Auxiliary prediction method and device for hepatic failure after hepatic resection, and electronic equipment
Through the training, the liver failure prediction model is used to predict changes in liver function after hepatic resection using multi-dimensional clinical data, which solves the problem of low prediction accuracy of liver failure after hepatic resection in the prior art, and improves prediction accuracy and postoperative safety.
Patent Information
- Application Number
- CN202510122691.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-26
- Publication Date
- 2025-06-03
AI Technical Summary
The prior art has low accuracy in predicting liver failure after hepatic resection, and it is impossible to effectively judge the risk of liver failure after hepatic resection.
By obtaining multiple sets of training data, including clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information and postoperative liver failure results, a liver failure prediction model was obtained. This model can input relevant data from the target patient and predict the probability of abnormal liver function indicators after hepatic resection.
It improves the prediction accuracy of liver failure after hepatic resection, reduces misdiagnosis and misdiagnosis, reduces the incidence of postoperative complications, and improves the safety and treatment effect of hepatic resection.
Smart Images
Figure CN120089361A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure belongs to the field of medical technologies, and particularly relates to a method, a device, and an electronic device for assisting in predicting liver failure after hepatectomy. Background Art
[0002] Hepatectomy is a technical means widely used in the treatment of diseases such as liver tumors and benign liver diseases. Although hepatectomy provides an effective treatment means for patients, the complication rate of hepatectomy is still relatively high. Among them, liver failure after hepatectomy is one of the most serious complications and an important cause of early death of hepatectomy patients.
[0003] Currently, clinically, the assessment of liver function mainly uses the Child-Pugh score and the MELD score. The Child-Pugh score mainly scores liver function based on hepatic encephalopathy, ascites, serum bilirubin, serum albumin level, and prothrombin time. This method has subjective factors, resulting in certain limitations in its use and cannot accurately judge the risk of liver failure after hepatectomy. It is necessary to further comprehensively judge and evaluate in combination with other examinations, and the prediction accuracy is relatively low; the MELD score mainly evaluates the liver function of patients based on the serum creatinine level and bilirubin level. This method is mainly for patients with end-stage liver diseases, and serum creatinine and bilirubin are easily affected by underlying diseases such as kidney diseases and cholestasis. Especially for patients with compensated liver function, it lacks specificity in judging the risk of liver failure after hepatectomy. Summary of the Invention
[0004] Embodiments of the present disclosure provide a solution to solve the problem of inaccurate prediction of liver failure after hepatectomy in the related art.
[0005] In a first aspect, the present disclosure provides a method for assisting in predicting liver failure after hepatectomy, the method including:
[0006] Obtain multiple groups of training data, where the training data includes the clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of sample patients who have undergone partial hepatectomy. The clinical baseline data at least includes the basic demographic information of the patients, the imaging data at least includes abdominal CT images, and the laboratory indicators at least include blood routine parameters, biochemical indicators, and coagulation function parameters; the surgical plan information includes the hepatectomy range, operation time, estimated intraoperative blood transfusion volume, and estimated intraoperative blood loss volume, and the postoperative liver failure results include abnormal liver function indicators or normal liver function indicators;
[0007] Based on the multiple groups of training data, train a liver failure prediction model;
[0008] Obtain the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient who has undergone partial hepatectomy;
[0009] Input the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient into the liver failure prediction model to obtain the probability of abnormal liver function indicators occurring after hepatectomy for the target patient.
[0010] In a second aspect, the present disclosure provides an auxiliary prediction device for liver failure after hepatectomy, and the device includes:
[0011] An acquisition unit for acquiring multiple sets of training data, where the training data includes the clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of sample patients who have undergone partial hepatectomy. The clinical baseline data includes at least the basic demographic information of the patients, the imaging data includes at least abdominal CT images, the laboratory indicators include at least blood routine parameters, biochemical indicators, and coagulation function parameters; the surgical plan information includes the hepatectomy range, operation time, estimated intraoperative blood transfusion volume, and estimated intraoperative blood loss volume, and the postoperative liver failure results include abnormal liver function indicators or normal liver function indicators;
[0012] A training unit for training a liver failure prediction model based on the multiple sets of training data;
[0013] The acquisition unit for acquiring the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient who has undergone partial hepatectomy;
[0014] A determination unit for inputting the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient into the liver failure prediction model to obtain the probability of abnormal liver function indicators occurring after hepatectomy for the target patient.
