Postoperative heart failure risk prediction method and system, electronic equipment and storage medium

Abstract

The invention relates to the technical field of machine learning, and particularly discloses a postoperative heart failure risk prediction method and system, electronic equipment and a storage medium, and the method comprises the steps: determining a target machine learning model for postoperative heart failure risk prediction and a plurality of target feature indexes, training the target machine learning model according to specific values of a plurality of target feature indexes of each sample patient and a sample label representing whether each sample patient has postoperative heart failure or not to obtain a trained machine learning model; and inputting specific values of a plurality of target characteristic indexes of a to-be-tested patient into the trained machine learning model to obtain a postoperative heart failure risk prediction result of the to-be-tested patient. The clinical working process can be simplified, the accuracy and efficiency of postoperative heart failure risk assessment are improved, and finally better patient monitoring and nursing are achieved.

Description

Background Art

[0002] Postoperative heart failure (referred to as "postoperative heart failure") is a common and serious complication during surgery, which may increase the patient's mortality rate and prolong the postoperative recovery time. The occurrence of heart failure is usually caused by the combined action of multiple factors, including the patient's preoperative cardiac function status, intraoperative operation conditions, and postoperative monitoring and management measures. Early identification and risk assessment of heart failure are crucial to reducing postoperative complications and improving prognosis.

[0003] Traditional methods for assessing the risk of postoperative heart failure mainly rely on doctors' experience and some statistically based scoring systems, such as the European Cardiac Surgery Risk Scoring System (EuroSCORE) and the American Society of Thoracic Surgeons Scoring System (STS score). Although these scoring systems can provide a basis for the risk of postoperative heart failure to a certain extent, their limitations are also very obvious: 1) The scale score depends to a certain extent on the doctor's subjective judgment, which may lead to differences in the score; 2) Low specificity: the scale may be misjudged, and a score that is too high cannot clearly distinguish between low-risk and high-risk states; 3) The simple total score of the scale may ignore the relationship between the scoring items, making it difficult to present the overall condition of the patient; 4) The scale score is usually a static assessment at a certain point in time and cannot dynamically reflect the changes in the patient's condition in real time.

[0004] In recent years, with the rapid growth of medical data and the continuous advancement of artificial intelligence technology, prediction models based on machine learning have received widespread attention in the medical field. Machine learning models can process large amounts of multi-source data and automatically extract potential high-dimensional features from them, which makes them show great potential in disease risk prediction. For example, through comprehensive analysis of electronic health records (EHR), laboratory test data, and intraoperative monitoring data, machine learning models can achieve more accurate personalized predictions. Therefore, it is urgent to provide a technical solution to solve the above problems. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a method, system, electronic device and storage medium for predicting the risk of postoperative heart failure.

[0006] In a first aspect, the present invention provides a method for predicting the risk of postoperative heart failure, and the technical solution of the method is as follows:

[0007] Determine a target machine learning model and multiple target feature indicators for postoperative heart failure risk prediction, and train the target machine learning model according to the specific values ​​of the multiple target feature indicators of each sample patient and the sample label characterizing whether each sample patient has postoperative heart failure, to obtain a trained machine learning model;

[0008] The specific values ​​of multiple target characteristic indicators of the patient to be tested are input into the trained machine learning model to obtain the postoperative heart failure risk prediction result of the patient to be tested.

[0009] The beneficial effects of a postoperative heart failure risk prediction method of the present invention are as follows:

[0010] The method of the present invention can simplify clinical workflow, improve the accuracy and efficiency of postoperative heart failure risk assessment, and ultimately achieve better patient monitoring and care.

[0011] Based on the above scheme, the method for predicting the risk of postoperative heart failure of the present invention can also be improved as follows.

[0012] In an optional manner, the step of determining a target machine learning model for postoperative heart failure risk prediction comprises:

[0013] Determine multiple original characteristic indicators for postoperative heart failure risk prediction from demographic characteristics, vital signs, medical history and laboratory test information;

[0014] According to the specific values ​​of multiple original feature indicators of each sample patient and the sample label, and using a five-fold cross-validation method, the performance of each preset machine learning model used for postoperative heart failure risk prediction was evaluated to obtain the performance measurement value corresponding to each preset machine learning model;

[0015] The preset machine learning model with the highest performance metric value is determined as the target machine learning model.

