Simulation monitoring interface adaptive display method and system for medical teaching

Abstract

The invention discloses a simulation monitoring interface adaptive display method and system for medical teaching, and the system specifically comprises a keyword extraction module which obtains a case and extracts keywords related to the condition of a characterization patient; the case updating module is used for acquiring medical intervention measures and updating medical data in cases; the monitoring interface prediction model comprises an initial monitoring interface prediction module and a real-time monitoring interface prediction module which are used for respectively obtaining an initial monitoring interface prediction result and a real-time monitoring interface prediction result; and the user display module is used for matching the prediction result of the monitoring interface prediction module from a pre-constructed monitoring interface standard library, and enabling the data contained in the case to correspond to the prediction result to be rendered and displayed. Through the pre-trained monitoring interface prediction model, real-time matching of the monitoring interface on each time node after the case initial state and the medical intervention measures is realized, and the scheme is flexible.

Description

Technical Field

[0001] The present application relates to the field of medical simulation, and in particular to a method and system for adaptively displaying a simulated monitoring interface for medical teaching. Background Art

[0002] In the field of medical simulation, simulation monitoring equipment is mainly used to synchronously and continuously monitor the ECG, blood pressure, respiration, body temperature and other vital signs of simulated patients. It can reflect the changes in the condition and provide a good teaching method for medical staff to comprehensively, intuitively and timely understand the condition.

[0003] In the process of medical simulation, there may be some restrictions, such as funding or venue, as well as some teaching task requirements, which do not support the setting of multiple simulation monitoring devices to display various physiological parameters of vital signs. Therefore, how to dynamically match the corresponding simulation monitoring interface of each node according to different cases is the research direction of this application. Summary of the invention

[0004] In order to solve the above problems, the present application provides a method for adaptively displaying a simulated monitoring interface for medical teaching, comprising the following steps: S1, obtaining the case to be displayed and extracting keywords related to characterizing the patient's condition; S2, inputting the keyword into a pre-trained monitoring interface prediction model to obtain an initial monitoring interface prediction result; S3, matching an initial monitoring interface from a pre-built monitoring interface standard library according to the initial monitoring interface prediction result, and rendering and displaying the data contained in the case on the initial monitoring interface; S4, real-time monitoring of medical intervention measures, updating case data corresponding to the medical intervention measures in the case, inputting the updated case into a pre-trained monitoring interface prediction model, and obtaining a real-time monitoring interface prediction result; S5, judging whether the prediction result of the real-time monitoring interface is consistent with the current monitoring interface, If yes, the updated case data is rendered and displayed on the current monitoring interface; No, a real-time monitoring interface rendering display is matched from the monitoring interface standard library.

[0005] Furthermore, the case to be displayed is generated from the current case based on the standard healthy SP model, and the specific steps include: S101, constructing a standard healthy SP model library, wherein the standard healthy SP model library is composed of a number of normal standardized patient information; S102, classifying the normal standardized patient information in the standard healthy SP model library according to gender, age, height, and weight; S103, selecting the standardized patient information with the highest similarity from each category as the reference standardized patient information of each category; S104, determining the category to which the current case belongs according to gender, age, height, and weight, and comparing it with the benchmark standardized patient information corresponding to the category; S105, determining whether the vital sign categories included in the current case include all the vital sign categories included in the benchmark standardized patient information corresponding to the category to which it belongs, Yes, confirm that the current case is the case to be displayed; If no, determine the missing vital sign categories in the current case, and add the corresponding vital sign category data in the benchmark standardized patient information corresponding to the category to the current case to generate a case to be displayed.

[0006] Furthermore, a classification threshold interval is set for at least one of the age, the height, and the weight.

