifically, when establishing the aortic arch recognition model, the system acquires a large amount of ultrasound image information as a data set, and each ultrasound image information includes the aortic arch. The aortic arch includes but is not limited to the transverse arch, isthmus, brachiocephalic artery, left common carotid artery, left subclavian artery and ductus arteriosus.

[0182] Furthermore, the system divides the data set into a training set, a validation set, and a test set, and inputs the training set into the aortic arch recognition model for training. The training steps are repeated until the classification error no longer changes or is in a preset range, and the peak signal-to-noise ratio and structural similarity of the aortic arch recognition model tested using the validation set and the test set reach the optimal level, and the trained aortic arch recognition model is output.

[0183] The fourth judgment module is used to judge whether the original ultrasound image information includes the anatomical landmarks necessary for diagnosing aortic coarctation. The anatomical landmarks necessary for diagnosing aorta include the transverse arch, isthmus, brachiocephalic artery, left common carotid artery, left subclavian artery and ductus arteriosus.

[0184] Specifically, the anatomical landmarks are the locations of the aortic arch, such as Figure 4The ascending aorta, transverse arch, isthmus, descending aorta, brachiocephalic artery, left common carotid artery, left subclavian artery, ductus arteriosus and pulmonary artery are all anatomical landmarks. Among them, the transverse arch, isthmus, brachiocephalic artery, left common carotid artery, left subclavian artery and ductus arteriosus are necessary anatomical landmarks for diagnosing aortic coarctation, and the system can obtain ultrasound indicators through these anatomical landmarks.

[0185] Specifically, after the system inputs the original ultrasound image information into the trained aortic arch recognition model, the aortic arch recognition model recognizes the aortic arch in the original ultrasound image information and marks the anatomical landmarks recognized by the aortic arch recognition model at the corresponding positions of the original ultrasound image information.

[0186] The ultrasound image information generation module is used to crop, rotate or scale the original ultrasound image information to generate ultrasound image information of a preset size when the fourth judgment module determines that the answer is yes, and mark the anatomical landmarks required for diagnosing the aorta included in the ultrasound image information.

[0187] Specifically, when the area of ​​the aortic arch region occupies a small area of ​​the original ultrasonic image information, most of the original ultrasonic image information will be irrelevant to the diagnosis of aortic coarctation. In this case, the area with the aortic arch is retained and the original ultrasonic image information is cut into a preset shape. Generally, the preset shape can be a fan shape.

[0188] Then, the cut ultrasound image information is scaled to a preset size to obtain ultrasound image information for diagnosis. The ultrasound image information includes the anatomical landmarks necessary for diagnosing aortic coarctation, accurately locates the lesion site, ensures data accuracy, and facilitates efficient diagnosis after marking.

[0189] Then, the ultrasound image information acquisition submodule can acquire ultrasound image information.

[0190] The prompt information generating module is used to generate prompt information to prompt the user to re-acquire the original ultrasound image information when the fourth judging module judges that the result is no.

[0191] Specifically, when the fourth judgment module judges as no, that is, the anatomical landmarks for diagnosing aortic coarctation in the current original ultrasound image information are incomplete and therefore cannot be used to diagnose aortic coarctation, the system generates a prompt message to prompt the user to re-acquire the original ultrasound image information.

[0192] Furthermore, the system may indicate in the prompt information the missing parts in the current original ultrasound image information.

[0193] Furthermore, as mentioned above, when the area of ​​the aortic arch occupies a small area of ​​the original ultrasound image information, most of the original ultrasound image information will be irrelevant to the diagnosis of aortic coarctation, and when the original ultrasound image information is input into the aortic arch recognition model, the recognition speed of the aortic arch recognition module will be reduced. Therefore, the aortic arch recognition module includes:

[0194] The contour recognition submodule is used to identify the black and white parts in the original ultrasonic image information, mark the intersection of black and white, and obtain contour mark image information.

[0195] Specifically, it can be seen from the black and white ultrasound image that the aortic arch appears in black, and the anatomical landmarks around the aortic arch appear in white. In order to highlight the aortic arch, the system marks the location where black and white intersect.

[0196] The cutting quantity determination submodule is used to determine the area size of the closed figure enclosed by the outline mark image information, calculate the ratio of the area of ​​the closed figure to the area of ​​the original ultrasound image information, and determine the number of sub-image information obtained after segmentation based on the ratio.

[0197] Specifically, the system puts the contour identification image information into a coordinate system, and then obtains the position of each point on the contour, so that the area of ​​the closed image can be automatically estimated.