[0015] In a third aspect, the present disclosure provides an electronic device, including:
[0016] A processor; and
[0017] A memory for storing executable instructions of the processor;
[0018] Wherein, the processor is configured to execute any method in the first aspect or any possible implementation manner of the first aspect by executing the executable instructions.
[0019] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements any method in the first aspect or any possible implementation manner of the first aspect.
[0020] The technical solution provided by the present disclosure obtains multiple sets of training data. The training data includes the clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of sample patients who have undergone partial hepatectomy. The clinical baseline data includes at least the basic demographic information of the patients, the imaging data includes at least abdominal CT images, and the laboratory indicators include at least blood routine parameters, biochemical indicators, and coagulation function parameters. The surgical plan information includes the hepatectomy range, operation time, estimated intraoperative blood transfusion volume, and estimated intraoperative blood loss volume. The postoperative liver failure results include abnormal liver function indicators or normal liver function indicators. Based on the multiple sets of training data, a liver failure prediction model is trained. The clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of a target patient who has undergone partial hepatectomy are obtained. The clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient are input into the liver failure prediction model to obtain the probability of abnormal liver function indicators after hepatectomy of the target patient. The technical solutions provided by the embodiments of the present disclosure train a liver failure prediction model by combining multi-dimensional clinical data, can accurately predict the changes in liver function after hepatectomy, improve the prediction accuracy of liver failure after hepatectomy, reduce misdiagnosis and missed diagnosis, reduce the incidence of postoperative complications, and further improve the safety and treatment effect of hepatectomy. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or related technologies, the following will briefly introduce the drawings required for the description of the embodiments or related technologies. Obviously, the drawings in the following description are some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. In the drawings:
[0022] Figure 1 is a schematic flowchart of an auxiliary prediction method for liver failure after hepatectomy provided by an embodiment of the present disclosure;
[0023] Figure 2 is a schematic structural diagram of an auxiliary prediction device for liver failure after hepatectomy provided by an embodiment of the present disclosure;
[0024] Figure 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0025] Embodiments of the present disclosure will be described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present disclosure, and should not be construed as limiting the present disclosure.
[0026] Terms such as "first" and "second" in the description, claims, and drawings of the embodiments of the present disclosure are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.
[0027] The auxiliary prediction method for liver failure after hepatectomy provided by the embodiments of the present disclosure can run on a terminal device or a server. Among them, the terminal device can be a local terminal device, including wearable devices such as VR (Virtual Reality), AR (Augmented Reality), and MR (Mixed Reality). Among them, the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.
[0028] Hepatectomy is a technical means widely used to treat diseases such as liver tumors and liver benign diseases. Although hepatectomy provides an effective treatment means for patients, the complication rate of hepatectomy is still relatively high. Among them, liver failure after hepatectomy is one of the most serious complications and an important cause of early death of hepatectomy patients.
[0029] Currently in clinical practice, the assessment of liver function mainly uses the Child-Pugh score and the MELD score. The Child-Pugh score mainly scores liver function based on hepatic encephalopathy, ascites, serum bilirubin, serum albumin level, and prothrombin time. This method has subjective factors, resulting in certain limitations in its application. It cannot accurately judge the risk of liver failure after hepatectomy and requires further comprehensive judgment and evaluation in combination with other examinations, with relatively low prediction accuracy. The MELD score mainly assesses the liver function of patients based on serum creatinine level and bilirubin level. This method is mainly for patients with end-stage liver disease, and serum creatinine and bilirubin are easily affected by underlying diseases such as kidney disease and cholestasis. Especially for patients with compensated liver function, it lacks specificity in judging the risk of liver failure after hepatectomy.
[0030] The following uses specific embodiments to elaborate in detail on the technical solutions of the present disclosure and how the technical solutions of the present disclosure solve the above technical problems. These specific embodiments below can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present disclosure will be described below in conjunction with the accompanying drawings.