[0007] Further, the pre-trained monitoring interface prediction model is generated according to the following steps, including: S201, from historical cases, obtain basic information, historical vital signs data, historical physiological waveform data, and historical medical intervention measures of each patient, and construct a category set {C1, C2, C3, C4} and a data set T={Xi,i=1,2,…n}, where C1, C2, C3, C4 correspond to basic information, vital signs data, physiological waveform data, and medical intervention measures, respectively, Xi is the i-th historical case data, n is the number of historical case data, and each historical case data in the data set corresponds to one or more category labels; S202, adding a monitoring interface type label to the initial status of each historical case and the time node of each medical intervention measure; S203, extracting keywords related to characterizing the patient's condition from the basic information, taking the keywords as input and the corresponding monitoring interface type labels on the initial state nodes as output results for training, and obtaining an initial monitoring interface prediction model; S204, taking the medical intervention measures as input, combining the category labels corresponding to the current historical cases, and the vital signs data and physiological waveform data corresponding to the current medical intervention measures, and taking the monitoring interface type labels corresponding to the current medical intervention measures nodes as the output results for training, a real-time monitoring interface prediction model is obtained.

[0008] Furthermore, step S202 also includes: calculating the relevance of the monitoring interface type labels at each time node of the historical cases, and weighting and sorting the monitoring interface type labels according to the relevance.

[0009] The present application also provides a medical teaching simulation monitoring interface adaptive display system, which is applied to the above display method, including a keyword extraction module, a case update module, a monitoring interface prediction model, and a user display module. A keyword extraction module is used to obtain cases and extract keywords related to characterizing the patient's condition from the cases; A case update module, used to obtain medical intervention measure data and update the medical data corresponding to the medical intervention measure in the case; The monitoring interface prediction model includes an initial monitoring interface prediction module and a real-time monitoring interface prediction module, wherein the initial monitoring interface prediction module is used to obtain an initial monitoring interface prediction result according to the keyword, and the real-time monitoring interface prediction module is used to obtain a real-time monitoring interface prediction result according to the updated medical data; The user display module is used to match the prediction results of the monitoring interface prediction module from the pre-built monitoring interface standard library, and render and display the data contained in the case corresponding to the prediction results. The beneficial effects of this application are: The present application uses a pre-trained monitoring interface prediction model to train the monitoring interface types at various time nodes of historical cases, thereby realizing the prediction of the monitoring interface at various time nodes from the initial state of the case to after medical intervention measures during the teaching simulation process. By calculating the correlation of the monitoring interface types at each time node, the monitoring interface types can be weighted and ranked, and dynamic switching of the monitoring interface can be achieved according to the key factors of the case. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a flow chart of the adaptive display method of the simulated monitoring interface for medical teaching according to the embodiment of the present application.

[0011] Figure 2 It is a schematic diagram of the steps for generating the case to be displayed in this application.

[0012] Figure 3 It is a training diagram of the monitoring interface prediction model pre-trained in this application.

[0013] Figure 4 It is a schematic diagram of the steps of generating a pre-trained monitoring interface prediction model in this application.

[0014] Figure 5 It is a schematic diagram of the steps of weight ranking calculation of the monitoring interface in another embodiment of the present application.

[0015] Figure 6 It is a structural schematic diagram of a simulated monitoring interface adaptive display system for medical teaching according to an embodiment of the present application. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be clearly and completely described below in conjunction with the drawings of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the protection scope of the present application.

[0017] like Figure 1 , Figure 2 As shown, the embodiment of the present application provides an adaptive display method for a simulated monitoring interface for medical teaching. The method can simulate vital signs throughout the entire cycle of case diagnosis and treatment, and dynamically match the currently required monitoring interface in real time according to the development of the disease. The method specifically includes the following steps: S1, obtaining the case to be displayed and extracting keywords related to characterizing the patient's condition; In this step, most of the cases to be displayed are from clinical practice, and some are written by experienced medical workers, teachers, etc. to provide simulations of different cases to diversify teaching scenarios and tasks. Usually, the cases include the patient's sociodemographic information, chief complaint, current medical history, examination records, disease diagnosis and treatment, etc. Therefore, the case is a complex structure carrier in multiple formats such as graphics, text, and tables, and the monitoring interface is a real-time monitoring and early warning of important data of the patient's vital signs such as electrocardiogram signals, heart rate, blood oxygen saturation, blood pressure, respiratory rate, temperature, etc. throughout the cycle. During the entire cycle, depending on the initial state and changes in vital signs, the monitoring interface may be adjusted as the disease progresses. Different cases have different patient conditions. By extracting keywords related to the characterization of the patient's condition, the monitor required for the initial state of the case can be determined, thereby determining the monitoring interface.