[0016] In an optional manner, the step of determining the target characteristic index for postoperative heart failure risk prediction includes:

[0017] According to the importance of all original feature indicators to the target machine learning model and the stability of different numbers of feature indicators among all original feature indicators to the target machine learning model, multiple target feature indicators for postoperative heart failure risk prediction are determined.

[0018] In an optional manner, the method further includes:

[0019] Using the SHAP method and based on the specific values ​​of multiple target characteristic indicators of the patient to be tested, the SHAP value that characterizes the contribution of each target characteristic indicator of the patient to be tested to the postoperative heart failure risk prediction result is calculated, and based on all the SHAP values, the SHAP force diagram of the patient to be tested is generated.

[0020] In an optional manner, the method further includes:

[0021] Based on the postoperative heart failure risk prediction result and in combination with the medical database, recommended measures for the patient to be tested are generated.

[0022] In an optional manner, the method further includes:

[0023] According to the specific values ​​of the multiple target characteristic indicators of the patient to be tested, the postoperative heart failure risk prediction result and the recommended measures, a risk assessment report for the patient to be tested is generated and visually output.

[0024] In an optional manner, the plurality of target characteristic indicators include: coronary heart disease, serum phosphorus, myocardial infarction, absolute neutrophil count, acute kidney injury, prothrombin time, pulse, acute poisoning, diastolic blood pressure and age.

[0025] In a second aspect, the present invention provides a postoperative heart failure risk prediction system, the technical solution of the system is as follows:

[0026] Includes: building module and prediction module;

[0027] The construction module is used to: determine a target machine learning model and multiple target feature indicators for postoperative heart failure risk prediction, and train the target machine learning model according to specific values ​​of the multiple target feature indicators of each sample patient and a sample label characterizing whether each sample patient has postoperative heart failure, to obtain a trained machine learning model;

[0028] The prediction module is used to input specific values ​​of multiple target characteristic indicators of the patient to be tested into the trained machine learning model to obtain a postoperative heart failure risk prediction result for the patient to be tested.

[0029] The beneficial effects of a postoperative heart failure risk prediction system of the present invention are as follows:

[0030] The system of the present invention can simplify clinical workflow, improve the accuracy and efficiency of postoperative heart failure risk assessment, and ultimately achieve better patient monitoring and care.

[0031] In a third aspect, a technical solution of an electronic device of the present invention is as follows:

[0032] The method comprises a memory, a processor and a program stored in the memory and running on the processor. When the processor executes the program, the steps of the method for predicting the risk of postoperative heart failure of the present invention are implemented.

[0033] In a fourth aspect, the present invention provides a computer-readable storage medium having the following technical solution:

[0034] Instructions are stored in the computer-readable storage medium. When the computer-readable storage medium reads the instructions, the computer-readable storage medium executes the steps of the postoperative heart failure risk prediction method of the present invention.

[0035] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings are only used to illustrate the embodiments and are not to be considered as limiting the present invention. In addition, the same reference symbols are used to represent the same components throughout the accompanying drawings. In the accompanying drawings:

[0037] Figure 1 A schematic diagram of a process of an embodiment of a method for predicting risk of postoperative heart failure according to the present invention;

[0038] Figure 2 AUC curves for eight preset machine learning models;

[0039] Figure 3 Schematic diagram of feature importance of XGBoost model;

[0040] Figure 4 The model performance diagram for different combinations of target feature indicators for the XGBoost model;

[0041] Figure 5 This is one of the schematic diagrams of the performance of the XGBoost model on the test set;

[0042] Figure 6 The second diagram is the performance of the XGBoost model on the test set;

[0043] Figure 7 This is one of the global explanation diagrams of the XGBoost model;

[0044] Figure 8 This is the second diagram of the global explanation of the XGBoost model;

