Congenital heart disease progression assessment method and system based on ultrasound data

Abstract

The present invention provides a congenital heart disease process assessment method and system based on ultrasound data, which obtains a dynamic checklist of the target detection personnel corresponding to the detection time, and retrieves the attribute information of the target detection personnel, and generates an initial time series diagnosis tree based on the attribute information and the dynamic checklist; retrieves the ultrasound data intercepted by the physician end, obtains the annotation information and the interception time of the ultrasound data, and fills the corresponding ultrasound data into the dynamic checklist corresponding to the inspection node in the initial time series diagnosis tree based on the annotation information and the interception time to obtain the ultrasound time series diagnosis tree; determines the abnormal items in the ultrasound time series diagnosis tree, and splices the dynamic checklists in the ultrasound time series diagnosis tree based on the abnormal items to obtain an abnormal sequence list corresponding to the abnormal items; generates a heart prediction change map based on the trend prediction strategy and the ultrasound data in the abnormal sequence list and sends it to the detection end.

Description

Technical Field

[0001] The present invention relates to data processing technology, and in particular to a congenital heart disease progression assessment method and system based on ultrasound data. Background Art

[0002] Congenital heart disease (CHD) is the most common birth defect in children. Its incidence rate remains high worldwide, seriously affecting the quality of life and life expectancy of children. Early assessment of the progression of congenital heart disease is of great significance for formulating treatment plans and improving children's physical fitness.

[0003] Traditional diagnostic methods mainly rely on the doctor's clinical experience and cardiac ultrasound examinations, but this method has problems such as strong subjectivity and low diagnostic efficiency. In addition, due to the large number of test reports, in order to ensure the accuracy of the diagnostic results, it is necessary to analyze multiple examination items in multiple test reports one by one. When personnel do not perform tests in a timely manner according to the doctor's instructions and lack test data at the corresponding time, it will affect the diagnostic efficiency and may also reduce the accuracy of the diagnostic results.

[0004] Therefore, how to automatically summarize, analyze and display the detection data of corresponding abnormal items based on the actual detection data of the target personnel so that personnel can view it intuitively and improve diagnostic accuracy and efficiency has become an urgent problem that needs to be solved. Summary of the Invention

[0005] The embodiment of the present invention provides a congenital heart disease progression assessment method and system based on ultrasound data, which can automatically analyze, summarize and display the detection data of corresponding abnormal items based on the actual detection data of the target person, so that the person can view it intuitively, thereby improving the accuracy and efficiency of diagnosis.

[0006] A first aspect of an embodiment of the present invention provides a method for assessing the progression of congenital heart disease based on ultrasound data, comprising:

[0007] Obtaining a dynamic checklist of target detection personnel corresponding to detection times, retrieving attribute information of the target detection personnel, and generating an initial time series diagnosis tree based on the attribute information and the dynamic checklist;

[0008] Retrieving ultrasound data intercepted by the physician, obtaining annotation information and interception time of the ultrasound data, and filling the corresponding ultrasound data into a dynamic inspection table corresponding to an inspection node in the initial time series diagnostic tree based on the annotation information and the interception time, thereby obtaining an ultrasound time series diagnostic tree;

[0009] Determine an abnormal item in the ultrasound time series diagnosis tree, and perform splicing processing on each of the dynamic inspection tables in the ultrasound time series diagnosis tree based on the abnormal item to obtain an abnormal sequence table corresponding to the abnormal item;

[0010] According to the trend prediction strategy and the ultrasound data in the abnormal sequence table, a heart prediction evaluation change graph is generated and sent to the detection end.

[0011] Optionally, in a possible implementation of the first aspect, obtaining a dynamic checklist of target detection personnel corresponding to detection times, retrieving attribute information of the target detection personnel, and generating an initial time series diagnostic tree based on the attribute information and the dynamic checklist includes:

[0012] Obtaining a dynamic inspection table corresponding to the inspection time of the target inspection person, and retrieving attribute information of the target inspection person, wherein the dynamic inspection table includes an ultrasonic filling area and a coefficient evaluation area corresponding to a plurality of dynamic inspection items;

[0013] Constructing a person node according to the attribute information, extracting the month information and date information in the detection time, constructing a month summary node based on the month information, and connecting the month summary node to the person node;

[0014] Constructing a date subdivision node based on the date information, connecting the date subdivision node with the corresponding month summary node, and constructing a check node connected to the date subdivision node;

[0015] The dynamic check table is bound to the corresponding check node according to the detection time to obtain an initial timing diagnosis tree.

[0016] Optionally, in a possible implementation of the first aspect, retrieving ultrasound data intercepted by the physician, obtaining annotation information and interception time of the ultrasound data, and filling the corresponding ultrasound data into a dynamic inspection table corresponding to an inspection node in the initial time series diagnostic tree based on the annotation information and the interception time, to obtain the ultrasound time series diagnostic tree, includes:

[0017] Retrieving ultrasound data intercepted by the physician, and obtaining annotation information and interception time of the ultrasound data;

[0018] Determine the corresponding month summary node as the month update node and the date subdivision node as the date update node based on the interception time, and use the check node connected to the date update node as the node to be updated;

[0019] Determine a dynamic check table corresponding to the node to be updated as a dynamic update table, and locate a dynamic check item in the dynamic update table as an update item based on the annotation information;

[0020] The ultrasound data is filled into the ultrasound filling area corresponding to the update item, and the evaluation coefficient of the ultrasound data by the physician is received, and the evaluation coefficient is filled into the coefficient evaluation area corresponding to the update item to obtain an ultrasound time series diagnosis tree.

[0021] Optionally, in a possible implementation of the first aspect, when it is determined that the dynamic examination item corresponding to the dynamic examination table in the ultrasound time series diagnosis tree does not have an evaluation coefficient, the corresponding dynamic examination item is used as an item to be completed;

[0022] Obtaining date information of the item to be completed that does not have an evaluation coefficient as the missing date, and using the remaining date information corresponding to the item to be completed as the base date, and retrieving the evaluation coefficient of the item to be completed corresponding to the base date as the base coefficient;

[0023] According to the missing date, the base date and the base coefficient, a filling coefficient of the to-be-filled item corresponding to the missing date is obtained.

[0024] Optionally, in a possible implementation of the first aspect, obtaining, according to the missing date, the base date, and the base coefficient, a completion coefficient for the to-be-completed item corresponding to the missing date includes:

[0025] Sort the missing dates and the reference dates in ascending order to obtain a sequence of completion dates corresponding to the items to be completed;

[0026] sequentially counting adjacent missing dates in the complemented date sequence to obtain multiple missing date sequences;

[0027] When it is determined that only one side of the missing date sequence has a reference date, the corresponding missing date sequence is used as a trend missing sequence, and the reference date is used as a trend completion date;

[0028] Obtaining a reference coefficient corresponding to the trend filling date as a trend filling coefficient, and determining a filling coefficient for an item to be filled corresponding to a missing date in the trend missing sequence according to the trend filling coefficient and the trend filling date;

[0029] When it is determined that both sides of the missing date sequence are reference dates, the corresponding missing date sequence is used as the missing sequence at both ends, and the corresponding reference date is used as the date for filling in both ends;

[0030] A reference coefficient corresponding to the two-end filling date is obtained as the two-end filling coefficient, and a filling coefficient of the item to be filled corresponding to the missing date in the two-end missing sequence is determined according to the two-end filling coefficient and the two-end filling date.

[0031] Optionally, in a possible implementation of the first aspect, determining, based on the trend filling coefficient and the trend filling date, the filling coefficient of the to-be-filled item corresponding to the missing date in the trend missing sequence includes:

[0032] Obtaining two adjacent trend filling dates in the filling date sequence as trend calculation dates, and retrieving trend filling coefficients corresponding to the trend calculation dates as trend calculation coefficients;

[0033] Calculate the absolute value of the trend date difference based on the trend to obtain the trend date difference, and calculate the absolute value of the trend coefficient difference based on the trend to obtain the trend coefficient difference;

[0034] Based on the ratio of the trend coefficient difference and the trend date difference, the trend change rate corresponding to the two adjacent trend completion dates is obtained, and the average value of the trend change rate is obtained to obtain the average change rate;

[0035] When it is determined that the trend missing sequence is at the end of the supplementary date sequence, a trend supplementary date adjacent to the trend missing sequence in the supplementary date sequence is obtained as a starting prediction date, and a trend supplementary coefficient corresponding to the starting prediction date is retrieved as a starting prediction coefficient;

[0036] Obtaining the missing date in the trend missing sequence, obtaining a predicted date difference according to the difference between the missing date and the starting predicted date, and obtaining a prediction coefficient difference according to the product of the predicted date difference and the average change rate;

[0037] Obtaining a filling coefficient for the item to be filled corresponding to the missing date according to the sum of the starting prediction coefficient and the difference between the prediction coefficients;

[0038] When it is determined that the trend missing sequence is located at the front end of the supplementary date sequence, a trend supplementary date adjacent to the trend missing sequence in the supplementary date sequence is obtained as the end prediction date, and a trend supplementary coefficient corresponding to the end prediction date is retrieved as the end prediction coefficient;

[0039] Obtaining the missing date in the trend missing sequence, obtaining a predicted date difference according to the difference between the last predicted date and the missing date, and obtaining a prediction coefficient difference according to the product of the predicted date difference and the average change rate;

[0040] According to the difference between the final prediction coefficient and the prediction coefficient difference, the filling coefficient of the item to be filled corresponding to the missing date is obtained.