[0018] In order to make the case content to be displayed more complete and meet the display requirements of different monitoring interfaces, in some embodiments, the case to be displayed is generated by the current case based on the standard health SP model, and the specific steps include: S101, build a standard healthy SP model library: In clinical practice, standardized patients (SP for short) refer to normal people or patients who have undergone standardized and systematic training and accurately represent the actual clinical problems of patients. In this step, several normal standardized patients' information is selected to build a standard healthy SP model library; S102, classify the standardized patient information in the standard health SP model library according to basic elements such as gender, age, height, and weight; under normal circumstances, age, height, and weight have different effects on human vital signs. For example, as age increases, the normal range of blood pressure also changes. Therefore, in this step, age, height, and weight can also be set to a classification threshold interval respectively. The standardized patient information of a certain age range can be set as one category, and the height and weight can be set accordingly; S103, selecting standardized patient information with the highest similarity from each category as the reference standardized patient information of each category; S104, determining the category to which the current case belongs according to gender, age, height, and weight, and comparing it with the benchmark standardized patient information corresponding to the category; S105, determining whether the vital sign categories included in the current case include all the vital sign categories included in the benchmark standardized patient information corresponding to the category to which it belongs, Yes, confirm that the current case is the case to be displayed; If no, determine the missing vital sign category in the current case, and add the corresponding vital sign category data in the benchmark standardized patient information corresponding to the category to the current case to generate a case to be displayed. S2, input the keyword into the pre-trained monitoring interface prediction model to obtain the initial monitoring interface prediction result; S3, matching an initial monitoring interface from a pre-built monitoring interface standard library according to the initial monitoring interface prediction result, and rendering and displaying the data contained in the case on the initial monitoring interface; Different monitors have different interfaces and monitor different vital signs data. In order to be closer to real monitors, most of the monitoring interfaces in the monitoring interface standard library are derived from real monitor interfaces. The interface can also be designed according to teaching needs.

[0019] S4, real-time monitoring of medical intervention measures, updating case data corresponding to the medical intervention measures in the case, inputting the updated case into a pre-trained monitoring interface prediction model, and obtaining a real-time monitoring interface prediction result; During the case simulation process, from the initial state, after medical intervention measures such as diagnosis and treatment, the vital signs data and physiological waveform data undergo corresponding changes, causing the condition to change accordingly. Through this step, the changes in the condition during the entire diagnosis and treatment process can be monitored in real time, and the monitoring interface required after the current medical intervention measures can be predicted.

[0020] S5, judging whether the prediction result of the real-time monitoring interface is consistent with the current monitoring interface, If yes, the updated case data is rendered and displayed on the current monitoring interface; No, a real-time monitoring interface rendering display is matched from the monitoring interface standard library.

[0021] like Figure 3 , Figure 4 As shown, the pre-trained monitoring interface prediction model is generated by the following steps, including: S201, obtain each patient's basic information, historical vital signs data, historical physiological waveform data, and historical medical intervention measures from historical cases, and construct a category set {C1, C2, C3, C4} and a data set T={Xi,i=1,2,…n}, where C1, C2, C3, C4 correspond to basic information, vital signs data, physiological waveform data, and medical intervention measures, respectively, Xi is the i-th historical case data, n is the number of historical case data, and each historical case data in the data set corresponds to one or more category labels.

[0022] In this step, the basic information mainly includes text information such as the patient's name, age, height, weight, and chief complaint. The vital signs data and physiological waveform data are graphic data that represent the patient's vital signs waveform. The medical intervention measures are mainly physician diagnosis and treatment measures, including but not limited to medication, manual or instrumental care, examination, surgery and other treatment operations. In this step, by classifying each data item, establishing a category set and a data set, and corresponding labels are given respectively, so as to bind each patient's initial state and the vital signs information in the time dimension after the medical intervention measures.

[0023] S202, adding a monitoring interface type label to the initial status of each historical case and the time node of each medical intervention measure.