[0031] Figure 1 FIG. is a schematic flowchart of an auxiliary prediction method for liver failure after hepatectomy provided by an exemplary embodiment of the present disclosure. This method can be applied to an electronic device with data processing capabilities. Taking the application of this method to an auxiliary decision-making device in a hospital as an example, this solution at least includes the following steps S101-S104:
[0032] S101, obtain multiple sets of training data.
[0033] The training data includes the clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of sample patients who have undergone partial hepatectomy.
[0034] In some embodiments, the clinical baseline data at least includes the basic demographic information of the patient. Among them, the basic demographic information of the patient may include: the gender information of the patient, the age information of the patient, etc.
[0035] In some embodiments, the imaging data at least includes abdominal CT images.
[0036] Among them, in order to improve the accuracy of training, the imaging data may also include magnetic resonance MRI images.
[0037] In some embodiments, the laboratory indicators at least include blood routine parameters, biochemical indicators, and coagulation function parameters.
[0038] Among them, the blood routine parameters at least include: red blood cell count, platelet count, white blood cell count, hemoglobin level, etc.
[0039] The biochemical indexes at least include: total bilirubin level, prothrombin time, serum transaminase value, albumin level, etc.
[0040] The coagulation function parameters at least include: PT reflecting exogenous coagulation function, INR for monitoring the effect of anticoagulants, APTT reflecting endogenous coagulation function, TT reflecting fibrinogen conversion time, FIB as a coagulation factor, and D-dimer indicating hyperfibrinolysis and thrombosis.
[0041] In some embodiments, in order to increase the accuracy of the data during the training process, the biochemical indexes may further include direct bilirubin, indirect bilirubin, international normalized ratio, alanine aminotransferase, aspartate aminotransferase, direct bilirubin, serum albumin, creatinine, etc.
[0042] In some embodiments, the surgical plan information includes the liver resection range, operation time, estimated intraoperative blood transfusion volume, and estimated intraoperative blood loss volume.
[0043] In some embodiments, the postoperative liver failure results include abnormal liver function indexes or normal liver function indexes.
[0044] In some embodiments, during the actual acquisition of training data, the auxiliary decision-making device is connected to the hospital information system (HIS), laboratory information system (LIS), and imaging management system (PACS) of the hospital, so the training data can be directly obtained from the hospital information system (HIS), laboratory information system (LIS), and imaging management system (PACS).
[0045] In some embodiments, in order to ensure the accuracy of the training model, it is preferable to obtain the training data of 1000 patients from different hospitals. Specifically, it can be set according to the actual situation.
[0046] In some embodiments, obtaining multiple groups of training data includes steps S11 - S12:
[0047] S11, obtaining multiple groups of original data.
[0048] In some embodiments, the original data includes: the original clinical data, original imaging data, original clinical baseline data, original imaging data, original preoperative laboratory indicators, and original surgical plan information of the sample patients who have undergone partial liver resection surgery.
[0049] S12, preprocessing the multiple groups of original data to obtain the multiple groups of training data.
[0050] Among them, the preprocessing includes data cleaning and normalization processing.
[0051] Specifically, in the specific acquisition process, the original data can be directly obtained from the hospital information system (HIS), laboratory information system (LIS), and picture archiving and communication system (PACS). However, the obtained original data may contain invalid data and abnormal data. Therefore, the original data needs to be preprocessed. For example, the original data can be cleaned, which can specifically include removing missing values and outliers, and normalization processing can be performed, which can specifically include specifying the standardized numerical range. By adopting this method, the consistency and reliability of the data can be ensured.
[0052] S102. Based on the multiple sets of training data, a liver failure prediction model is trained.
[0053] In some embodiments, based on the multiple sets of training data, training a liver failure prediction model includes: based on the multiple sets of training data, using an ensemble learning algorithm to train a liver failure prediction model.
[0054] Specifically, when training the liver failure prediction model, algorithms such as random forest, gradient boosting tree, support vector machine, decision tree, artificial neural network, extreme gradient boosting, and light gradient boosting machine can also be used.
[0055] In some embodiments, in order to ensure the accuracy of the obtained liver failure prediction model, the trained liver failure prediction model needs to be verified. The specific method is as follows in steps S21 - S24:
[0056] S21. Obtain multiple sets of verification data.