[0041] Optionally, in a possible implementation of the first aspect, determining, based on the two-end-padding coefficient and the two-end-padding date, the padding coefficient for the to-be-padded item corresponding to the missing date in the two-end-missing sequence includes:

[0042] Obtaining a sum of coefficients at both ends based on the sum of the padded coefficients at both ends, and obtaining a sum of dates at both ends based on the sum of the padded dates at both ends;

[0043] Obtain the missing date in the sequence with missing dates at both ends, and obtain the date proportion based on the ratio of the missing date to the sum of the dates at both ends;

[0044] Based on the date proportion and the product of the coefficient and value at both ends, the filling coefficient of the item to be filled corresponding to the missing date in the missing sequence at both ends is obtained.

[0045] Optionally, in a possible implementation of the first aspect, determining an abnormal item in the ultrasound time series diagnostic tree, and splicing each of the dynamic checklists in the ultrasound time series diagnostic tree based on the abnormal item to obtain an abnormal sequence list corresponding to the abnormal item includes:

[0046] Obtaining medical indicators of ultrasound data corresponding to each dynamic examination item in the ultrasound time series diagnosis tree, wherein the medical indicators include at least inner diameter, area, and length;

[0047] Retrieving the standard index interval of the dynamic examination item corresponding to each date information in the ultrasound time series diagnosis tree;

[0048] When it is determined that the medical indicator is not within the standard indicator range, treating the corresponding dynamic examination item as an abnormal item;

[0049] Based on the abnormal items, the ultrasonic filling areas in each of the dynamic inspection tables are intercepted as areas to be spliced, and the areas to be spliced ​​are sequentially spliced ​​according to the detection time to obtain an abnormal sequence table corresponding to the abnormal items.

[0050] Optionally, in a possible implementation of the first aspect, generating a predicted cardiac change map based on a trend prediction strategy and the ultrasound data in the abnormal sequence table and sending the map to the detection end includes:

[0051] Obtaining the medical indicators of the ultrasound data corresponding to each detection time in the abnormal sequence table as construction indicators, and the standard indicator interval corresponding to each detection time;

[0052] When it is determined that the construction index is less than the standard index interval, the minimum value of the standard index interval is retrieved as the minimum reference value;

[0053] Obtain a negative anomaly difference based on the difference between the construction indicator and the minimum benchmark value, obtain a negative anomaly proportion based on the ratio of the negative anomaly difference to the minimum benchmark value, and obtain a detection time of the construction indicator corresponding to the negative anomaly proportion as the negative anomaly time;

[0054] When it is determined that the construction index is greater than the standard index interval, the maximum value of the standard index interval is retrieved as the maximum reference value;

[0055] Obtaining a positive anomaly difference based on a difference between the construction indicator and the maximum reference value, obtaining a positive anomaly proportion based on a ratio of the positive anomaly difference to the maximum reference value, and obtaining a detection time of the construction indicator corresponding to the positive anomaly proportion as a positive anomaly time;

[0056] When it is determined that the construction indicator is within the standard indicator range, the detection time corresponding to the construction indicator is used as the standard time, and a preset standard ratio is configured for the standard time;

[0057] Retrieving a preset baseplate image and a preset change coordinate system, obtaining a center point of the preset baseplate image, aligning an origin of the preset change coordinate system with the center point, wherein the horizontal axis of the preset change coordinate system has a plurality of preset time points, and the vertical axis of the preset change coordinate system has a plurality of preset proportions;

[0058] Determine a negative trend point in the preset change coordinate system according to the negative abnormality time and the negative abnormality ratio;

[0059] Determining a positive trend point in the preset change coordinate system based on the positive anomaly time and the positive anomaly ratio;

[0060] Determine a standard trend point in the preset change coordinate system according to the standard time and the preset standard proportion;

[0061] Sort the detection times in the abnormal sequence table in ascending order to obtain a time connection sequence, and connect the negative trend point, the positive trend point and the standard trend point in sequence based on the order of the detection times in the time connection sequence to generate a heart prediction change map and send it to the detection end.

[0062] A second aspect of an embodiment of the present invention provides a congenital heart disease progression assessment system based on ultrasound data, comprising:

[0063] A generation module is used to obtain a dynamic checklist of target detection personnel corresponding to detection time, retrieve attribute information of the target detection personnel, and generate an initial time series diagnosis tree based on the attribute information and the dynamic checklist;

[0064] a filling module, configured to retrieve ultrasound data intercepted by the physician, obtain annotation information and interception time of the ultrasound data, and fill the corresponding ultrasound data into a dynamic inspection table corresponding to an inspection node in the initial time series diagnostic tree based on the annotation information and the interception time, thereby obtaining an ultrasound time series diagnostic tree;

[0065] a splicing module, configured to determine abnormal items in the ultrasound time series diagnosis tree, and splice the dynamic checklists in the ultrasound time series diagnosis tree based on the abnormal items to obtain an abnormal sequence list corresponding to the abnormal items;

[0066] The sending module is used to generate a heart prediction change map based on the trend prediction strategy and the ultrasound data in the abnormal sequence table and send it to the detection end.

[0067] According to a third aspect of an embodiment of the present invention, an electronic device is provided, comprising: a memory, a processor, and a computer program, wherein the computer program is stored in the memory, and the processor runs the computer program to execute the first aspect of the present invention and various methods that may be involved in the first aspect.

[0068] According to a fourth aspect of an embodiment of the present invention, a storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the first aspect of the present invention and various methods that may be involved in the first aspect.

[0069] The beneficial effects of the present invention are as follows:

[0070] 1. The present invention can automatically summarize and display the detection data of the corresponding abnormal items based on the actual detection data of the target personnel, so that the personnel can view it intuitively, thereby improving the accuracy and efficiency of diagnosis. First, the present invention can obtain an initial time series diagnostic tree based on the attribute information and the dynamic inspection table, so that the diagnostic data can be predicted and supplemented according to the initial time series diagnostic tree in the future, so as to provide more diagnostic data as a reference and improve the accuracy of the diagnostic results. Secondly, according to the annotation information and interception time of the ultrasound data, the corresponding ultrasound data is filled in the corresponding dynamic inspection table to generate an ultrasound time series diagnostic tree, so as to facilitate the dynamic management and analysis of the ultrasound data. Thus, the present invention can splice the dynamic inspection tables corresponding to the determined abnormal items to obtain an abnormal sequence table corresponding to the abnormal items, so that personnel can quickly locate and analyze heart abnormalities. Finally, the present invention can generate a heart prediction change map based on the ultrasound data in the abnormal sequence table and send it to the detection end for intuitive viewing by personnel, thereby improving the accuracy and efficiency of diagnosis.

[0071] 2. The present invention can accurately predict missing data and fill the obtained supplementary data into the corresponding dynamic inspection table, thereby increasing the amount of inspection data for diagnostic analysis, facilitating intuitive viewing by personnel, and thus improving the accuracy of diagnostic results. Specifically, the present invention can distinguish between multiple situations based on the position of the missing date sequence corresponding to the missing date in the supplementary date sequence, thereby adopting different methods to calculate the supplementary coefficient corresponding to the corresponding missing date, increasing the amount of diagnostic data, and thus improving the accuracy of the supplementary coefficient, thereby facilitating personnel to perform diagnostic analysis, thereby improving the accuracy of diagnostic results and the efficiency of root diagnosis.

[0072] 3. The present invention can generate a heart prediction change graph based on ultrasound data, thereby displaying the data changes of multiple examination results, so that personnel can intuitively view the changing trends of corresponding examination items and quickly determine the diagnosis results. Specifically, the present invention can generate a heart prediction change graph based on ultrasound data in the abnormal sequence table, clearly displaying the changes in the examination results of corresponding dynamic examination items, so that physicians can intuitively view the changing trends of the examination results of corresponding dynamic examination items through the heart prediction change graph, and quickly diagnose the corresponding fetal heart condition. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 This is a flow chart of a method for evaluating the progression of congenital heart disease based on ultrasound data provided by the present invention;

[0074] Figure 2 A schematic diagram of an initial timing diagnosis tree provided by the present invention;

[0075] Figure 3 A schematic diagram of a heart prediction change diagram provided by the present invention;

[0076] Figure 4 This is a schematic structural diagram of a congenital heart disease progression assessment system based on ultrasound data provided by the present invention;

[0077] Figure 5 This is a schematic diagram of the hardware structure of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0078] The following specific embodiments are used to describe the technical solution of the present invention in detail. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0079] like Figure 1 As shown, the present invention provides a flow chart of a method for evaluating the progression of congenital heart disease based on ultrasound data, the method for evaluating the progression of congenital heart disease based on ultrasound data comprising:

[0080] S1, obtaining a dynamic checklist of target detection personnel corresponding to detection time, and retrieving attribute information of the target detection personnel, and generating an initial time series diagnosis tree based on the attribute information and the dynamic checklist.

[0081] It should be noted that congenital heart disease is a congenital cardiovascular malformation caused by developmental defects of the fetal heart in the mother's body. Therefore, during pregnancy, fetal cardiac color Doppler ultrasound can be used to check the development of the fetal heart, mainly including the normal morphology and cardiac function of the heart. Since the time of physical examination and gestational cycle of each person are different, the corresponding fetal development is also different. Therefore, a dynamic checklist with corresponding examination items is required for different periods. The items to be checked in the dynamic checklist can change with time. Therefore, in order to facilitate the medical staff to review and diagnose the dynamic checklist, the dynamic checklist of the target detection personnel can be obtained to generate an initial time series diagnostic tree, so as to facilitate the subsequent prediction and completion of missing data, and improve the accuracy of the diagnostic results.

[0082] It can be understood that the target detection personnel are the personnel who perform the color ultrasound examination, for example, person A, the examination time is the time of the physical examination during pregnancy, for example, March 4, March 22, April 8, etc., the dynamic inspection form is the item form for the personnel to perform the inspection, the attribute information is the information of the target detection personnel, for example, it can be a name, and the initial time series diagnosis tree is the time series structure tree for the target detection personnel to perform the result diagnosis.