[0045] Fig. 9 This is one of the SHAP diagrams of the patient to be tested;

[0046] Fig.10 The second SHAP diagram for the patient to be tested;

[0047] Fig.11 It is a structural schematic diagram of an embodiment of a postoperative heart failure risk prediction system of the present invention;

[0048] Fig.12 The figure is a schematic structural diagram of an embodiment of an electronic device of the present invention. DETAILED DESCRIPTION

[0049] Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0050] Figure 1 A flow chart of an embodiment of a method for predicting the risk of postoperative heart failure provided by the present invention is shown, and the method for predicting the risk of postoperative heart failure can be executed by an electronic device such as a terminal device or a server. The terminal device can be any fixed or mobile terminal such as a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. The server can be a single server or a server cluster composed of multiple servers. Any electronic device can implement the method for predicting the risk of postoperative heart failure by means of a processor calling computer-readable instructions stored in a memory. Figure 1 As shown, the following steps are included:

[0051] S1. Determine a target machine learning model and multiple target feature indicators for predicting the risk of postoperative heart failure, and train the target machine learning model according to the specific values ​​of the multiple target feature indicators of each sample patient and the sample label that characterizes whether each sample patient has postoperative heart failure to obtain a trained machine learning model.

[0052] Among them, the target machine learning model defaults to the XGBoost model, which can also be set according to the actual situation, and there is no restriction here. Multiple target feature indicators include: coronary heart disease (excluding coronary heart disease with myocardial infarction), serum phosphorus, myocardial infarction, absolute neutrophil value, acute kidney injury, prothrombin time, pulse, acute poisoning (alcohol, carbon monoxide, chemical poisons), diastolic blood pressure (low blood pressure) and age. The specific values ​​of the target feature indicators are all represented by the quantitative values ​​of the target feature indicators. When the sample patient has postoperative heart failure, the sample patient's sample label is 1; when the sample patient does not have postoperative heart failure, the sample patient's sample label is 0.

[0053] It should be noted that the specific value of coronary heart disease is 0 (indicating no coronary heart disease) or 1 (indicating having coronary heart disease), the specific value of myocardial infarction is 0 (indicating no myocardial infarction) or 1 (indicating having myocardial infarction), the specific value of acute kidney injury is 0 (indicating no acute kidney injury) or 1 (indicating having acute kidney injury), the specific value of acute poisoning is 0 (indicating no acute poisoning) or 1 (indicating having acute poisoning), and the specific values of the remaining target characteristic indicators (such as serum phosphorus, absolute neutrophil count, prothrombin time, pulse, and diastolic blood pressure, etc.) are determined according to the actual values.

[0054] S2. Input the specific values of multiple target characteristic indicators of the patient to be tested into the trained machine learning model to obtain the postoperative heart failure risk prediction result of the patient to be tested.

[0055] Among them, the XGBoost model is defaulted to a binary classification model, and the postoperative heart failure risk prediction result takes a value of 1 or 0. When the postoperative heart failure risk prediction result is 1, it indicates that the patient to be tested has a postoperative heart failure risk; when the postoperative heart failure risk prediction result is 0, it indicates that the patient to be tested does not have a postoperative heart failure risk. The XGBoost model can also be a regression model, and the postoperative heart failure risk prediction result takes a value of 0 - 1. Assuming the threshold is 0.6, when the postoperative heart failure risk prediction result is greater than 0.6, it indicates that the patient to be tested has a postoperative heart failure risk; when the postoperative heart failure risk prediction result is less than or equal to 0.6, it indicates that the patient to be tested does not have a postoperative heart failure risk.

[0056] In an optional manner, the steps for determining the target machine learning model for postoperative heart failure risk prediction include:

[0057] Determine multiple original characteristic indicators for postoperative heart failure risk prediction from demographic characteristic information, vital sign information, past medical history information, and laboratory test information.