[0024] This step refers to the vital signs information corresponding to the real monitor display interface, combines the corresponding vital signs data and physiological waveform data at each time node to add the corresponding monitoring interface type label, and constructs a monitoring interface training set to provide basic support for predicting the monitoring interface in subsequent steps.

[0025] S203, extracting keywords related to characterizing the patient's condition from the basic information, taking the keywords as input and the corresponding monitoring interface type labels on the initial state nodes as output results for training, and obtaining an initial monitoring interface prediction model.

[0026] S204, taking the medical intervention measures as input, combining the category labels corresponding to the current historical cases, and the vital signs data and physiological waveform data corresponding to the current medical intervention measures, and taking the monitoring interface type labels corresponding to the current medical intervention measures nodes as the output results for training, a real-time monitoring interface prediction model is obtained.

[0027] In this embodiment, step S202 further includes: calculating the relevance of the monitoring interface type labels at each time node of the historical cases, and weighting and sorting the monitoring interface type labels according to the relevance.

[0028] In some cases, it may be necessary to monitor multiple vital signs simultaneously at a certain time node. Therefore, multiple monitoring interface type labels may be added to the node. In order to facilitate subsequent display, this step calculates the relevance of each category of monitoring interfaces to the condition, and weights the monitoring interfaces according to the relevance. Based on this, before displaying the predicted monitoring interface on the user display interface, the user can select a monitoring interface with a high weight to display, or set a display threshold, select monitoring interfaces within a certain threshold range, and set display areas on the user display interface to display them simultaneously.

[0029] The following is a case of an elderly pregnant woman with grade 2 hypertension who was admitted to the hospital and was about to give birth. Figure 5 As shown in the figure, after the monitoring interface prediction model is calculated, two monitoring interface prediction results are obtained, namely CTG equipment (fetal heart monitoring equipment) and ECG monitor. By calculating the relevance of the two monitoring equipment to the condition, weight calculation is performed, such as calculating the two key factors of hypertension and childbirth, that is, when childbirth is the key factor, the fetal heart monitoring equipment performs the final stage of fetal heart and uterine contraction waveform analysis, and the weight accounts for 87%; but the patient also suffers from severe hypertension and needs to keep blood pressure detection. At this time, the weight of the ECG monitor accounts for 64%. Finally, dynamic analysis and switching are performed according to the current patient status. For example, the current patient is admitted to the hospital voluntarily. At this time, the patient is in good consciousness and normal condition. The prediction model will increase the weight of the CTG device to 92%. At this time, CTG is in normal monitoring of fetal heart and uterine contraction status. If the current patient is admitted to the hospital through the emergency department, has blurred consciousness, and is waiting for full-term delivery, the prediction model will believe that the current patient needs real-time monitoring of vital signs parameters, the weight of the CTG device will be reduced to 73%, and the weight of the ECG monitor will be increased to 96%.

[0030] like Figure 6 As shown, this embodiment also provides a medical teaching simulation monitoring interface adaptive display system, including: a keyword extraction module, a case update module, a monitoring interface prediction module, and a user display module: A keyword extraction module is used to obtain case data and extract keywords related to characterizing the patient's condition; A case update module is used to obtain medical intervention measures data and update the medical data corresponding to the medical intervention measures in the case; The monitoring interface prediction module includes an initial monitoring interface prediction module and a real-time monitoring interface prediction module. The initial monitoring interface prediction module is used to obtain the initial monitoring interface prediction result according to the keywords, and the real-time monitoring interface prediction module is used to obtain the real-time monitoring interface prediction result according to the updated medical data; The user display module is used to match the prediction results of the monitoring interface prediction module from the pre-built monitoring interface standard library, and render and display the data contained in the case corresponding to the prediction results.