[0083] Through the above implementation, the present invention can obtain an initial time series diagnostic tree, so as to facilitate subsequent prediction and completion of diagnostic data based on the initial time series diagnostic tree, thereby providing more diagnostic data as a reference and improving the accuracy of the diagnostic result.

[0084] In some embodiments, the specific implementation of step S1 (obtaining a dynamic checklist of target detection personnel corresponding to detection times, retrieving attribute information of the target detection personnel, and generating an initial time series diagnostic tree based on the attribute information and the dynamic checklist) includes:

[0085] S11, obtaining a dynamic inspection table corresponding to the inspection time of the target inspection person, and retrieving the attribute information of the target inspection person, wherein the dynamic inspection table includes an ultrasonic filling area and a coefficient evaluation area corresponding to a plurality of dynamic inspection items.

[0086] It can be understood that in order to customize the initial timing diagnostic tree of the target detection personnel, a dynamic inspection table and corresponding attribute information of the target detection personnel corresponding to the detection time can be obtained, and the dynamic inspection table includes ultrasonic filling areas and coefficient evaluation areas corresponding to multiple dynamic inspection items.

[0087] Among them, the dynamic examination items are items for ultrasound examination, such as heart development, cardiac blood flow, heart rhythm, heart valve function, etc. The ultrasound filling area is the area filled with ultrasound images, and the coefficient evaluation area is the area filled with the evaluation coefficients of the examination results.

[0088] S12, constructing a person node according to the attribute information, extracting the month information and date information in the detection time, constructing a month summary node based on the month information, and connecting the month summary node with the person node.

[0089] It should be noted that the target detection personnel may conduct inspections multiple times. In order to facilitate subsequent data viewing and processing, the month information and date information can be extracted from the detection time, so that corresponding nodes can be constructed according to the month information and date information, so that information can be summarized according to time.

[0090] It is not difficult to understand that the detection time contains month information and date information, that is, the month corresponding to March 4th is March, and the date information is the 4th, so a month summary node can be constructed based on the month information.

[0091] It can be understood that the person node is the node corresponding to the person information, the month information is the month in the detection time, the date information is the date in the detection time, and the month summary node is the node corresponding to the month information.

[0092] For example: Figure 2 As shown, when the extracted month information includes March, April, and May, corresponding month summary nodes can be constructed according to March, April, and May, thereby connecting each month summary node with the person node.

[0093] S13: constructing a date subdivision node based on the date information, connecting the date subdivision node with the corresponding month summary node, and constructing a check node connected to the date subdivision node.

[0094] It can be understood that the date subdivision nodes are nodes corresponding to date information, and the check nodes are nodes that are checked in a one-to-one correspondence with the date subdivision nodes.

[0095] For example: Figure 2As shown, when the corresponding date information of 4th, 3.22nd, 4.8th and 4.25th is extracted from the detection times 3.4th, 3.22nd, 4.8th and 4.25th, the date subdivision nodes corresponding to 4th, 22nd, 8th and 25th are constructed respectively, so that the date subdivision nodes corresponding to 4th and 22nd are connected with the corresponding month summary node of March, and the date subdivision nodes corresponding to 8th and 25th are connected with the corresponding month summary node of April. Moreover, since the detection time needs to be checked, there is a corresponding dynamic check table, so that the check nodes corresponding to each date subdivision node can be generated one by one and connected accordingly.

[0096] S14 , binding the dynamic check table with the corresponding check node according to the detection time to obtain an initial timing diagnosis tree.

[0097] It is understandable that the corresponding dynamic check table is bound to the corresponding check node according to the detection time, thereby generating an initial time series diagnostic tree, so that the dynamic check table bound to the corresponding check node can be retrieved and analyzed according to the date information, thereby improving data processing efficiency.

[0098] S2, retrieve the ultrasound data intercepted by the physician end, obtain the annotation information and interception time of the ultrasound data, and fill the corresponding ultrasound data into the dynamic inspection table corresponding to the inspection node in the initial time series diagnosis tree based on the annotation information and the interception time to obtain the ultrasound time series diagnosis tree.

[0099] It should be noted that in order to enable doctors to obtain accurate diagnosis results, doctors can intercept ultrasound data during the ultrasound examination and annotate the ultrasound data so that doctors can carefully review the detection information in the ultrasound examination during the annotation process.

[0100] It is understandable that when the physician performs an ultrasound examination on the target person, the data of the examination process will be intercepted to obtain the corresponding ultrasound data, such as color ultrasound images, so as to fill the corresponding ultrasound data in the corresponding dynamic examination table, thereby facilitating the personnel to view the examination results.

[0101] Among them, the physician end is the instrument terminal used by the physician for examination, the ultrasound data is the image data in the ultrasound examination, the annotation information is the information used to annotate and describe the ultrasound data, such as small tricuspid valve hypoplasia, etc., the interception time is the time when the data is intercepted during the ultrasound detection, and the ultrasound time series diagnosis tree is a result tree filled with ultrasound data in the dynamic inspection table corresponding to the corresponding inspection node.

[0102] It is not difficult to understand that the corresponding annotation information and interception time can be obtained through the ultrasound data, so that the corresponding ultrasound data can be filled in the dynamic inspection table of the inspection node connected to the corresponding date subdivision node in the initial time series diagnosis tree, thereby completing the filling of the detection data, so as to facilitate the subsequent prediction and completion of the data corresponding to the missing date, which is convenient for personnel to view and accurately diagnose.

[0103] In some embodiments, a specific implementation of step S2 (retrieving ultrasound data intercepted by the physician, obtaining annotation information and interception time of the ultrasound data, and filling the corresponding ultrasound data into the dynamic inspection table corresponding to the inspection node in the initial time series diagnostic tree based on the annotation information and the interception time to obtain the ultrasound time series diagnostic tree) includes:

[0104] S21, retrieve the ultrasound data intercepted by the physician, and obtain the annotation information and interception time of the ultrasound data.

[0105] It is understandable that the corresponding annotation information and the interception time are obtained through the ultrasound data so that the corresponding ultrasound data can be filled in the corresponding dynamic inspection table later, thereby facilitating personnel viewing.

[0106] S22, based on the interception time, determining the corresponding month summary node as the month update node and the date subdivision node as the date update node, and taking the check node connected to the date update node as the node to be updated.

[0107] It is understandable that the corresponding month information and date information can be obtained from the intercepted time, thereby determining the nodes that need to be filled and updated with information later.

[0108] Among them, the month update node is the month summary node corresponding to the month information in the interception time, the date update node is the date subdivision node corresponding to the date information in the interception time, and the node to be updated is the check node connected to the date update node.

[0109] For example: when the interception time is March 4, 2024, the corresponding month update node is the month summary node corresponding to March, and the date update node is the date subdivision node corresponding to the 4th under the month update node. Therefore, the inspection node connected to the date update node corresponding to the 4th is the node to be updated, so that the dynamic inspection table bound to the node to be updated can be filled with and updated with ultrasound data.

[0110] S23: Determine the dynamic check table corresponding to the node to be updated as the dynamic update table, and locate the dynamic check items in the dynamic update table as update items based on the annotation information.

[0111] It can be understood that the dynamic update table is a dynamic check table corresponding to the node to be updated, and the update item is a dynamic check item corresponding to the annotation information in the dynamic update table.

[0112] For example, when the annotation information is that the tricuspid valve is small and hypoplastic, the heart valve function dynamic examination item in the dynamic update table can be used as an update item.

[0113] Through the above implementation, the present invention can determine the update items so as to subsequently fill the corresponding ultrasound data in the corresponding ultrasound filling area.

[0114] S24, filling the ultrasound data into the ultrasound filling area corresponding to the update item, receiving the evaluation coefficient of the ultrasound data from the physician end, filling the evaluation coefficient into the coefficient evaluation area corresponding to the update item, and obtaining an ultrasound timing diagnosis tree.

[0115] It is understandable that medical staff can evaluate the corresponding dynamic examination items based on the detected ultrasound data. For example, based on the detected inner diameter comment data of the left atrium and right atrium of the fetus, they can compare them with the normal inner diameter corresponding to the standard cycle to obtain the evaluation coefficient, and then fill the corresponding evaluation coefficient in the updated coefficient evaluation area corresponding to the updated data to obtain the ultrasound timing diagnosis tree.

[0116] Among them, the evaluation coefficient is a coefficient for evaluating and diagnosing the results of the fetal heart dynamic detection project based on the detected ultrasound data. It is a coefficient that is manually evaluated based on the examination results. When the examination data differs more from the standard normal data, the corresponding evaluation coefficient differs more from 0. When the evaluation coefficient tends to 0, it can be said that the corresponding heart examination result is more normal.

[0117] Through the above implementation, the present invention can obtain an ultrasound time series diagnostic tree, so that the dynamic inspection table corresponding to the missing date can be supplemented with data through the ultrasound time series diagnostic tree, so as to provide physicians with more comprehensive reference data, thereby enhancing the accuracy of the diagnosis results.

[0118] It should be noted that the target detection personnel may not perform the examination at the regular detection time due to personal reasons, resulting in missing diagnostic data in the dynamic inspection table of the inspection node connected to the corresponding date subdivision node in the ultrasound time series diagnosis tree, which in turn affects the physician's diagnosis of the fetal heart. Therefore, in order to enhance the accuracy of the diagnostic results, the missing data can be accurately predicted and supplemented. Therefore, in some embodiments, the following is further included:

[0119] A1: When it is determined that the dynamic examination item corresponding to the dynamic examination table in the ultrasound time series diagnosis tree does not have an evaluation coefficient, the corresponding dynamic examination item is set as an item to be supplemented.