[0058] Among them, the demographic characteristic information includes: basic demographic statistics such as the patient's name, gender, age, contact information, and address. The vital sign information includes: the patient's heart rate, pulse, diastolic blood pressure, systolic blood pressure, BMI, height, weight, etc. The past medical history information includes: atrial fibrillation, acute coronary syndrome, primary cardiomyopathy, pulmonary embolism, congenital heart disease, coronary heart disease, myocardial infarction, etc. The laboratory test information includes: hematological examinations (such as complete blood count), biochemical examinations, coagulation index examinations, etc.

[0059] It should be noted that demographic information, vital signs information, medical history information and laboratory test information were extracted from the hospital database through structured query language (SQL). All of the above information can be used as original feature indicators. In addition, the specific values ​​of the above original feature indicators were preprocessed, and the preprocessing methods included but were not limited to: ① Data cleaning: eliminating features with a missing value ratio of more than 30%; ② Label encoding or one-hot encoding of categorical variables; ③ Treating outliers under each feature as null values; ④ Using Impute.SimpleImputer in the preprocessing module of the python machine learning library Sklearn to fill the missing values ​​with the median.

[0060] According to the specific values ​​of multiple original feature indicators of each sample patient and the sample label, and using a five-fold cross-validation method, each preset machine learning model for postoperative heart failure risk prediction is evaluated for performance, and the performance metric value corresponding to each preset machine learning model is obtained. The preset machine learning model with the highest performance metric value is determined as the target machine learning model.

[0061] Among them, the number of preset machine learning models defaults to eight, including: Naive Bayes, K-Nearest Neighbors, Support Vector Machines, Logistic Regression, Decision Tree, AdaBoost, XGBoost and Random Forest.

[0062] Among them, 80% of the samples were randomly selected from the entire data set as the training set (36,788 samples in the training set, 292 positive cases, and 36,496 negative cases) to train the model; 20% of the samples were used as the test set (9,197 samples in the test set, 73 positive cases, and 9,124 negative cases) to evaluate the performance of the model. In order to avoid the extreme imbalance between negative and positive samples in the training set, the training set was randomly undersampled. In order to ensure the robustness of the model and prevent overfitting, the five-fold cross-validation method was used, and six model evaluation indicators were calculated: accuracy, sensitivity, specificity, precision, F1 score, and area under the curve (AUC). The model was tested using the area under the receiver operating characteristic curve (ROC) (AUC), which was divided into the following levels: 0.5-0.6 (poor), 0.6-0.7 (poor), 0.7-0.8 (better), 0.8-0.9 (good), and 0.9-1.0 (excellent). Among the eight models constructed, the random forest and XGBoost models performed best, with an accuracy of 0.984 and AUROCs of 0.983 and 0.982, respectively. The XGBoost model has a high sensitivity, and the XGBoost model was finally selected as the target machine learning model. The model evaluation indicators are shown in Table 1. The AUC curves of the eight preset machine learning models are shown in Table 1. Figure 2 shown.

[0063] Table 1:

[0064] Model Name Accuracy Sensitivity Specificity Accuracy F1 score AUROC NB 0.885 0.86 0.886 0.21 0.335 0.905 KNN 0.935 0.808 0.939 0.308 0.445 0.906 SVM 0.863 0.863 0.863 0.175 0.29 0.94 LR 0.921 0.87 0.923 0.274 0.416 0.962 DT 0.97 0.705 0.979 0.529 0.603 0.867 AdaBoost 0.969 0.781 0.976 0.518 0.622 0.974 XGBoost 0.984 0.767 0.991 0.746 0.754 0.982 RF 0.984 0.733 0.992 0.764 0.746 0.983

[0065] Among them, the prediction results of the XGBoost model are displayed and evaluated through the confusion matrix. The confusion matrix is ​​used to describe the accuracy of the prediction results of the XGBoost model. The following is a detailed explanation of the prediction results of the binary classification model: ①TN represents the number of samples predicted as negative and actually negative. ②FP represents the number of samples predicted as positive but actually negative (i.e. false positives). ③FN represents the number of samples predicted as negative but actually positive (i.e. missed negatives). ④TP represents the number of samples predicted as positive and actually positive (i.e. correct predictions).