[0198] Furthermore, the system will segment the original ultrasound image to obtain multiple sub-image information. Through further analysis, the area irrelevant to the diagnosis can be deleted, thereby retaining the sub-image information with the aortic arch.

[0199] Therefore, the system presets the number of sub-image information corresponding to each ratio. When the ratio is larger, the aortic arch occupies a larger area in the original ultrasound image, which also means that there are fewer areas irrelevant to the diagnosis and there is no need to delete too much sub-image information. In order to speed up the calculation, the system sets the corresponding number of sub-image information to a smaller number.

[0200] However, the smaller the ratio, the smaller the area of ​​the aortic arch in the original ultrasound image information, the larger the area irrelevant to the diagnosis, and the more difficult it is to identify the aortic arch. Therefore, the more the system sets the number of corresponding sub-image information, the more refined the analysis can be, and most of the areas irrelevant to the diagnosis can be deleted.

[0201] The image information segmentation submodule is used to segment the contour mark image information according to the quantity to obtain a plurality of non-overlapping sub-image information, and to segment the original ultrasonic image information into ultrasonic sub-image information corresponding to the sub-image information one by one.

[0202] Generally, the contour mark image information is in the shape of a sector.

[0203] Reference Figure 7 The system determines multiple first segmentation points (such as a) at equal intervals on the outer arc of the sector, and multiple second segmentation points (such as b) at equal intervals on the inner arc of the sector, and connects the first segmentation points and the second segmentation points one by one to determine multiple first segmentation lines; the system determines multiple third segmentation points (such as c) at equal intervals on the two straight sides of the sector, draws multiple arcs with the same curvature as the outer arc, and makes the arcs pass through the third segmentation points on the two straight sides to obtain multiple second segmentation lines, so that the area between the first segmentation line and the second segmentation line is the sub-image information. The contour mark image information is divided into a preset number of non-overlapping sub-image information.

[0204] It should be noted that each number corresponds to a segmentation method, that is, the number of sub-image information is different, and the spacing of the first segmentation points determined by the system on the outer arc of the sector and the spacing of the second segmentation points determined on the straight edge are also different.

[0205] The position and size of the ultrasound sub-image information correspond one-to-one to the sub-image information.

[0206] The fifth determination submodule is used to determine whether the sub-image information includes a logo.

[0207] Specifically, the system may enable the fifth judgment submodule to judge each sub-image information in sequence, or may judge each sub-image information at the same time, so as to obtain the judgment result quickly.

[0208] The first image information determination submodule is used to determine the sub-image information as the first image information after the fifth determination submodule determines that the sub-image information includes the identifier.

[0209] The sixth judgment submodule is used to determine whether there is any adjacent first image information to the current first image information, taking the first image information located in the center of the contour mark image information as a starting point.

[0210] Specifically, when the user obtains the original ultrasound image information while operating the ultrasound instrument, due to human subjective factors, the aortic arch is likely to be located in the center of the original ultrasound image information. Therefore, it is faster and more accurate to start judging from the center of the contour mark image information. The final recognition result also guides the user to select the original ultrasound image with the aortic arch placed in the center, so as to observe the aortic arch more comprehensively.

[0211] The second image information determination submodule determines the current first image information as the second image information and determines the adjacent first image information as the current image information after the sixth judgment submodule determines as yes, and the sixth judgment submodule executes again; after the sixth judgment submodule determines as no, the current first image information is determined as the second image information, and the sixth judgment submodule stops executing.

[0212] The image information combination submodule is used to determine the ultrasonic sub-image information corresponding to the second image information, and obtain regular core ultrasonic image information according to a combination of a plurality of corresponding ultrasonic sub-image information.

[0213] Specifically, the system can determine the first image information gathered together as the second image information by judging, referring to Figure 7 , the sub-image information of the N region and the M region are both the first image information, the M region is the second image information, and the original ultrasound image information corresponding to the positions of the multiple second image information is likely to include the aortic arch. When the combined ultrasound sub-image information is irregular, the system selects the ultrasound sub-image information adjacent to the ultrasound sub-image information, fills the irregular position, and then obtains the regular core ultrasound image information. For example Figure 7 In the figure, the O region is the corresponding ultrasonic sub-image information, and the P region is the adjacent ultrasonic sub-image information.

[0214] The recognition module is used to input the core ultrasound image information into a trained aortic arch recognition model to recognize the aortic arch.

[0215] Specifically, through the processing of the above modules, the core ultrasound image information includes the anatomical landmarks necessary for diagnosing aortic coarctation, and the area that is useless for diagnosis is small, which can speed up the analysis speed of the aortic arch recognition model. The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be covered within the protection scope of the present application.