[0057] In some embodiments, the verification data includes the clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of the verified patients who have undergone partial hepatectomy.
[0058] Specifically, in the acquisition process, the verification data can be directly obtained from the hospital information system (HIS), laboratory information system (LIS), and picture archiving and communication system (PACS). Among them, the verification data is different from the training data obtained in the above step S101 and is the relevant data of different patients.
[0059] S22. Input the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the verified patients into the liver failure prediction model to obtain the probability of abnormal liver function indicators after hepatectomy of the verified patients.
[0060] S23. Determine whether the probability of abnormal postoperative liver function indicators of the verified patient is the same as the postoperative liver failure outcome of the verified patient.
[0061] Among them, if the probability of abnormal postoperative liver function indicators is greater than the preset threshold, it indicates abnormal liver function indicators. Conversely, the liver function is normal.
[0062] Specifically, a probability of abnormal postoperative liver function indicators greater than the preset threshold means that the risk of liver failure is very high after the patient undergoes hepatectomy. Conversely, the risk is very low.
[0063] S24. If they are different, retrain the liver failure prediction model.
[0064] In some embodiments, if the probability of abnormal postoperative liver function indicators of the verified patient is different from the postoperative liver failure outcome of the verified patient, it means that the liver failure prediction model is inaccurate and needs to be optimized and updated or retrained.
[0065] In this embodiment, the verified patient and the sample patient are not in the same group and are completely different patients.
[0066] In some embodiments, retraining the liver failure prediction model includes steps S241 - S242:
[0067] S241. Obtain multiple sets of test data.
[0068] In some embodiments, the test data includes the clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure outcome of test patients who have undergone partial hepatectomy.
[0069] S242. Based on the multiple sets of test data, retrain to obtain a new liver failure prediction model.
[0070] Specifically, during the acquisition process, the test data can be directly obtained from the hospital information system (HIS), laboratory information system (LIS), and picture archiving and communication system (PACS). The test data is different from the training data obtained in step S101 and the verification data in step S21 above, and is the relevant data of different patients.
[0071] S103. Obtain the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient who has undergone partial hepatectomy.
[0072] In some embodiments, the target patient refers to a patient who currently intends to undergo hepatectomy.
[0073] Among them, the clinical baseline data, imaging data, and preoperative laboratory indicators of the target patient can be obtained from the hospital information system (HIS), laboratory information system (LIS), and picture archiving and communication system (PACS). The surgical plan information is the plan specified by the doctor after diagnosing the condition of the target patient and is stored in the hospital information system (HIS).
[0074] S104, input the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient into the liver failure prediction model to obtain the probability of abnormal liver function indicators after hepatectomy for the target patient.
[0075] Specifically, if the probability of abnormal liver function indicators after hepatectomy for the target patient output is not greater than the preset threshold, it means that the target patient is to undergo hepatectomy. If the probability of abnormal liver function indicators after hepatectomy for the target patient is greater than the preset threshold, it means that the target patient does not need to undergo hepatectomy.
[0076] In some embodiments, in order to improve the treatment efficiency, after obtaining the probability of abnormal liver function indicators after hepatectomy for the target patient, the method further includes: determining the intervention treatment plan for the target patient based on the probability of abnormal liver function indicators after hepatectomy for the target patient, and the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient.
[0077] Among them, the intervention treatment plan refers to whether to suspend the surgery or perform further examinations, treatments, or drug interventions, etc.
[0078] Specifically, if the probability of abnormal liver function indicators after hepatectomy for the target patient output is not greater than the preset threshold, it means that the target patient has normal liver function after hepatectomy and is suitable for hepatectomy. Therefore, the intervention treatment plan for the target patient is output, that is, further examination of the hepatectomy. The doctor makes corresponding decisions based on the intervention treatment plan and the probability of liver failure after hepatectomy. If the probability of abnormal liver function indicators after hepatectomy output is greater than the preset threshold, it means that the target patient will have liver failure after hepatectomy, and the current physical condition of the target patient is not suitable for hepatectomy. Therefore, the intervention treatment plan for the target patient is output, that is, suspend the surgery for further examination. The doctor makes corresponding decisions based on the intervention treatment plan and the probability of liver failure after hepatectomy. By adopting this method, the work efficiency of doctors can be improved, and the risk of patient complications can be reduced.