[0120] It can be understood that the items to be supplemented are dynamic examination items for which the dynamic examination items corresponding to the dynamic examination table in the ultrasound time series diagnosis tree do not have evaluation coefficients.

[0121] For example: when it is determined that the dynamic examination item (heart valve function) corresponding to the dynamic examination table does not have a corresponding evaluation coefficient, it can be said that the corresponding detection time has not been checked, and thus the dynamic examination item in the corresponding dynamic examination table does not have a corresponding evaluation coefficient. The corresponding dynamic examination item (heart valve function) can be used as an item to be filled in, so that the missing data of the item to be filled can be predicted and filled in according to the evaluation coefficient of the same dynamic examination item.

[0122] A2, obtaining the date information of the item to be completed that does not have an evaluation coefficient as the missing date, and using the remaining date information corresponding to the item to be completed as the reference date, and retrieving the evaluation coefficient of the item to be completed corresponding to the reference date as the reference coefficient.

[0123] It can be understood that the missing date is the date information when the item to be completed does not have an evaluation coefficient, the base date is the date information when the item to be completed has an evaluation coefficient, and the base coefficient is the evaluation coefficient of the item to be completed corresponding to the base date.

[0124] For example: when the item to be completed is heart valve function, and it is determined that the date information corresponding to the dynamic inspection table without an evaluation coefficient for heart valve function is 22nd in the March information, then March 22nd can be used as the missing date. When the other date information containing the dynamic inspection item of heart valve function is April 8th and April 25th, the corresponding benchmark dates are April 8th and April 25th, so that the evaluation coefficients of the items to be completed (heart valve function) in the dynamic inspection table corresponding to April 8th and April 25th can be used as benchmark coefficients.

[0125] Through the above implementation, the present invention can determine the missing date, the base date and the base coefficient, so as to subsequently obtain the filling coefficient of the item to be filled in the corresponding missing date.

[0126] A3, according to the missing date, the base date and the base coefficient, obtains a completion coefficient for the item to be completed corresponding to the missing date.

[0127] It can be understood that the filling coefficient of the item to be filled in corresponding to the missing date is determined based on the missing date, the benchmark date and the benchmark coefficient, so that it can be filled in the corresponding area later, thereby increasing the reference data for diagnosis and improving the accuracy of the diagnosis result.

[0128] The completion coefficient is a coefficient for predicting and completing the evaluation coefficient of the to-be-completed items corresponding to the dynamic checklist.

[0129] In some embodiments, a specific implementation of step A3 (obtaining a completion coefficient for the item to be completed corresponding to the missing date based on the missing date, the base date, and the base coefficient) includes:

[0130] A31 , sorting the missing dates and the reference dates in ascending order to obtain a sequence of completion dates corresponding to the items to be completed.

[0131] It is understandable that in order to determine the calculation method for obtaining the corresponding filling coefficient, the missing dates and the reference dates can be sorted in ascending order, so as to obtain a filling date sequence corresponding to the items to be filled.

[0132] The completion date sequence is a date sequence in which the missing dates and the base dates corresponding to the items to be completed are sorted in ascending order.

[0133] A32, sequentially counting adjacent missing dates in the complemented date sequence to obtain multiple missing date sequences.

[0134] It should be noted that, in multiple inspection times with the same dynamic inspection item, the target inspection personnel may miss multiple inspections, so when the date series is completed to count the missing dates, there may be multiple missing date series.

[0135] It can be understood that the missing date sequence is a date sequence of statistically adjacent missing dates.

[0136] For example, when the completed date sequence is (3.4, 3.22, 4.8, 4.25, 5.12, 5.30), where the missing dates are 3.22, 5.12, and 5.30, the two missing date sequences obtained are (3.22) and (5.12, 5.30).

[0137] Through the above-mentioned implementation, the present invention can obtain multiple missing date sequences, so as to subsequently determine the method of obtaining the corresponding missing date filling coefficient according to the position of the missing date sequence in the filled date sequence, so as to obtain an accurate filling coefficient and improve the accuracy of the diagnosis result.

[0138] A33: When it is determined that only one side of the missing date sequence has a reference date, the corresponding missing date sequence is used as a trend missing sequence, and the reference date is used as a trend completion date.

[0139] It can be understood that the trend missing sequence is a missing date sequence having a reference date on only one side of the missing date sequence, and the trend supplementary date is the reference date in the corresponding supplementary date.

[0140] For example: when the completed date sequence is (3.4, 3.22, 4.8, 4.25, 5.12, 5.30), the missing date sequences are (3.22) and (5.12, 5.30), among which the missing date sequence (5.12, 5.30) has a base date on only one side, so the missing date sequence (5.12, 5.30) can be used as a trend missing sequence, and since the completed date sequence (3.4, 3.22, 4.8, 4.25, 5.12, 5.30) has base dates of 3.4, 4.8 and 4.25, the base dates of 3.4, 4.8 and 4.25 can be used as trend completed dates.

[0141] Through the above implementation, the present invention can determine the trend filling date, so as to facilitate the subsequent determination of the filling coefficient corresponding to the missing date in the trend missing sequence according to the evaluation coefficient of the trend filling date.

[0142] A34 is to obtain a reference coefficient corresponding to the trend filling date as a trend filling coefficient, and determine a filling coefficient for the item to be filled corresponding to the missing date in the trend missing sequence according to the trend filling coefficient and the trend filling date.

[0143] It can be understood that the trend filling coefficient and the trend filling date are used to predict and fill the filling coefficient of the item to be filled corresponding to the missing date in the trend missing sequence.

[0144] Among them, the trend filling coefficient is the benchmark coefficient corresponding to the trend filling date.

[0145] Through the above implementation, the present invention can determine the corresponding padding coefficient, so that the padding coefficient can be subsequently filled in the corresponding coefficient evaluation area for personnel to view and reference, thereby improving the accuracy of the diagnosis result.

[0146] In some embodiments, a specific implementation of step A34 (determining the filling coefficient of the item to be filled corresponding to the missing date in the trend missing sequence based on the trend filling coefficient and the trend filling date) includes:

[0147] A341, obtaining two adjacent trend filling dates in the filling date sequence as trend calculation dates, and retrieving trend filling coefficients corresponding to the trend calculation dates as trend calculation coefficients.

[0148] It can be understood that the trend calculation dates are the two adjacent trend completion dates in the completion date sequence. For example, when the completion date sequence is (3.4, 3.22, 4.8, 4.25, 5.12, 5.30), among them, 3.4, 4.8 and 4.25 are trend completion dates. Since 4.8 and 4.25 are two adjacent dates, 4.8 and 4.25 can be used as trend calculation dates.

[0149] The trend calculation coefficient is the trend filling coefficient corresponding to the trend calculation date.

[0150] Through the above implementation, the present invention can obtain the trend calculation date and the trend calculation coefficient, so that an accurate filling coefficient can be calculated for the corresponding missing date later.

[0151] A342, calculates the absolute value of the trend date difference based on the trend to obtain the trend date difference, and calculates the absolute value of the trend coefficient difference based on the trend to obtain the trend coefficient difference.

[0152] It can be understood that the trend date difference is the absolute value of the trend calculation date difference. For example, when the trend calculation dates are April 8 and April 25, the trend date difference can be obtained as 17 days.

[0153] Among them, the trend coefficient difference is the absolute value of the trend calculation coefficient difference. For example, when the trend calculation dates are April 8 and April 25, the trend calculation coefficient corresponding to April 8 is -2, and the trend calculation coefficient corresponding to April 25 is -0.3, so the trend coefficient difference is 1.7.

[0154] A343, based on the ratio of the trend coefficient difference and the trend date difference, obtain the trend change rate corresponding to the two adjacent trend completion dates, and obtain the average value of the trend change rate to obtain the average change rate.

[0155] It can be understood that the trend change rate is the numerical change rate of the trend calculation coefficient, that is, the ratio of the trend coefficient difference to the trend date difference. For example, when the trend coefficient difference is 1.7 and the trend date difference is 17, the corresponding trend change rate is 1.7 / 17=0.1.

[0156] It is not difficult to understand that there may be multiple adjacent trend filling dates in the filling date sequence, so there may be multiple trend date difference values ​​and corresponding trend coefficient difference values, and there will be multiple trend change rates. In order to make the predicted filling coefficient closer to the actual data, the average value of multiple trend change rates can be obtained to obtain a more accurate filling coefficient.

[0157] The average change rate is the average of all trend change rates. For example, when the multiple trend change rates corresponding to the dynamic inspection item are 0.1, 0.05, and 0.15, the average change rate can be obtained as (0.1+0.05+0.15) / 3=0.1.

[0158] A344, when it is determined that the trend missing sequence is at the end of the supplementary date sequence, the trend supplementary date adjacent to the trend missing sequence in the supplementary date sequence is obtained as the starting prediction date, and the trend supplementary coefficient corresponding to the starting prediction date is retrieved as the starting prediction coefficient.

[0159] It should be noted that, since the data results of fetal heart examinations are constantly changing with the fetal development cycle, when filling in missing data based on existing data, the data at the middle position is usually predicted and filled based on the data at both ends of the missing position. However, in some special implementations, since the fetus is constantly developing, new examination items will appear at different times, resulting in no data available for reference prediction before the missing date. It is necessary to determine the position of the trend missing sequence in the filling date sequence, and thus it is necessary to determine whether the date of missing data is before or after the trend filling date corresponding to the trend filling coefficient, and then determine whether the change trend of the calculated filling coefficient is downward or upward, so as to obtain a more accurate filling coefficient for reference by personnel diagnosis.