[0066] Among them, the calculation formula for accuracy is: (TP+TN) / (TP+TN+FP+FN); the calculation formula for precision is: TP / (TP+FP); the calculation formula for sensitivity (Recall) is: TP / (TP+FN); the calculation formula for F1 score is: 2*Precision*Recall / (Precision+Recall).

[0067] It should be noted that the inclusion criteria for the sample patients in the data set are: ① age ≥ 18 years old; ② surgery during hospitalization; exclusion criteria are: ① age < 18 years old; ② patients with heart failure before surgery. A total of 45,620 negative diagnoses and 365 positive diagnoses were included in the data set.

[0068] In an optional manner, the step of determining the target characteristic index for postoperative heart failure risk prediction includes:

[0069] According to the importance of all original feature indicators to the target machine learning model and the stability of different numbers of feature indicators among all original feature indicators to the target machine learning model, multiple target feature indicators for postoperative heart failure risk prediction are determined.

[0070] Among them, the characteristics of the XGBoost model are used to obtain the importance of each original feature index to the target machine learning model. Selecting the most relevant and influential feature index from all the original feature indexes can improve the performance and interpretability of the model while saving storage and computing resources. First, the original feature indexes are sorted according to the importance of the XGBoost model, such as Figure 3 As shown. Then, the least important features are discarded one by one, and the performance of the model is evaluated using the AUROC of five-fold cross-validation to find the best feature combination. Figure 4 As shown in the figure, the performance of the XGBoost model tends to be stable after ten feature indicators. The target feature indicators finally screened out include: coronary heart disease (excluding coronary heart disease with myocardial infarction), serum phosphorus, myocardial infarction, absolute neutrophil count, acute kidney injury, prothrombin time, pulse, acute poisoning (alcohol, carbon monoxide, chemical poisons), diastolic blood pressure (low blood pressure), and age.

[0071] It should be noted that the test set data consists of 9197 cases (sample patients), including 9124 negative cases and 73 positive cases. The verification results of the XGBoost model on the test set are shown in Table 2. Figure 5 and Figure 6 As shown in the figure, the XGBoost model performed well on the test set, with an accuracy of 0.987 and an AUROC of 0.975. Among the 9124 negative samples, 9017 were predicted correctly; among the 73 positive samples, 62 were predicted correctly.

[0072] In an optional manner, the method further includes:

[0073] Using the SHAP method and based on the specific values ​​of multiple target characteristic indicators of the patient to be tested, the SHAP value that characterizes the contribution of each target characteristic indicator of the patient to be tested to the postoperative heart failure risk prediction result is calculated, and based on all the SHAP values, the SHAP force diagram of the patient to be tested is generated.

[0074] Among them, SHAP (SHapley Additive exPlanations) explains the machine learning model by quantifying the contribution of each target feature indicator to the prediction results of postoperative heart failure risk. The main advantage of SHAP is that it provides global and local interpretability. The global explanation can highlight the most influential target feature indicators in the model decision-making process. The local explanation explains a single prediction by assigning a specific value to each target feature indicator, showing the contribution of each target feature indicator to the prediction results of postoperative heart failure risk. Figure 7 As shown in the figure, the importance of the feature to the prediction of the heart failure model is shown by the average of the SHAP absolute value. The ten target feature indicators are ranked from large to small according to their contribution to the model: serum phosphorus, coronary heart disease (excluding coronary heart disease with myocardial infarction), pulse, diastolic blood pressure (low blood pressure), prothrombin time, age, absolute neutrophil count, myocardial infarction, acute kidney injury, and acute poisoning (alcohol, carbon monoxide, chemical poisons). Figure 8 The SHAP summary graph of ten target feature indicators is shown. Figure 8 It can be seen that coronary heart disease (excluding coronary heart disease with myocardial infarction), pulse, prothrombin time, age, absolute neutrophil count, myocardial infarction, acute kidney injury, acute poisoning (alcohol, carbon monoxide, chemical poisons) are positively correlated with the occurrence of heart failure; serum phosphorus and diastolic blood pressure (low blood pressure) are negatively correlated with the occurrence of heart failure. The correlation between the target characteristic indicators and heart failure is consistent with the logic of clinical medicine or with the literature reports.