Claims

1. A method for adaptively displaying a monitoring simulation interface for medical teaching, characterized in that: The following steps are involved: S1, obtaining the case to be displayed and extracting keywords related to characterizing the patient's condition; S2, inputting the keyword into a pre-trained monitoring interface prediction model to obtain an initial monitoring interface prediction result; S3, matching an initial monitoring interface from a pre-built monitoring interface standard library according to the initial monitoring interface prediction result, and rendering and displaying the data contained in the case on the initial monitoring interface; S4, real-time monitoring of medical intervention measures, updating case data corresponding to the medical intervention measures in the case, inputting the updated case into a pre-trained monitoring interface prediction model, and obtaining a real-time monitoring interface prediction result; S5, judging whether the prediction result of the real-time monitoring interface is consistent with the current monitoring interface, If yes, the updated case data is rendered and displayed on the current monitoring interface; No, a real-time monitoring interface rendering display is matched from the monitoring interface standard library.

2. The method for adaptively displaying a monitoring simulation interface for medical teaching according to claim 1, characterized in that: The case to be displayed is generated from the current case based on the standard healthy SP model, and the specific steps include: S101, constructing a standard healthy SP model library, wherein the standard healthy SP model library is composed of a number of normal standardized patient information; S102, classifying the normal standardized patient information in the standard healthy SP model library according to gender, age, height, and weight; S103, selecting the standardized patient information with the highest similarity from each category as the reference standardized patient information of each category; S104, determining the category to which the current case belongs according to gender, age, height, and weight, and comparing it with the benchmark standardized patient information corresponding to the category; S105, determining whether the vital sign categories included in the current case include all the vital sign categories included in the benchmark standardized patient information corresponding to the category to which it belongs, Yes, confirm that the current case is the case to be displayed; If no, determine the missing vital sign categories in the current case, and add the corresponding vital sign category data in the benchmark standardized patient information corresponding to the category to the current case to generate a case to be displayed.

3. The method for adaptively displaying a monitoring simulation interface for medical teaching according to claim 2, characterized in that: Step S102 also includes setting a classification threshold interval for at least one of the age, the height, and the weight.

4. The method for adaptively displaying a monitoring simulation interface for medical teaching according to claim 1, characterized in that: The following steps are used to generate a pre-trained monitoring interface prediction model, including: S201, from historical cases, obtain basic information, historical vital signs data, historical physiological waveform data, and historical medical intervention measures of each patient, and construct a category set {C1, C2, C3, C4} and a data set T={Xi,i=1,2,…n}, where C1, C2, C3, C4 correspond to basic information, vital signs data, physiological waveform data, and medical intervention measures, respectively, Xi is the i-th historical case data, n is the number of historical case data, and each historical case data in the data set corresponds to one or more category labels; S202, adding a monitoring interface type label to the initial status of each historical case and the time node of each medical intervention measure; S203, extracting keywords related to characterizing the patient's condition from the basic information, taking the keywords as input and the corresponding monitoring interface type labels on the initial state nodes as output results for training, and obtaining an initial monitoring interface prediction model; S204, taking the medical intervention measures as input, combining the category labels corresponding to the current historical cases, and the vital signs data and physiological waveform data corresponding to the current medical intervention measures, and taking the monitoring interface type labels corresponding to the current medical intervention measures nodes as the output results for training, a real-time monitoring interface prediction model is obtained.

5. The method for adaptively displaying a monitoring simulation interface for medical teaching according to claim 4, characterized in that: Step S202 also includes: calculating the relevance of the monitoring interface type labels at each time node of the historical cases, and weighting and sorting the monitoring interface type labels according to the relevance.

6. A monitoring simulation interface adaptive display system for medical teaching, applied to the display method according to any one of claims 1 to 5, characterized in that: It includes keyword extraction module, case update module, monitoring interface prediction model, and user display module. A keyword extraction module is used to obtain cases and extract keywords related to characterizing the patient's condition from the cases; A case update module, used to obtain medical intervention measure data and update the medical data corresponding to the medical intervention measure in the case; The monitoring interface prediction model includes an initial monitoring interface prediction module and a real-time monitoring interface prediction module, wherein the initial monitoring interface prediction module is used to obtain an initial monitoring interface prediction result according to the keyword, and the real-time monitoring interface prediction module is used to obtain a real-time monitoring interface prediction result according to the updated medical data; The user display module is used to match the prediction results of the monitoring interface prediction module from a pre-built monitoring interface standard library, and render and display the data contained in the case corresponding to the prediction results.

Images