[0079] In some embodiments, after obtaining the postoperative liver failure result of the target patient, the method further includes: sending the probability of abnormal liver function indicators after hepatectomy to the client so that the user can view the probability of abnormal liver function indicators after hepatectomy; if the probability of abnormal liver function indicators after hepatectomy is greater than the preset threshold, feedback a suggestion to suspend the operation; if the probability of abnormal liver function indicators after hepatectomy is not greater than the preset threshold, feedback a suggestion for surgical treatment.
[0080] Specifically, the auxiliary decision-making device in the hospital will send the probability of abnormal liver function indicators after hepatectomy of the target patient to the client, that is, the doctor's terminal, through the web platform or medical terminal device. The doctor can view the probability of abnormal liver function indicators after hepatectomy of the target patient and then make corresponding decisions.
[0081] To better understand the embodiments of the present solution and further explain the process of the auxiliary prediction method for liver failure after hepatectomy in the present solution, the specific content is as follows:
[0082] Patient basic information: 45 years old, male, weight 70 kg, normal preoperative liver function, suffering from hepatocellular carcinoma;
[0083] Preoperative laboratory indicators: serum transaminase (ALT) value 100 U / L, total bilirubin (TBIL) level 1.5 mg / dL, albumin (ALB) level 3.2 g / d, prothrombin time (PT) 13 s;
[0084] Imaging data: Preoperative CT scan shows that the liver resection range and the remaining liver function are normal.
[0085] Inputting the above data into the trained liver failure prediction model, the obtained data is as follows:
[0086] Probability of abnormal liver function indicators after hepatectomy: 70% (high risk);
[0087] Intervention treatment plan: Strengthen liver function protection treatment, consider using liver protection drugs, and closely monitor the patient's liver function indicators, such as transaminase and bilirubin levels.
[0088] In this embodiment, the doctor conducts corresponding liver protection treatment on the patient and increases the monitoring frequency according to the probability of abnormal liver function indicators after hepatectomy and the intervention treatment plan to ensure that severe liver failure does not occur during the postoperative recovery process. In this way, the doctor's work efficiency can be improved, and the incidence of postoperative complications can be effectively reduced, ensuring the patient's postoperative recovery.
[0089] The technical solution provided by the present disclosure is to obtain multiple sets of training data. The training data includes the clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of sample patients who have undergone partial hepatectomy. The clinical baseline data includes at least the basic demographic information of the patients, the imaging data includes at least abdominal CT images, and the laboratory indicators include at least blood routine parameters, biochemical indicators, and coagulation function parameters. The surgical plan information includes the liver resection range, operation time, estimated intraoperative blood transfusion volume, and estimated intraoperative blood loss volume. The postoperative liver failure results include abnormal liver function indicators or normal liver function indicators. Based on the multiple sets of training data, a liver failure prediction model is trained. The clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of a target patient who has undergone partial hepatectomy are obtained. The clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient are input into the liver failure prediction model to obtain the probability of abnormal liver function indicators after hepatectomy for the target patient. The technical solutions provided by the embodiments of the present disclosure train a liver failure prediction model by combining multi-dimensional clinical data, can accurately predict the changes in liver function after hepatectomy, improve the prediction accuracy of liver failure after hepatectomy, reduce misdiagnosis and missed diagnosis, reduce the incidence of postoperative complications, and further improve the safety and treatment effect of hepatectomy.
[0090] Figure 2 FIG. 4 is a schematic structural diagram of an auxiliary prediction device for liver failure after hepatectomy provided by an exemplary embodiment of the present disclosure;
[0091] Among them, the device includes: an acquisition unit 201, a training unit 202, and a determination unit 203;
[0092] The acquisition unit 201 is configured to obtain multiple sets of training data. The training data includes the clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of sample patients who have undergone partial hepatectomy. The clinical baseline data includes at least the basic demographic information of the patients, the imaging data includes at least abdominal CT images, and the laboratory indicators include at least blood routine parameters, biochemical indicators, and coagulation function parameters. The surgical plan information includes the liver resection range, operation time, estimated intraoperative blood transfusion volume, and estimated intraoperative blood loss volume. The postoperative liver failure results include abnormal liver function indicators or normal liver function indicators.