[0075] Among them, the SHAP force diagram is used to intuitively display the patient's SHAP value and its impact on the model prediction results; through the SHAP force diagram, you can clearly see the contribution of each target feature indicator to the patient's prediction value. Fig. 9 The SHAP force diagram of the positive patients (patients to be tested) is shown. Fig.10 The SHAP force plot for negative patients (patients to be tested) is shown.

[0076] It should be noted that the baseline value of the SHAP force graph refers to the value predicted by the model without any feature value information.

[0077] Feature contribution means: the contribution of each target feature indicator is represented by a bar, and the length of the bar represents the impact of the feature on the final predicted value. Dark black bars represent positive contributions, that is, the feature increases the predicted value, and light gray bars represent negative contributions, that is, the feature decreases the predicted value. The value of the feature is marked below the drawing arrow. The output value represents the final predicted value of the model for the sample, which is the sum of the baseline value plus the contribution of all features.

[0078] In an optional manner, the method further includes:

[0079] Based on the postoperative heart failure risk prediction result and in combination with the medical database, recommended measures for the patient to be tested are generated.

[0080] The medical database includes, but is not limited to, the latest medical guidelines and expert opinions. Specifically, large language models (LLMs) are used in combination with the latest medical guidelines and expert opinions to automatically generate decisions (suggested measures) for the patients’ postoperative heart failure risk prediction results, thereby assisting doctors in making clinical decisions. This approach not only improves the accuracy of predictions, but also promotes the combination of standardization and personalization of prevention programs, thereby improving clinical outcomes.

[0081] In an optional manner, the method further includes:

[0082] According to the specific values ​​of the multiple target characteristic indicators of the patient to be tested, the postoperative heart failure risk prediction result and the recommended measures, a risk assessment report for the patient to be tested is generated and visually output.

[0083] The risk assessment report is designed to ensure that even clinicians without a statistical background can quickly understand the risk status of the patient being tested.

[0084] The technical solution of this embodiment can simplify the clinical workflow, improve the accuracy and efficiency of postoperative heart failure risk assessment, and ultimately achieve better patient monitoring and care.

[0085] Fig.11 FIG. 2 is a schematic diagram showing a structural diagram of an embodiment of a postoperative heart failure risk prediction system 200 provided by the present invention. Fig.11 As shown, the system 200 includes: a construction module 210 and a prediction module 220;

[0086] The construction module 210 is used to: determine a target machine learning model and multiple target feature indicators for postoperative heart failure risk prediction, and train the target machine learning model according to specific values ​​of multiple target feature indicators of each sample patient and a sample label characterizing whether each sample patient has postoperative heart failure, to obtain a trained machine learning model;

[0087] The prediction module 220 is used to input the specific values ​​of multiple target characteristic indicators of the patient to be tested into the trained machine learning model to obtain the postoperative heart failure risk prediction result of the patient to be tested.

[0088] In an optional manner, the construction module 210 is specifically used to:

[0089] Determine multiple original characteristic indicators for postoperative heart failure risk prediction from demographic characteristics, vital signs, medical history and laboratory test information;

[0090] According to the specific values ​​of multiple original feature indicators of each sample patient and the sample label, and using a five-fold cross-validation method, the performance of each preset machine learning model used for postoperative heart failure risk prediction was evaluated to obtain the performance measurement value corresponding to each preset machine learning model;

[0091] The preset machine learning model with the highest performance metric value is determined as the target machine learning model.

[0092] In an optional manner, the construction module 210 is specifically used to:

[0093] According to the importance of all original feature indicators to the target machine learning model and the stability of different numbers of feature indicators among all original feature indicators to the target machine learning model, multiple target feature indicators for postoperative heart failure risk prediction are determined.

[0094] In an optional manner, the method further includes: a generating module; the generating module is used to:

[0095] Using the SHAP method and based on the specific values ​​of multiple target characteristic indicators of the patient to be tested, the SHAP value that characterizes the contribution of each target characteristic indicator of the patient to be tested to the postoperative heart failure risk prediction result is calculated, and based on all the SHAP values, the SHAP force diagram of the patient to be tested is generated.