[0160] It can be understood that when the trend missing sequence is at the end of the supplementary date sequence, the starting prediction date is the trend supplementary date adjacent to the trend missing sequence in the supplementary date sequence. For example, when the trend missing sequence (5.12, 5.30) is at the end of the supplementary date sequence (3.4, 3.22, 4.8, 4.25, 5.12, 5.30), the trend supplementary date 4.25 adjacent to the trend missing sequence (5.12, 5.30) can be used as the starting prediction date.

[0161] The starting prediction coefficient is the trend filling coefficient corresponding to the starting prediction date, for example, it can be -0.3.

[0162] Through the above implementation, the present invention can obtain the starting prediction date, so as to facilitate the subsequent determination of the filling coefficient corresponding to the missing date.

[0163] A345, obtains the missing date in the trend missing sequence, obtains the predicted date difference based on the difference between the missing date and the starting predicted date, and obtains the prediction coefficient difference based on the product of the predicted date difference and the average change rate.

[0164] It can be understood that the difference in predicted dates is the difference between the missing date and the starting predicted date. For example, when the missing dates in the trend missing sequence are 5.12 and 5.30, and the starting predicted date is 4.25, the difference in predicted dates between the starting predicted date 4.25 and the missing date 5.12 is 17, and the difference in predicted dates between the starting predicted date 4.25 and the missing date 5.30 is 35, so that the corresponding prediction coefficient difference can be obtained according to the difference in predicted dates.

[0165] Among them, the prediction coefficient difference is the difference between the corresponding evaluation coefficients between the missing date and the starting prediction date, that is, the product of the prediction date difference and the average change rate. For example, when the prediction date difference is 17 and the average change rate is 0.1, the prediction coefficient difference is 1.7.

[0166] Through the above implementation, the present invention can obtain the prediction coefficient difference between each missing date and the starting prediction date, so as to facilitate the subsequent determination of the filling coefficient of the corresponding missing date.

[0167] A346, based on the sum of the starting prediction coefficient and the difference between the prediction coefficients, obtains the filling coefficient of the item to be filled corresponding to the missing date.

[0168] It should be noted that when it is determined that the trend missing sequence is at the end of the filled date sequence, since the development of the fetus is gradually strengthening, the filling coefficient corresponding to the missing date at the end should be greater than the initial prediction coefficient. Therefore, when predicting the filling coefficient corresponding to the missing date based on the trend filling coefficient of the trend filling date, it is necessary to add the initial prediction coefficient and the prediction coefficient to obtain the sum value.

[0169] It can be understood that when the starting prediction coefficient and the prediction coefficient difference are determined, the corresponding starting prediction coefficient and prediction coefficient difference can be added to obtain the sum value, and the sum value is the supplementary coefficient of the corresponding missing date. For example, when the starting prediction coefficient is -0.3, and the difference in prediction coefficients corresponding to the missing date 5.8 and the starting prediction date 4.25 is 1.7, then the supplementary coefficient for the missing date 5.8 can be obtained as -0.3+1.7=1.4.

[0170] Through the above implementation, the present invention can determine the filling coefficient corresponding to the missing date, so that doctors can refer to the data, thereby improving the accuracy of the diagnosis result.

[0171] A347, when it is determined that the trend missing sequence is at the front end of the supplementary date sequence, the trend supplementary date adjacent to the trend missing sequence in the supplementary date sequence is obtained as the end prediction date, and the trend supplementary coefficient corresponding to the end prediction date is retrieved as the end prediction coefficient.

[0172] It can be understood that when the trend missing sequence is at the front end of the filling date sequence, the end prediction date is the trend filling date adjacent to the trend missing sequence in the filling date sequence. For example, when the trend missing sequence (2.20) is at the front end of the filling date sequence (2.20, 3.4, 3.22, 4.8, 4.25, 5.12, 5.30), the trend filling date 3.4 adjacent to the trend missing sequence (2.20) can be used as the end prediction date.

[0173] The end prediction coefficient is the trend filling coefficient corresponding to the end prediction date, for example, it can be -0.4.

[0174] Through the above implementation, the present invention can obtain the end prediction date and the corresponding end prediction coefficient, so as to obtain the filling coefficient corresponding to the missing date at the front end of the filling date sequence.

[0175] A348, obtains the missing date in the trend missing sequence, obtains the predicted date difference based on the difference between the last predicted date and the missing date, and obtains the prediction coefficient difference based on the product of the predicted date difference and the average change rate.

[0176] It can be understood that when the trend missing sequence is at the front end, the corresponding prediction date difference is the difference between the last prediction date and the missing date. For example, when the missing date is February 20 and the last prediction date is March 4, the corresponding prediction date difference is 12 days, and then the prediction coefficient difference is the product of the prediction date difference and the average change rate. For example, when the prediction date difference is 12 and the average change rate is 0.1, the prediction coefficient difference is 1.2.

[0177] A349, based on the difference between the final prediction coefficient and the prediction coefficient difference, obtains the completion coefficient of the item to be completed corresponding to the missing date.

[0178] It should be noted that when it is determined that the trend missing sequence is at the front end of the filling date sequence, since the development of the fetus is gradually strengthening, the filling coefficient corresponding to the missing date at the front end should be smaller than the prediction coefficient at the end. Therefore, when predicting the filling coefficient corresponding to the missing date based on the trend filling coefficient of the trend filling date, it is necessary to subtract the prediction coefficient difference from the prediction coefficient at the end.

[0179] It can be understood that the difference between the final prediction coefficient and the prediction coefficient difference is calculated to obtain the completion coefficient of the item to be completed corresponding to the missing date.

[0180] A35, when it is determined that both sides of the missing date sequence are reference dates, the corresponding missing date sequence is used as the missing sequence at both ends, and the corresponding reference date is used as the completion date at both ends.

[0181] It can be understood that the missing sequence at both ends is the missing date sequence corresponding to when both sides of the missing date sequence are the base dates, and the complementary dates at both ends are the base dates on both sides of the missing date sequence. For example, when the missing date sequence is (March 22nd), and the corresponding dates 3.4th and 4.8th on both sides are the base dates, the missing date sequence (March 22nd) can be used as the missing sequence at both ends, and the base dates 3.4th and 4.8th can be used as the complementary dates at both ends.

[0182] Through the above implementation, the present invention can determine the two-end filling date, so that the filling coefficient of the missing date can be determined according to the reference coefficient corresponding to the two-end filling date.

[0183] A36 is to obtain a reference coefficient corresponding to the two-end filling date as the two-end filling coefficient, and determine a filling coefficient for the item to be filled corresponding to the missing date in the two-end missing sequence based on the two-end filling coefficient and the two-end filling date.

[0184] It can be understood that the padding coefficient at both ends is the base coefficient corresponding to the padding date at both ends.

[0185] It is not difficult to understand that the evaluation coefficient of the missing date in the middle can be predicted based on the filling coefficients at both ends and the filling dates at both ends, thereby obtaining the filling coefficients of the items to be filled corresponding to the missing dates in the missing sequences at both ends.

[0186] In some embodiments, a specific implementation of step A36 (determining the padding coefficients of the items to be padded corresponding to the missing dates in the missing-at-end sequence based on the padding coefficients and the padding dates) includes:

[0187] A361, obtains the sum of the coefficients at both ends based on the sum of the coefficients padded at both ends, and obtains the sum of the dates at both ends based on the sum of the dates padded at both ends.

[0188] It can be understood that the sum of the coefficients at both ends is the sum of the coefficients padded at both ends, and the sum of the dates at both ends is the sum of the dates padded at both ends.

[0189] For example: when the padding coefficients at both ends are -0.4 and -2 respectively, the corresponding sum of the coefficients at both ends is -0.4+(-2)=-2.4. When the corresponding padding dates at both ends are March 4th and April 8th respectively, starting from January 1st, March 4th corresponds to the 63rd day and April 8th to the 97th day, so the corresponding sum of the dates at both ends is 160 days.

[0190] Through the above implementation, the present invention can obtain the coefficient sum value at both ends and the date sum value at both ends, so as to facilitate the subsequent calculation of the filling coefficient of the corresponding missing date.

[0191] A362, obtaining the missing dates in the sequence with missing dates at both ends, and obtaining the date proportion according to the ratio of the missing dates to the sum of the dates at both ends.

[0192] It can be understood that the date ratio is the ratio of the missing date to the sum of the dates at both ends. For example, when the missing date is March 22 and January 1 corresponds to the 82nd day, and the sum of the dates at both ends is 160 days, the date ratio is 82 / 160.

[0193] A363, based on the date proportion and the product of the coefficient and value at both ends, obtains the filling coefficient of the item to be filled corresponding to the missing date in the missing sequence at both ends.

[0194] It can be understood that the date proportion is multiplied by the coefficient and value at both ends to obtain the filling coefficient of the missing date corresponding to the item to be filled in the missing sequence at both ends.

[0195] For example, when the date ratio is 82 / 160 and the sum of the coefficients at both ends is -2.4, the padding coefficient can be obtained as -2.4×(82 / 160)=-123 / 120.

[0196] S3, determining abnormal items in the ultrasound time series diagnosis tree, and splicing the dynamic checklists in the ultrasound time series diagnosis tree based on the abnormal items to obtain an abnormal sequence list corresponding to the abnormal items.

[0197] It should be noted that in order to facilitate the diagnosis of medical personnel, the dynamic inspection tables with abnormal items can be spliced ​​together, so that the inspection result data of the same dynamic inspection item can be statistically summarized, so that personnel can quickly obtain diagnostic results through the abnormal sequence table, thereby improving diagnostic efficiency.

[0198] It can be understood that the abnormal items are dynamic inspection items with abnormal inspection results, and the abnormal sequence table is a form obtained by splicing multiple dynamic inspection tables with the same abnormal items.