[0093] The training unit 202 is configured to train a liver failure prediction model based on the multiple sets of training data;
[0094] The acquisition unit 201 is configured to obtain the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of a target patient who has undergone partial hepatectomy;
[0095] A determination unit 203, configured to input the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient into the liver failure prediction model, and obtain the probability of abnormal liver function indicators after hepatectomy of the target patient.
[0096] In some embodiments, the device is further configured to:
[0097] Obtain multiple groups of verification data, where the verification data includes the clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of verification patients who have undergone partial hepatectomy;
[0098] Input the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the verification patients into the liver failure prediction model, and obtain the probability of abnormal liver function indicators after hepatectomy of the verification patients;
[0099] Determine whether the probability of abnormal liver function indicators after hepatectomy of the verification patients is the same as the postoperative liver failure results of the verification patients, where if the probability of abnormal liver function indicators after hepatectomy is greater than a preset threshold, it indicates abnormal liver function indicators, and vice versa, the liver function is normal;
[0100] If they are different, retrain the liver failure prediction model.
[0101] In some embodiments, the device is configured to retrain the liver failure prediction model, and the device is specifically configured to:
[0102] Obtain multiple groups of test data, where the test data includes the clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of test patients who have undergone partial hepatectomy;
[0103] Based on the multiple groups of test data, retrain to obtain a new liver failure prediction model.
[0104] In some embodiments, the device is configured to train a liver failure prediction model based on the multiple groups of training data, and the device is specifically configured to:
[0105] Based on the multiple groups of training data, use an ensemble learning algorithm to train the liver failure prediction model.
[0106] In some embodiments, the device is configured to obtain multiple groups of training data, and the device is specifically configured to:
[0107] Obtain multiple sets of original data, where the original data includes: the original clinical data of sample patients who have undergone partial hepatectomy, original imaging data, original clinical baseline data, original imaging data, original preoperative laboratory indicators, and original surgical plan information;
[0108] Preprocess the multiple sets of original data to obtain the multiple sets of training data;
[0109] Among them, the preprocessing includes data cleaning and normalization processing.
[0110] In some embodiments, after the device is used to obtain the probability of abnormal liver function indicators after hepatectomy of the target patient, the device is specifically used for:
[0111] Based on the probability of abnormal liver function indicators after hepatectomy of the target patient and the clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient, determine the intervention treatment plan for the target patient.
[0112] In some embodiments, the device is used to obtain the probability of abnormal liver function indicators after hepatectomy of the target patient. Specifically, the device is used to: send the probability of abnormal liver function indicators after hepatectomy to the client so that the user can view the probability of abnormal liver function indicators after hepatectomy; if the probability of abnormal liver function indicators after hepatectomy is greater than the preset threshold, feedback a suggestion to suspend the operation; if the probability of abnormal liver function indicators after hepatectomy is not greater than the preset threshold, feedback a suggestion for surgical treatment.
[0113] The technical solution provided by the present disclosure obtains multiple sets of training data. The training data includes the clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of sample patients who have undergone partial hepatectomy. The clinical baseline data includes at least the basic demographic information of the patients, the imaging data includes at least abdominal CT images, and the laboratory indicators include at least blood routine parameters, biochemical indicators, and coagulation function parameters. The surgical plan information includes the liver resection range, operation time, estimated intraoperative blood transfusion volume, and estimated intraoperative blood loss volume. The postoperative liver failure results include abnormal liver function indicators or normal liver function indicators. Based on the multiple sets of training data, a liver failure prediction model is trained. The clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of a target patient who has undergone partial hepatectomy are obtained. The clinical baseline data, imaging data, preoperative laboratory indicators, and surgical plan information of the target patient are input into the liver failure prediction model to obtain the probability of abnormal liver function indicators occurring after hepatectomy for the target patient. The technical solutions provided by the embodiments of the present disclosure train a liver failure prediction model by combining multi-dimensional clinical data, can accurately predict the changes in liver function after hepatectomy, improve the prediction accuracy of liver failure after hepatectomy, reduce misdiagnosis and missed diagnosis, reduce the incidence of postoperative complications, and further improve the safety and treatment effect of hepatectomy.