[0216] In addition, it is to be understood that relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprises," "comprising," or any other variation thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device that includes a series of elements includes not only those elements, but also other elements that are not explicitly listed, or elements that are inherent to such process, method, article, or device.

Claims

1. A prenatal diagnosis system for aortic coarctation, characterized in that: include: A fetal week acquisition module is used to obtain the fetal week; An ultrasound index data acquisition module is used to obtain actual measured values ​​of multiple ultrasound indexes of the fetal aortic arch; A prenatal diagnosis model for aortic coarctation, used to analyze the fetal age and the measured values ​​of the multiple ultrasound indicators to generate a diagnosis result; The ultrasound indexes include: the angle between the transverse arch and the isthmus, the inner diameter of the transverse arch, the Z value of the isthmus and additional indexes, and the additional indexes include at least: the Z value of the transverse arch, the inner diameter of the isthmus, the distance between the left subclavian artery and the left common carotid artery, the ratio of the distance between the branch arteries and the ratio of the inner diameter of the isthmus to the left subclavian artery; The aortic coarctation prenatal diagnosis model comprises: A diagnostic critical value acquisition submodule, used to acquire the diagnostic critical values ​​of various ultrasound indicators corresponding to the fetal week; The first judgment submodule is used to compare the measured value of the angle between the transverse arch and the isthmus with the corresponding diagnostic critical value to determine whether the angle between the transverse arch and the isthmus is normal; wherein the diagnostic critical value of the angle between the transverse arch and the isthmus is 117°; A second judgment submodule is used for comparing the measured value of the inner diameter of the transverse arch and the measured value of the isthmus Z value with the corresponding diagnostic critical values ​​one by one to judge whether the inner diameter of the transverse arch and the isthmus Z value are normal when the first judgment submodule judges that the angle between the transverse arch and the isthmus is normal; A first output submodule, configured to output a diagnosis result of a normal arch when the second judgment submodule determines that the inner diameter of the transverse arch portion and the Z value of the isthmus portion are both normal; A third judgment submodule is used for comparing the measured value of the inner diameter of the transverse arch and the measured value of the isthmus Z value with the corresponding diagnostic critical values ​​one by one to judge whether the inner diameter of the transverse arch and the isthmus Z value are normal when the first judgment submodule judges that the angle between the transverse arch and the isthmus is abnormal; A second output submodule, configured to output a diagnosis result of aortic coarctation when the third judgment submodule judges that both the inner diameter of the transverse arch and the Z value of the isthmus are abnormal; A detailed analysis submodule, used for obtaining the measured value of the additional index for analysis to obtain a diagnosis result when the second judgment submodule judges that there are abnormal items in the inner diameter of the transverse arch and the Z value of the isthmus; The detailed analysis submodule is further used to obtain the measured value of the additional index for analysis to obtain a diagnosis result when the third judgment submodule judges that there is a normal item in the inner diameter of the transverse arch part and the Z value of the isthmus; The detailed analysis submodule includes: The fourth judgment submodule is used to compare the measured values ​​of the additional indicators with the corresponding diagnostic critical values ​​one by one to determine whether the additional indicators are abnormal; The score calculation submodule is used to calculate the score according to the weight ratio corresponding to each normal additional indicator; The third output submodule is used to perform any of the following: When the score reaches a preset score, outputting a diagnosis result of aortic coarctation; When the score does not reach the preset score, outputting the diagnosis result of normal arch; The ultrasonic index data acquisition module includes: An ultrasound image information acquisition submodule, used for acquiring ultrasound image information with aortic arch section; A first line generation submodule, for generating a first line from the opening of the left common carotid artery to the opening of the left subclavian artery in response to first operation information of the user; A second line generation submodule, for generating a second line parallel to the isthmus lumen from the proximal end to the distal end of the isthmus in response to second operation information of the user; An angle measurement module, used to measure the angle between the first line and the second line, and obtain the angle between the transverse arch part and the isthmus part; The system further comprises: A storage module, used to store the measured values ​​and diagnosis results of ultrasound indicators of historical cases; An aorta actual information storage module is used to store the actual aorta information of the fetus after birth; A historical information screening module is used to compare the diagnosis results of historical cases with the actual information of the aorta after the fetus is born, and screen out historical cases with false positive diagnosis results; A standby database generation module, used to generate a standby database using the measured values ​​of various ultrasound indicators of historical cases with false positive diagnosis results and the corresponding fetal weeks; a review module, for comparing the actual measured values ​​of the abnormal ultrasound index and / or the actual measured values ​​of the additional index with the actual measured values ​​of the corresponding ultrasound index and / or the actual measured values ​​of the additional index in the standby database when the detailed analysis submodule determines that there is aortic coarctation, and generating a prompt message of possible false positive when there are actual measured values ​​of the ultrasound index and / or the actual measured values ​​of the additional index in the standby database that are consistent with the comparison; The system further comprises: An original ultrasonic image information acquisition module is used to acquire original ultrasonic image information; an aortic arch recognition module, used for inputting the original ultrasound image information into a trained aortic arch recognition model to recognize the aortic arch; a fourth determination module, for determining whether the original ultrasound image information includes anatomical landmarks necessary for diagnosing the aorta, wherein the anatomical landmarks necessary for diagnosing the aorta include the transverse arch, isthmus, brachiocephalic artery, left common carotid artery, left subclavian artery and ductus arteriosus; an ultrasound image information generating module, configured to, when the fourth determining module determines that the result is yes, crop, rotate or scale the original ultrasound image information to generate ultrasound image information of a preset size, and mark the anatomical landmarks necessary for diagnosing the aorta included in the ultrasound image information; a prompt information generating module, configured to generate prompt information prompting the user to reacquire the original ultrasound image information when the fourth judging module judges that the result is no; The aortic arch identification module comprises: A contour recognition submodule, used to recognize the black and white parts in the original ultrasonic image information, mark the intersection of black and white, and obtain contour mark image information; a cutting quantity determination submodule, used to determine the area size of the closed figure surrounded by the mark in the contour mark image information, calculate the ratio of the area of ​​the closed figure to the area of ​​the original ultrasound image information, and determine the number of sub-image information obtained after segmentation according to the ratio; the larger the ratio, the smaller the number of sub-image information corresponding to the ratio is set; the smaller the ratio, the more the number of sub-image information corresponding to the ratio is set; An image information segmentation submodule, configured to segment the contour mark image information according to the quantity to obtain a plurality of non-overlapping sub-image information, and segment the original ultrasound image information into ultrasound sub-image information corresponding to the sub-image information one by one; A fifth determination submodule, used to determine whether the sub-image information includes a logo; a first image information determination submodule, configured to determine the sub-image information as first image information after the fifth determination submodule determines that the sub-image information includes an identifier; A sixth judgment submodule is used to determine whether there is any adjacent first image information of the current first image information, taking the first image information located in the center of the contour mark image information as a starting point; The second image information determination submodule determines the current first image information as the second image information and determines the adjacent first image information as the current image information after the sixth determination submodule determines that it is yes, and the sixth determination submodule is executed again; after the sixth determination submodule determines that it is no, the current first image information is determined as the second image information, and the sixth determination submodule stops executing; An image information combination submodule, used to determine the ultrasonic sub-image information corresponding to the second image information, and obtain regular core ultrasonic image information according to a combination of a plurality of corresponding ultrasonic sub-image information; The recognition module is used to input the core ultrasound image information into a trained aortic arch recognition model to recognize the aortic arch.