[0199] In some embodiments, the specific implementation of step S3 (determining abnormal items in the ultrasound time series diagnostic tree, and splicing the dynamic checklists in the ultrasound time series diagnostic tree based on the abnormal items to obtain an abnormal sequence list corresponding to the abnormal items) includes:

[0200] S31, obtaining medical indicators of ultrasound data corresponding to each dynamic examination item in the ultrasound time series diagnosis tree, wherein the medical indicators at least include inner diameter, area and length.

[0201] It is understandable that in order to determine whether each dynamic examination item has abnormalities, the medical indicators of the corresponding ultrasound data can be retrieved for subsequent comparison with the corresponding standard normal data, thereby automatically obtaining abnormal results.

[0202] Among them, medical indicators are indicators of medical examination, including inner diameter, area and length, such as the inner diameter of the left ventricle and the inner diameter of the right ventricle of the heart.

[0203] S32, retrieving the standard index interval of the dynamic examination item corresponding to each date information in the ultrasound time series diagnosis tree.

[0204] It should be noted that since the fetus is constantly developing with the changes in the gestational cycle, for example, the ventricles grow from nothing to something and then from small to large, the standard values ​​corresponding to the medical indicators when examined at different times are also different. Therefore, the standard indicator ranges for the corresponding dynamic examination items can be retrieved based on the date information.

[0205] It can be understood that the standard indicator range is the standard normal data range of the corresponding indicators of the dynamic detection items. For example, the normal right atrial diameter range corresponding to the gestational period of 17 to 28 weeks is (7.1mm, 11.3mm), and the left atrial diameter range is (6.3mm, 8.9mm).

[0206] S33: When it is determined that the medical indicator is not within the standard indicator range, the corresponding dynamic examination item is regarded as an abnormal item.

[0207] It is understandable that when the corresponding medical indicator is not within the standard indicator range, it can be said that the examination result of the corresponding dynamic examination item is abnormal, and thus the dynamic examination item can be regarded as an abnormal item.

[0208] Through the above implementation, the present invention can determine abnormal items, so as to facilitate subsequent splicing of dynamic check tables with corresponding abnormal items to obtain an abnormal sequence table.

[0209] S34, based on the abnormal items, intercepting the ultrasonic filling areas in each of the dynamic inspection tables as areas to be spliced, and sequentially splicing the areas to be spliced ​​according to the detection time to obtain an abnormal sequence table corresponding to the abnormal items.

[0210] It can be understood that the area to be spliced ​​is the ultrasonic filling area corresponding to the abnormal item in the dynamic inspection table, so that the intercepted area to be spliced ​​can be spliced ​​in chronological order according to the detection time, thereby obtaining the abnormal sequence list corresponding to the spliced ​​abnormal item.

[0211] Through the above-mentioned implementation, the present invention can obtain an abnormal sequence list, so that doctors can quickly diagnose congenital heart disease according to the abnormal sequence list of corresponding abnormal items, thereby improving diagnostic efficiency.

[0212] It should be noted that when there are abnormal items in the dynamic inspection table corresponding to the missing date, there is no corresponding ultrasound image data in the area to be spliced ​​corresponding to the detection time in the spliced ​​abnormal sequence table. In order for personnel to clearly observe the image data corresponding to the abnormal item, the ultrasound data corresponding to the base date can be used to fill in the ultrasound filling area of ​​the missing date for easy viewing by personnel. Therefore, in some embodiments, the following is further included:

[0213] A medical indicator corresponding to the missing date is obtained based on the supplemented data, and the ultrasound data is scaled according to the medical indicator to obtain a supplementary image corresponding to the missing date.

[0214] It can be understood that the medical indicators of the abnormal items corresponding to the missing date can be obtained by comparing the completed data corresponding to the missing date with the evaluation coefficient corresponding to the reference time. For example, when the medical indicator of the reference date is an inner diameter of 6mm, the corresponding evaluation coefficient is -1, and the completion coefficient of the missing date is -1.5, so the medical indicator corresponding to the missing date can be obtained as an inner diameter of 4mm. Therefore, the medical indicator of the reference date can be compared with the medical indicator of the missing date to obtain the corresponding scaling factor, for example, the scaling factor is 2 / 3, and then the ultrasound data of the inner diameter of 6mm on the reference date can be scaled according to the scaling factor to obtain the supplementary image of the missing date.

[0215] The supplementary image is the ultrasound data image corresponding to the missing date, and the scaling factor is the scaling factor for scaling the ultrasound data of the reference date.

[0216] Through the above implementation, the present invention can obtain a supplementary image corresponding to the missing date, so as to facilitate subsequent filling in the corresponding area to be spliced.

[0217] The supplementary image is retrieved and filled into the area to be spliced ​​corresponding to the missing time in the abnormal sequence table.

[0218] It can be understood that the supplementary image is filled in the area to be spliced ​​corresponding to the corresponding missing time in the abnormal sequence table, so as to complete the ultrasound data of the corresponding detection time in the abnormal sequence table, and obtain an abnormal sequence table with complete data for diagnosis and review by personnel, so as to quickly locate the corresponding cause of the heart disease based on the examination results of the abnormal items and improve the diagnostic efficiency.

[0219] S4, generating a heart prediction change graph based on the trend prediction strategy and the ultrasound data in the abnormal sequence table and sending it to the detection end.

[0220] It should be noted that in order to clearly display the changes in the examination results of the corresponding dynamic examination items, a heart prediction change diagram can be generated based on the ultrasound data in the abnormal sequence table, so that doctors can intuitively view the changing trends of the examination results of the corresponding dynamic examination items through the heart prediction change diagram, and thus quickly diagnose the corresponding fetal heart condition.

[0221] It can be understood that the heart prediction change graph is a data change graph of the heart ultrasound examination results, and the detection end is a signal terminal corresponding to the heart detection diagnostic personnel, such as a mobile phone, computer, etc.

[0222] In some embodiments, the specific implementation of step S4 (generating a cardiac prediction change map based on the trend prediction strategy and the ultrasound data in the abnormal sequence table and sending it to the detection end) includes:

[0223] S401 , obtaining medical indicators of ultrasound data corresponding to each detection time in the abnormal sequence table as construction indicators, and standard indicator intervals corresponding to each detection time.

[0224] It should be noted that in order to determine the data coordinates corresponding to each detection time in the heart prediction change diagram, the medical indicators of the corresponding ultrasound data and the corresponding standard indicator intervals can be obtained, so as to facilitate the subsequent determination of the corresponding data change curve.

[0225] It can be understood that the constructed index is a medical index corresponding to the ultrasound data at each detection time in the abnormal sequence table.

[0226] S402: When it is determined that the construction index is smaller than the standard index interval, the minimum value of the standard index interval is retrieved as the minimum reference value.

[0227] It can be understood that when the constructed indicator is smaller than the standard indicator interval, the minimum benchmark value is the minimum value of the standard indicator interval.

[0228] For example, when the detected construction index of the right atrium is 6.1 mm in inner diameter, which is smaller than the standard index interval (7.1 mm, 11.3 mm), the minimum value of the standard index interval, 7.1 mm, can be used as the minimum benchmark value.

[0229] S403, obtain a negative anomaly difference based on the difference between the construction indicator and the minimum benchmark value, obtain a negative anomaly proportion based on the ratio of the negative anomaly difference to the minimum benchmark value, and obtain the detection time of the construction indicator corresponding to the negative anomaly proportion as the negative anomaly time.

[0230] It can be understood that the negative abnormal difference is the difference between the construction index and the minimum benchmark value. For example, when the construction index is an inner diameter of 6.1 mm and the minimum benchmark value is 7.1 mm, the negative abnormal difference is 6.1-7.1=-1, and the negative abnormality ratio is the ratio of the negative abnormal difference to the minimum benchmark value, that is, -1 / 7.1.

[0231] Among them, the negative anomaly time is the detection time of the indicator corresponding to the negative anomaly ratio.

[0232] Through the above-mentioned implementation, the present invention can determine the negative abnormality ratio and negative abnormality time, so as to obtain the corresponding heart prediction change map subsequently.

[0233] S404: When it is determined that the construction index is greater than the standard index interval, the maximum value of the standard index interval is retrieved as the maximum reference value.

[0234] It can be understood that when the constructed indicator is greater than the standard indicator interval, the maximum benchmark value is the maximum value of the standard indicator interval.

[0235] For example, when the constructed indicator is that the inner diameter of the right atrium is 12.43 mm, which is greater than the maximum value of the standard indicator range (7.1 mm, 11.3 mm), the maximum benchmark value can be obtained as 11.3 mm.

[0236] S405: Obtain a positive anomaly difference based on the difference between the construction indicator and the maximum benchmark value, obtain a positive anomaly proportion based on the ratio of the positive anomaly difference to the maximum benchmark value, and obtain the detection time of the construction indicator corresponding to the positive anomaly proportion as the positive anomaly time.

[0237] It can be understood that the positive abnormal difference is the difference between the construction index and the maximum reference value. For example, when the construction index is 12.43 mm and the maximum reference value is 11.3 mm, the corresponding positive abnormal difference is 12.43-11.3=1.13 mm.

[0238] Among them, the positive anomaly ratio is the ratio of the positive anomaly difference to the maximum benchmark value. For example, when the positive anomaly difference is 1.13mm and the maximum benchmark value is 11.3mm, the positive anomaly ratio is 0.2, and the positive anomaly time is the detection time of the indicator constructed corresponding to the positive anomaly ratio, such as March 22.

[0239] Through the above-mentioned implementation, the present invention can determine the corresponding positive abnormality ratio and the corresponding positive abnormality time, so as to subsequently generate the corresponding heart prediction change map.