[0114] It should be understood that the device embodiments and the method embodiments can correspond to each other, and similar descriptions can refer to the method embodiments. To avoid repetition, it will not be elaborated here. Specifically, the device can execute the above method embodiments, and the foregoing and other operations and / or functions of each module in the device respectively correspond to the corresponding processes in each method in the above method embodiments. For the sake of brevity, it will not be elaborated here.
[0115] In the foregoing, the device of the embodiments of the present disclosure has been described from the perspective of functional modules. It should be understood that the functional modules can be implemented in the form of hardware, can also be implemented by instructions in the form of software, or can be implemented by a combination of hardware and software modules. Specifically, the steps of the method embodiments in the present disclosure can be completed by the integrated logic circuit of the hardware in the processor and / or instructions in the form of software. The steps of the method disclosed in combination with the embodiments of the present disclosure can be directly embodied as being executed and completed by the hardware decoding processor, or can be executed and completed by a combination of the hardware and software modules in the decoding processor. Optionally, the software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the steps in the above method embodiments.
[0116] Figure 3It is a schematic block diagram of an electronic device provided by an embodiment of the present disclosure. The electronic device may include:
[0117] A memory 301 and a processor 302. The memory 301 is used to store a computer program and transmit the program code to the processor 302. In other words, the processor 302 can call and run the computer program from the memory 301 to implement the method in the embodiment of the present disclosure.
[0118] For example, the processor 302 can be used to execute the above method embodiment according to the instructions in the computer program.
[0119] In some embodiments of the present disclosure, the processor 302 may include but is not limited to:
[0120] A general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and so on.
[0121] In some embodiments of the present disclosure, the memory 301 includes but is not limited to:
[0122] Volatile memory and / or non-volatile memory. Among them, the non-volatile memory can be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically EPROM (EEPROM), or a flash memory. The volatile memory can be a Random Access Memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double DataRate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), synch link DRAM (SLDRAM), and Direct Rambus RAM (DR RAM).
[0123] In some embodiments of the present disclosure, the computer program may be divided into one or more modules, and the one or more modules are stored in the memory 301 and executed by the processor 302 to complete the method provided by the present disclosure. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.
[0124] As Figure 3 shown, the electronic device may further include:
[0125] A transceiver 303, which may be connected to the processor 302 or the memory 301.
[0126] Among them, the processor 302 can control the transceiver 303 to communicate with other devices. Specifically, it can send information or data to other devices, or receive information or data sent by other devices. The transceiver 303 may include a transmitter and a receiver. The transceiver 303 may further include an antenna, and the number of antennas may be one or more.
[0127] It should be understood that the various components in the electronic device are connected through a bus system. Among them, the bus system includes not only a data bus, but also a power bus, a control bus, and a status signal bus.
[0128] The present disclosure also provides a computer storage medium, on which a computer program is stored. When the computer program is executed by a computer, the computer can execute the methods in the above method embodiments. Or rather, the embodiments of the present disclosure also provide a computer program product containing instructions. When the instructions are executed by a computer, the computer executes the methods in the above method embodiments.
[0129] When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions according to the embodiments of the present disclosure are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or a wireless manner (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that the computer can access, or a data storage device such as a server or a data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a digital video disc (DVD)), or a semiconductor medium (such as a solid state disk (SSD)).
[0130] Those of ordinary skill in the art can realize that the modules and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present disclosure.
[0131] In several embodiments provided by the present disclosure, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or modules can be in electrical, mechanical, or other forms.
[0132] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they can be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. For example, in each embodiment of the present disclosure, the various functional modules can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.
[0133] The above is only the specific implementation manner of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present disclosure can easily think of changes or substitutions, which should all be covered by the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.