[0096] In an optional manner, the method further includes: a decision module; the decision module is used to:

[0097] Based on the postoperative heart failure risk prediction result and in combination with the medical database, recommended measures for the patient to be tested are generated.

[0098] In an optional manner, the method further includes: a visualization module; the visualization module is used to:

[0099] According to the specific values ​​of the multiple target characteristic indicators of the patient to be tested, the postoperative heart failure risk prediction result and the recommended measures, a risk assessment report for the patient to be tested is generated and visually output.

[0100] In an optional manner, the plurality of target characteristic indicators include: coronary heart disease, serum phosphorus, myocardial infarction, absolute neutrophil count, acute kidney injury, prothrombin time, pulse, acute poisoning, diastolic blood pressure and age.

[0101] It should be noted that the beneficial effects of the postoperative heart failure risk prediction system provided by the above embodiment are the same as the beneficial effects of the postoperative heart failure risk prediction method described above, which will not be repeated here. In addition, when the system provided by the above embodiment realizes its functions, it only takes the division of the above functional modules as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the system is divided into different functional modules according to actual conditions to complete all or part of the functions described above. In addition, the system and method embodiments provided by the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0102] Among them, the postoperative heart failure risk prediction system of the present invention can be a computer program (including program code) running in a computer device. For example, the postoperative heart failure risk prediction system of the present invention is an application software that can be used to execute the corresponding steps in the postoperative heart failure risk prediction method of the present invention.

[0103] In some embodiments, the postoperative heart failure risk prediction system of the present invention can be implemented in a combination of software and hardware. As an example, the postoperative heart failure risk prediction system of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the postoperative heart failure risk prediction method of the present invention. For example, the processor in the form of a hardware decoding processor can adopt one or more application-specific integrated circuits (ASIC, Application Specific Integrated Circuit), DSP, programmable logic device (PLD, Programmable Logic Device), complex programmable logic device (CPLD, Complex Programmable Logic Device), field programmable gate array (FPGA, Field-Programmable Gate Array) or other electronic components.

[0104] The modules involved in the embodiments of the present invention may be implemented in software or hardware. The name of a module does not limit the module itself in some cases.

[0105] An electronic device according to an embodiment of the present invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, any one of the above-mentioned methods for predicting the risk of postoperative heart failure is implemented. That is to say, an electronic device according to an embodiment of the present invention may include but is not limited to: a processor and a memory; a memory for storing a computer program; a processor for executing a method for predicting the risk of postoperative heart failure shown in any embodiment of the present invention by calling a computer program.

[0106] In an alternative embodiment, an electronic device is provided, such as Fig.12 As shown, Fig.12 The electronic device 4000 shown includes: a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, such as through a bus 4002. Optionally, the electronic device 4000 may also include a transceiver 4004, which may be used for data interaction between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0107] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the present invention. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0108] The bus 4002 may include a path to transmit information between the above components. The bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.12In the figure, only one thick line is used to represent the bus 4002, but this does not mean that there is only one bus or one type of bus.

[0109] The memory 4003 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to this.

[0110] The memory 4003 is used to store the application code (computer program) for executing the solution of the present invention, and the execution is controlled by the processor 4001. The processor 4001 is used to execute the application code stored in the memory 4003 to implement the content shown in the above method embodiment.

[0111] Among them, the electronic device can also be a terminal device, and the terminal device can be any terminal device that can install applications and access web pages through applications, including at least one of a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, a smart TV, and a smart car device.

[0112] It should be noted that Fig.12 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0113] A computer-readable storage medium according to an embodiment of the present invention stores a computer program, and when the computer program is executed by a processor, any of the above-mentioned methods for predicting the risk of postoperative heart failure is implemented.

[0114] Optionally, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0115] In an exemplary embodiment, a computer program product or computer program is also provided, the computer program product or computer program comprising computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device performs the above-mentioned postoperative heart failure risk prediction method.