2. The system according to claim 1, characterized in that The system also includes a guidance module; The guidance module comprises: The guidance submodule is used to guide the user to check the corresponding examination items of the fetus in sequence according to the preset examination order; The storage submodule is used to store the ultrasound index data and additional index data measured by the ultrasound device when receiving the user's inspection completion operation.

3. The system according to any one of claims 1-2, characterized in that: The system also includes a report generation module; the report generation module includes: The difference calculation submodule is used to calculate the difference between the measured value of each abnormal ultrasound index and the corresponding diagnostic critical value; An abnormal proportion calculation submodule, used to calculate the abnormal proportion of each abnormal ultrasound index based on the difference and the diagnostic critical value; A key indicator selection submodule is used to select a preset number of ultrasound indicators with high abnormal proportions as key indicators; A recommendation submodule, used for recommending the ultrasound image information corresponding to the acquisition of the key indicators to the user for selection; The report generation submodule is used to generate a report based on the ultrasound image selected by the user and the diagnosis result generated by the aortic coarctation prenatal diagnosis model.

4. A method for establishing a prenatal diagnosis model of aortic coarctation applied to the system as claimed in any one of claims 1 to 3, characterized in that: The method comprises: Obtain diagnostic accuracy information for multiple candidate ultrasound indicators; Determining a plurality of ultrasound indices for establishing a model from the plurality of candidate ultrasound indices according to the diagnostic accuracy information of the candidate ultrasound indices; A prenatal diagnosis model for aortic coarctation is established based on the diagnostic critical values ​​and diagnostic procedures of the ultrasound indicators corresponding to different fetal weeks.