[0240] S406: When it is determined that the construction indicator is within the standard indicator range, the detection time corresponding to the construction indicator is used as the standard time, and a preset standard proportion is configured for the standard time.

[0241] It can be understood that when the construction indicator is in the standard indicator range, the standard time is the detection time corresponding to the construction indicator, and the preset standard proportion is the proportion value of the standard configured according to the standard time, which can be manually preset.

[0242] S407, call the preset base plate image and the preset change coordinate system, obtain the center point of the preset base plate image, align the origin of the preset change coordinate system with the center point, the horizontal axis of the preset change coordinate system has multiple preset time points, and the vertical axis of the preset change coordinate system has multiple preset proportions.

[0243] It can be understood that the preset base image is a preset blank base image, which is used to display the negative abnormality ratio, positive abnormality ratio and preset standard ratio corresponding to different detection times to generate a blank interface of the heart prediction change map, and the preset change coordinate system is a preset coordinate system, which is used to determine the coordinate points of the corresponding negative abnormality ratio, positive abnormality ratio and preset standard ratio, so as to obtain the heart prediction change map.

[0244] Among them, the preset time point is the corresponding time coordinate point on the horizontal axis, which can be one-to-one corresponding to the detection time, and the preset proportion is the proportion coordinate point on the vertical axis, so as to determine the coordinate position of the positive abnormal proportion and the negative abnormal proportion.

[0245] S408: Determine a negative trend point in the preset change coordinate system according to the negative abnormality time and the negative abnormality ratio.

[0246] It can be understood that the negative trend point is a position point determined in the preset change coordinate system based on the negative abnormal time and the negative abnormal proportion, that is, the horizontal coordinate of the negative trend point corresponds to the negative abnormal time, and the vertical coordinate corresponds to the position point of the negative abnormal proportion.

[0247] S409: Determine a positive trend point in the preset change coordinate system based on the positive abnormality time and the positive abnormality ratio.

[0248] It can be understood that the positive trend point is a position point determined in the preset changing coordinate system based on the positive abnormality time and the positive abnormality ratio, that is, the horizontal coordinate of the positive trend point corresponds to the positive abnormality time, and the vertical coordinate corresponds to the position point of the positive abnormality ratio.

[0249] S410: Determine a standard trend point in the preset change coordinate system according to the standard time and the preset standard proportion.

[0250] It can be understood that the standard trend point is a position point determined in a preset changing coordinate system based on the standard time and the preset standard proportion, that is, the horizontal coordinate of the standard trend point corresponds to the standard time, and the vertical coordinate corresponds to the position point of the preset standard proportion.

[0251] S411, sorting the detection times in the abnormal sequence table in ascending order to obtain a time connection sequence, connecting the negative trend point, the positive trend point and the standard trend point in sequence based on the order of the detection times in the time connection sequence, generating a heart prediction change diagram and sending it to the detection end.

[0252] It should be noted that in order to connect the corresponding position points according to the order of actual detection time and thus generate the corresponding predicted heart change diagram, the detection times in the abnormal sequence table can be sorted in ascending order so that they can be connected according to the time series to obtain the predicted heart change diagram.

[0253] It can be understood that the time connection sequence is a time sequence in which the detection times in the abnormal sequence table are sorted in ascending order, so that the negative trend points, positive trend points and standard trend points corresponding to the detection time can be connected in sequence according to the time connection sequence to obtain a heart prediction change diagram and send it to the detection end for intuitive viewing by personnel.

[0254] For example, Figure 3 As shown, when the time connection sequence is (3.4, 3.22, 4.8, 4.25, 5.12), the corresponding trend points can be connected according to the time connection sequence to obtain the heart prediction change diagram of the corresponding dynamic detection item.

[0255] It is not difficult to understand that different dynamic detection items have corresponding predicted heart change graphs, so that detection and diagnosis can be performed based on the predicted heart change graphs corresponding to the dynamic detection items, so as to improve the diagnosis efficiency and improve the accuracy of the diagnosis results.

[0256] See also Figure 4, is a schematic structural diagram of a congenital heart disease progression assessment system based on ultrasound data provided by an embodiment of the present invention. The congenital heart disease progression assessment system based on ultrasound data includes:

[0257] The generation module is used to obtain a dynamic checklist of target detection personnel corresponding to the detection time, retrieve the attribute information of the target detection personnel, and generate an initial time series diagnosis tree based on the attribute information and the dynamic checklist.

[0258] The filling module is used to retrieve the ultrasound data intercepted by the physician, obtain the annotation information and interception time of the ultrasound data, and fill the corresponding ultrasound data into the dynamic inspection table corresponding to the inspection node in the initial time series diagnosis tree based on the annotation information and the interception time to obtain the ultrasound time series diagnosis tree.

[0259] The splicing module is used to determine the abnormal items in the ultrasound time series diagnosis tree, and splice the dynamic inspection tables in the ultrasound time series diagnosis tree based on the abnormal items to obtain an abnormal sequence table corresponding to the abnormal items.

[0260] The sending module is used to generate a heart prediction change map based on the trend prediction strategy and the ultrasound data in the abnormal sequence table and send it to the detection end.

[0261] See also Figure 5 , is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention, the electronic device 50 includes: a processor 51, a memory 52 and a computer program; wherein

[0262] The memory 52 is used to store the computer program, which may also be a flash memory. The computer program is, for example, an application program or a functional module for implementing the above method.

[0263] The processor 51 is configured to execute the computer program stored in the memory to implement the various steps performed by the device in the above method. For details, please refer to the relevant description in the above method embodiment.

[0264] Optionally, the memory 52 may be independent or integrated with the processor 51 .

[0265] When the memory 52 is a device independent of the processor 51, the device may further include:

[0266] The bus 53 is used to connect the memory 52 and the processor 51 .

[0267] The present invention also provides a readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the methods provided in the various embodiments described above.

[0268] Among them, the readable storage medium can be a computer storage medium or a communication medium. Communication media include any medium that facilitates the transmission of computer programs from one place to another. Computer storage media can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application-specific integrated circuit (ASIC). In addition, the ASIC can be located in a user device. Of course, the processor and the readable storage medium can also exist in a communication device as discrete components. The readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0269] The present invention also provides a program product, which includes execution instructions stored in a readable storage medium. At least one processor of a device can read the execution instructions from the readable storage medium, and at least one processor executes the execution instructions so that the device implements the methods provided in the various embodiments described above.

[0270] In the embodiments of the above-mentioned devices, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly implemented by a hardware processor or implemented by a combination of hardware and software modules in the processor.

[0271] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating the progression of congenital heart disease based on ultrasound data, characterized in that: include: Obtaining a dynamic checklist of target detection personnel corresponding to detection times, retrieving attribute information of the target detection personnel, and generating an initial time series diagnosis tree based on the attribute information and the dynamic checklist; Retrieving ultrasound data intercepted by the physician, obtaining annotation information and interception time of the ultrasound data, and filling the corresponding ultrasound data into a dynamic inspection table corresponding to an inspection node in the initial time series diagnostic tree based on the annotation information and the interception time, thereby obtaining an ultrasound time series diagnostic tree; Determine an abnormal item in the ultrasound time series diagnosis tree, and perform splicing processing on each of the dynamic inspection tables in the ultrasound time series diagnosis tree based on the abnormal item to obtain an abnormal sequence table corresponding to the abnormal item; According to the trend prediction strategy and the ultrasound data in the abnormal sequence table, a heart prediction change map is generated and sent to the detection end, including: Obtaining the medical indicators of the ultrasound data corresponding to each detection time in the abnormal sequence table as construction indicators, and the standard indicator interval corresponding to each detection time; When it is determined that the construction index is less than the standard index interval, the minimum value of the standard index interval is retrieved as the minimum reference value; Obtain a negative anomaly difference based on the difference between the construction indicator and the minimum benchmark value, obtain a negative anomaly proportion based on the ratio of the negative anomaly difference to the minimum benchmark value, and obtain a detection time of the construction indicator corresponding to the negative anomaly proportion as the negative anomaly time; When it is determined that the construction index is greater than the standard index interval, the maximum value of the standard index interval is retrieved as the maximum reference value; Obtaining a positive anomaly difference based on a difference between the construction indicator and the maximum reference value, obtaining a positive anomaly proportion based on a ratio of the positive anomaly difference to the maximum reference value, and obtaining a detection time of the construction indicator corresponding to the positive anomaly proportion as a positive anomaly time; When it is determined that the construction indicator is within the standard indicator range, the detection time corresponding to the construction indicator is used as the standard time, and a preset standard ratio is configured for the standard time; Retrieving a preset baseplate image and a preset change coordinate system, obtaining a center point of the preset baseplate image, aligning an origin of the preset change coordinate system with the center point, wherein the horizontal axis of the preset change coordinate system has a plurality of preset time points, and the vertical axis of the preset change coordinate system has a plurality of preset proportions; Determine a negative trend point in the preset change coordinate system according to the negative abnormality time and the negative abnormality ratio; Determining a positive trend point in the preset change coordinate system based on the positive anomaly time and the positive anomaly ratio; Determine a standard trend point in the preset change coordinate system according to the standard time and the preset standard proportion; Sort the detection times in the abnormal sequence table in ascending order to obtain a time connection sequence, and connect the negative trend point, the positive trend point and the standard trend point in sequence based on the order of the detection times in the time connection sequence to generate a heart prediction change map and send it to the detection end.