[0116] Computer program code for performing the operations of the present invention may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0117] It should be understood that the flow charts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the module, the program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0118] The computer-readable storage medium provided by the embodiment of the present invention may be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or component.

[0119] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the method shown in the above embodiment.

[0120] The above description is only a preferred embodiment of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the disclosure scope involved in the present invention is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present invention (but not limited to) by each other.

[0121] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects and represent the definition of a specific order or sequence. The order of use of similar objects can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than the order shown or described.

[0122] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0123] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A method for predicting the risk of postoperative heart failure, characterized in that: include: Determine a target machine learning model and multiple target feature indicators for postoperative heart failure risk prediction, and train the target machine learning model according to the specific values ​​of the multiple target feature indicators of each sample patient and the sample label characterizing whether each sample patient has postoperative heart failure, to obtain a trained machine learning model; The specific values ​​of multiple target characteristic indicators of the patient to be tested are input into the trained machine learning model to obtain the postoperative heart failure risk prediction result of the patient to be tested.

2. The method for predicting postoperative heart failure risk according to claim 1, characterized in that: Steps to determine the target machine learning model for postoperative heart failure risk prediction include: Determine multiple original characteristic indicators for postoperative heart failure risk prediction from demographic characteristics, vital signs, medical history and laboratory test information; According to the specific values ​​of multiple original feature indicators of each sample patient and the sample label, and using a five-fold cross-validation method, the performance of each preset machine learning model used for postoperative heart failure risk prediction was evaluated to obtain the performance measurement value corresponding to each preset machine learning model; The preset machine learning model with the highest performance metric value is determined as the target machine learning model.

3. The method for predicting postoperative heart failure risk according to claim 2, characterized in that: The step of determining the target characteristic index for postoperative heart failure risk prediction comprises: According to the importance of all original feature indicators to the target machine learning model and the stability of different numbers of feature indicators among all original feature indicators to the target machine learning model, multiple target feature indicators for postoperative heart failure risk prediction are determined.

4. The method for predicting postoperative heart failure risk according to claim 1, characterized in that: Also includes: Using the SHAP method and based on the specific values ​​of multiple target characteristic indicators of the patient to be tested, the SHAP value that characterizes the contribution of each target characteristic indicator of the patient to be tested to the postoperative heart failure risk prediction result is calculated, and based on all the SHAP values, the SHAP force diagram of the patient to be tested is generated.

5. The method for predicting postoperative heart failure risk according to claim 1, characterized in that: Also includes: Based on the postoperative heart failure risk prediction result and in combination with the medical database, recommended measures for the patient to be tested are generated.

6. The method for predicting postoperative heart failure risk according to claim 5, characterized in that: Also includes: According to the specific values ​​of the multiple target characteristic indicators of the patient to be tested, the postoperative heart failure risk prediction result and the recommended measures, a risk assessment report for the patient to be tested is generated and visually output.

7. The method for predicting postoperative heart failure risk according to any one of claims 1 to 6, characterized in that: Multiple target characteristic indicators included: coronary heart disease, serum phosphorus, myocardial infarction, absolute neutrophil count, acute kidney injury, prothrombin time, pulse, acute poisoning, diastolic blood pressure, and age.

8. A postoperative heart failure risk prediction system, characterized in that: include: Building modules and prediction modules; The construction module is used to: determine a target machine learning model and multiple target feature indicators for postoperative heart failure risk prediction, and train the target machine learning model according to specific values ​​of the multiple target feature indicators of each sample patient and a sample label characterizing whether each sample patient has postoperative heart failure, to obtain a trained machine learning model; The prediction module is used to input specific values ​​of multiple target characteristic indicators of the patient to be tested into the trained machine learning model to obtain a postoperative heart failure risk prediction result for the patient to be tested.

9. An electronic device, characterized in that: The electronic device includes a processor, which is coupled to a memory. The memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor so that the electronic device implements the postoperative heart failure risk prediction method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one computer program, and the at least one computer program is loaded and executed by a processor so that the computer-readable storage medium implements the postoperative heart failure risk prediction method as described in any one of claims 1 to 7.

Images