Claims
1. A method for assisting prediction of liver failure after liver resection, characterized in that: The method comprises: Acquire multiple sets of training data, wherein the training data include clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of sample patients who underwent partial liver resection surgery, wherein the clinical baseline data at least include basic patient demographic information, the imaging data at least include abdominal CT images, and the laboratory indicators at least include routine blood parameters, biochemical indicators, and coagulation function parameters; the surgical plan information includes the range of liver resection, operation time, estimated intraoperative blood transfusion volume, and estimated intraoperative blood loss; and the postoperative liver failure results include abnormal liver function indicators or normal liver function indicators; Based on the multiple sets of training data, a liver failure prediction model is trained; Obtain clinical baseline information, imaging data, preoperative laboratory indicators, and surgical plan information of target patients undergoing partial liver resection; The clinical baseline information, imaging data, preoperative laboratory indicators and surgical plan information of the target patient are input into the liver failure prediction model to obtain the probability of abnormal liver function indicators of the target patient after liver resection.
2. The method according to claim 1, characterized in that The method further comprises: Acquire multiple groups of validation data, including clinical baseline information, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of validation patients who underwent partial liver resection surgery; Inputting the clinical baseline data, imaging data, preoperative laboratory indicators and surgical plan information of the verification patient into the liver failure prediction model to obtain the probability of abnormal liver function indicators after liver resection of the verification patient; Determine whether the probability of abnormal liver function index after liver resection of the verification patient is the same as the result of postoperative liver failure of the verification patient, wherein if the probability of abnormal liver function index after liver resection is greater than a preset threshold, it indicates that the liver function index is abnormal, otherwise, the liver function is normal; If different, the liver failure prediction model is retrained.
3. The method according to claim 2, characterized in that Retraining the liver failure prediction model, including: Acquiring multiple groups of test data, the test data including clinical baseline information, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of test patients who underwent partial liver resection surgery; Based on the multiple groups of test data, a new liver failure prediction model is retrained.
4. The method according to claim 1, characterized in that: Based on the multiple sets of training data, a liver failure prediction model is trained, including: Based on the multiple groups of training data, the liver failure prediction model is obtained by training using an integrated learning algorithm.
5. The method according to claim 1, characterized in that Get multiple sets of training data, including: Acquire multiple groups of original data, wherein the original data include: original clinical data, original imaging data, original clinical baseline data, original imaging data, original preoperative laboratory indicators, and original surgical plan information of sample patients undergoing partial liver resection surgery; Preprocessing the multiple groups of original data to obtain the multiple groups of training data; Wherein, the preprocessing includes data cleaning and normalization processing.
6. The method according to claim 1, characterized in that After obtaining the probability of abnormal liver function indexes after liver resection of the target patient, the method further includes: Based on the probability of abnormal liver function indicators after liver resection of the target patient and the target patient's clinical baseline data, imaging data, preoperative laboratory indicators and surgical plan information, the intervention treatment plan for the target patient is determined.
7. The method according to claim 1, characterized in that After obtaining the probability of abnormal liver function indicators after liver resection of the target patient, the method further includes: sending the probability of abnormal liver function indicators after liver resection to a client so that a user can view the probability of abnormal liver function indicators after liver resection; if the probability of abnormal liver function indicators after liver resection is greater than a preset threshold, feedback is given to recommend suspension of the operation; if the probability of abnormal liver function indicators after liver resection is not greater than the preset threshold, feedback is given to recommend surgical treatment.
8. An auxiliary prediction device for liver failure after liver resection, characterized in that: The device comprises: an acquisition unit, for acquiring multiple sets of training data, wherein the training data include clinical baseline data, imaging data, preoperative laboratory indicators, surgical plan information, and postoperative liver failure results of sample patients undergoing partial liver resection surgery, wherein the clinical baseline data at least include basic patient demographic information, the imaging data at least include abdominal CT images, and the laboratory indicators at least include routine blood parameters, biochemical indicators, and coagulation function parameters; the surgical plan information includes the range of liver resection, operation time, estimated intraoperative blood transfusion volume, and estimated intraoperative blood loss; and the postoperative liver failure results include abnormal liver function indicators or normal liver function indicators; A training unit, used for training a liver failure prediction model based on the multiple sets of training data; The acquisition unit is used to acquire clinical baseline data, imaging data, preoperative laboratory indicators and surgical plan information of the target patient undergoing partial liver resection surgery; The determination unit is used to input the clinical baseline information, imaging data, preoperative laboratory indicators and surgical plan information of the target patient into the liver failure prediction model to obtain the probability of abnormal liver function indicators after liver resection of the target patient.
9. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to perform the method of any one of claims 1 to 7 by executing the executable instructions.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.