2. The method according to claim 1, characterized in that The step of obtaining a dynamic checklist of target detection personnel corresponding to detection times, retrieving attribute information of the target detection personnel, and generating an initial time series diagnosis tree based on the attribute information and the dynamic checklist includes: Obtaining a dynamic inspection table corresponding to the inspection time of the target inspection person, and retrieving attribute information of the target inspection person, wherein the dynamic inspection table includes an ultrasonic filling area and a coefficient evaluation area corresponding to a plurality of dynamic inspection items; Constructing a person node according to the attribute information, extracting the month information and date information in the detection time, constructing a month summary node based on the month information, and connecting the month summary node to the person node; Constructing a date subdivision node based on the date information, connecting the date subdivision node with the corresponding month summary node, and constructing a check node connected to the date subdivision node; The dynamic check table is bound to the corresponding check node according to the detection time to obtain an initial timing diagnosis tree.

3. The method according to claim 2, characterized in that The ultrasound data intercepted by the physician end is retrieved, annotation information and interception time of the ultrasound data are obtained, and the corresponding ultrasound data is filled into a dynamic inspection table corresponding to the inspection node in the initial time series diagnosis tree based on the annotation information and the interception time, to obtain the ultrasound time series diagnosis tree, including: Retrieving ultrasound data intercepted by the physician, and obtaining annotation information and interception time of the ultrasound data; Determine the corresponding month summary node as the month update node and the date subdivision node as the date update node based on the interception time, and use the check node connected to the date update node as the node to be updated; Determine a dynamic check table corresponding to the node to be updated as a dynamic update table, and locate a dynamic check item in the dynamic update table as an update item based on the annotation information; The ultrasound data is filled into the ultrasound filling area corresponding to the update item, and the evaluation coefficient of the ultrasound data by the physician is received, and the evaluation coefficient is filled into the coefficient evaluation area corresponding to the update item to obtain an ultrasound time series diagnosis tree.

4. The method according to claim 3, characterized in that Also includes: When it is determined that the dynamic examination item corresponding to the dynamic examination table in the ultrasound time series diagnosis tree does not have an evaluation coefficient, the corresponding dynamic examination item is used as an item to be supplemented; Obtaining date information of the item to be completed that does not have an evaluation coefficient as the missing date, and using the remaining date information corresponding to the item to be completed as the base date, and retrieving the evaluation coefficient of the item to be completed corresponding to the base date as the base coefficient; According to the missing date, the base date and the base coefficient, a filling coefficient of the to-be-filled item corresponding to the missing date is obtained.

5. The method according to claim 4, characterized in that Obtaining the filling coefficient of the item to be filled corresponding to the missing date according to the missing date, the base date, and the base coefficient includes: Sort the missing dates and the reference dates in ascending order to obtain a sequence of completion dates corresponding to the items to be completed; sequentially counting adjacent missing dates in the complemented date sequence to obtain multiple missing date sequences; When it is determined that only one side of the missing date sequence has a reference date, the corresponding missing date sequence is used as a trend missing sequence, and the reference date is used as a trend completion date; Obtaining a reference coefficient corresponding to the trend filling date as a trend filling coefficient, and determining a filling coefficient for an item to be filled corresponding to a missing date in the trend missing sequence according to the trend filling coefficient and the trend filling date; When it is determined that both sides of the missing date sequence are reference dates, the corresponding missing date sequence is used as the missing sequence at both ends, and the corresponding reference date is used as the date for filling in both ends; A reference coefficient corresponding to the two-end filling date is obtained as the two-end filling coefficient, and a filling coefficient of the item to be filled corresponding to the missing date in the two-end missing sequence is determined according to the two-end filling coefficient and the two-end filling date.

6. The method according to claim 5, characterized in that The determining, based on the trend filling coefficient and the trend filling date, of the item to be filled corresponding to the missing date in the trend missing sequence includes: Obtaining two adjacent trend filling dates in the filling date sequence as trend calculation dates, and retrieving trend filling coefficients corresponding to the trend calculation dates as trend calculation coefficients; Calculate the absolute value of the trend date difference based on the trend to obtain the trend date difference, and calculate the absolute value of the trend coefficient difference based on the trend to obtain the trend coefficient difference; Based on the ratio of the trend coefficient difference and the trend date difference, the trend change rate corresponding to the two adjacent trend completion dates is obtained, and the average value of the trend change rate is obtained to obtain the average change rate; When it is determined that the trend missing sequence is at the end of the supplementary date sequence, a trend supplementary date adjacent to the trend missing sequence in the supplementary date sequence is obtained as a starting prediction date, and a trend supplementary coefficient corresponding to the starting prediction date is retrieved as a starting prediction coefficient; Obtaining the missing date in the trend missing sequence, obtaining a predicted date difference according to the difference between the missing date and the starting predicted date, and obtaining a prediction coefficient difference according to the product of the predicted date difference and the average change rate; Obtaining a filling coefficient for the item to be filled corresponding to the missing date according to the sum of the starting prediction coefficient and the difference between the prediction coefficients; When it is determined that the trend missing sequence is located at the front end of the supplementary date sequence, a trend supplementary date adjacent to the trend missing sequence in the supplementary date sequence is obtained as the end prediction date, and a trend supplementary coefficient corresponding to the end prediction date is retrieved as the end prediction coefficient; Obtaining the missing date in the trend missing sequence, obtaining a predicted date difference according to the difference between the last predicted date and the missing date, and obtaining a prediction coefficient difference according to the product of the predicted date difference and the average change rate; According to the difference between the final prediction coefficient and the prediction coefficient difference, the filling coefficient of the item to be filled corresponding to the missing date is obtained.

7. The method according to claim 5, characterized in that The determining, based on the two-end filling coefficient and the two-end filling date, of the item to be filled corresponding to the missing date in the two-end missing sequence includes: Obtaining a sum of coefficients at both ends based on the sum of the padded coefficients at both ends, and obtaining a sum of dates at both ends based on the sum of the padded dates at both ends; Obtain the missing date in the sequence with missing dates at both ends, and obtain the date proportion based on the ratio of the missing date to the sum of the dates at both ends; Based on the date proportion and the product of the coefficient and value at both ends, the filling coefficient of the item to be filled corresponding to the missing date in the missing sequence at both ends is obtained.

8. The method according to claim 3, characterized in that The determining of abnormal items in the ultrasound time series diagnostic tree, and splicing the dynamic checklists in the ultrasound time series diagnostic tree based on the abnormal items to obtain an abnormal sequence list corresponding to the abnormal items, includes: Obtaining medical indicators of ultrasound data corresponding to each dynamic examination item in the ultrasound time series diagnosis tree, wherein the medical indicators include at least inner diameter, area, and length; Retrieving the standard index interval of the dynamic examination item corresponding to each date information in the ultrasound time series diagnosis tree; When it is determined that the medical indicator is not within the standard indicator range, treating the corresponding dynamic examination item as an abnormal item; Based on the abnormal items, the ultrasonic filling areas in each of the dynamic inspection tables are intercepted as areas to be spliced, and the areas to be spliced ​​are sequentially spliced ​​according to the detection time to obtain an abnormal sequence table corresponding to the abnormal items.

9. A congenital heart disease progression assessment system based on ultrasound data, characterized in that: include: A generation module is used to obtain a dynamic checklist of target detection personnel corresponding to detection time, retrieve attribute information of the target detection personnel, and generate an initial time series diagnosis tree based on the attribute information and the dynamic checklist; a filling module, configured to retrieve ultrasound data intercepted by the physician, obtain annotation information and interception time of the ultrasound data, and fill the corresponding ultrasound data into a dynamic inspection table corresponding to an inspection node in the initial time series diagnostic tree based on the annotation information and the interception time, thereby obtaining an ultrasound time series diagnostic tree; a splicing module, configured to determine abnormal items in the ultrasound time series diagnosis tree, and splice the dynamic checklists in the ultrasound time series diagnosis tree based on the abnormal items to obtain an abnormal sequence list corresponding to the abnormal items; A sending module, configured to generate a heart prediction change map based on the trend prediction strategy and the ultrasound data in the abnormal sequence table and send the map to the detection end, comprising: Obtaining the medical indicators of the ultrasound data corresponding to each detection time in the abnormal sequence table as construction indicators, and the standard indicator interval corresponding to each detection time; When it is determined that the construction index is less than the standard index interval, the minimum value of the standard index interval is retrieved as the minimum reference value; Obtain a negative anomaly difference based on the difference between the construction indicator and the minimum benchmark value, obtain a negative anomaly proportion based on the ratio of the negative anomaly difference to the minimum benchmark value, and obtain a detection time of the construction indicator corresponding to the negative anomaly proportion as the negative anomaly time; When it is determined that the construction index is greater than the standard index interval, the maximum value of the standard index interval is retrieved as the maximum reference value; Obtaining a positive anomaly difference based on a difference between the construction indicator and the maximum reference value, obtaining a positive anomaly proportion based on a ratio of the positive anomaly difference to the maximum reference value, and obtaining a detection time of the construction indicator corresponding to the positive anomaly proportion as a positive anomaly time; When it is determined that the construction indicator is within the standard indicator range, the detection time corresponding to the construction indicator is used as the standard time, and a preset standard ratio is configured for the standard time; Retrieving a preset baseplate image and a preset change coordinate system, obtaining a center point of the preset baseplate image, aligning an origin of the preset change coordinate system with the center point, wherein the horizontal axis of the preset change coordinate system has a plurality of preset time points, and the vertical axis of the preset change coordinate system has a plurality of preset proportions; Determine a negative trend point in the preset change coordinate system according to the negative abnormality time and the negative abnormality ratio; Determining a positive trend point in the preset change coordinate system based on the positive anomaly time and the positive anomaly ratio; Determine a standard trend point in the preset change coordinate system according to the standard time and the preset standard proportion; Sort the detection times in the abnormal sequence table in ascending order to obtain a time connection sequence, and connect the negative trend point, the positive trend point and the standard trend point in sequence based on the order of the detection times in the time connection sequence to generate a heart prediction change map and send it to the detection end.

Images