Disease course prediction method, apparatus, device, and storage medium
By using a disease progression prediction model and regression analysis based on multi-dimensional medical record data, the problem of low efficiency in disease progression prediction has been solved, achieving efficient and accurate disease progression prediction and support for treatment plans.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- WINNING HEALTH TECHNOLOGY GROUP CO LTD
- Filing Date
- 2022-09-26
- Publication Date
- 2026-05-15
AI Technical Summary
Existing technologies for predicting disease progression are inefficient, require significant manual labor, have low data accuracy, and are prone to errors.
Using a pre-defined disease progression data model, the system predicts disease progression through multi-dimensional medical record data, and combines regression analysis algorithms and linear fitting to generate the disease progression direction and predict the cure time.
It improves the efficiency of disease course prediction, provides data support for the generation of subsequent treatment plans, reduces labor costs, and improves the accuracy of data statistics.
Smart Images
Figure CN115565687B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of data processing technology, and more specifically, to a method, apparatus, device, and storage medium for predicting the course of a disease. Background Technology
[0002] With the development of technology, hospitals need to rationally allocate medical and nursing care and bed resources based on patients' admission and discharge status.
[0003] Current technologies generally require multi-dimensional comprehensive model analysis to predict the progression of a patient's disease. This involves deriving a patient's disease progression model based on existing records of changes in the patient's disease progression. On this basis, treatment resources can be allocated more rationally, bed turnover rate can be improved, and patients can be cured as early as possible. Data extraction and integration are carried out in various systems; manual backend retrieval of database data puts performance pressure on the running system.
[0004] However, this method of processing consumes a lot of labor costs, has low data statistics accuracy, and is prone to errors. Summary of the Invention
[0005] The purpose of this application is to provide a method, apparatus, device, and storage medium for predicting the course of a disease, in order to address the shortcomings of the prior art and solve the problem of low prediction efficiency in the prior art.
[0006] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows:
[0007] In a first aspect, one embodiment of this application provides a method for predicting the course of a disease, the method comprising:
[0008] Obtain medical record data of the patient to be predicted across multiple dimensions;
[0009] Based on the medical record data from the multiple dimensions, a preset disease course data model is used to predict the disease course and obtain the target disease course development direction of the patient to be predicted.
[0010] Based on the target disease progression direction, calculate the predicted cure time for the patient to be predicted.
[0011] Optionally, the step of predicting the course of disease using a preset disease course data model based on the multiple dimensions of medical record data to obtain the target disease progression direction for the patient to be predicted includes:
[0012] Based on the medical record data of the multiple dimensions, the disease course is predicted using the disease course data model corresponding to the multiple dimensions, and the disease course development direction corresponding to the multiple dimensions is obtained.
[0013] Based on the disease progression direction corresponding to the multiple dimensions, the target disease progression direction of the patient to be predicted is obtained.
[0014] Optionally, the step of predicting the course of disease using a preset disease course data model based on the multiple dimensions of medical record data to obtain the target disease progression direction for the patient to be predicted includes:
[0015] A preset regression analysis algorithm is used to fit and analyze the disease progression direction corresponding to the multiple dimensions to obtain the target disease progression direction.
[0016] Optionally, before predicting the course of disease based on the medical record data of the multiple dimensions using the corresponding disease course data models for each of the multiple dimensions, and obtaining the course of disease development direction corresponding to the multiple dimensions, the method further includes:
[0017] Obtain sample data from multiple dimensions, each dimension including: medical records of multiple historically cured patients within the corresponding treatment time period under each dimension;
[0018] Linear fitting is performed on the medical record data of multiple historical cured patients under each dimension within the corresponding treatment time period to obtain the change curve of the medical record data under each dimension within the corresponding treatment time period.
[0019] Based on the change curve, mark the correspondence between medical record data and disease progression direction under each dimension;
[0020] Based on the correspondence of the labels, generate the disease course data model corresponding to each dimension.
[0021] Optionally, the step of performing linear fitting on the medical record data of the multiple historical cured patients under each dimension within the corresponding treatment time period to obtain the change curve of the medical record data under each dimension within the corresponding treatment time period includes:
[0022] The aforementioned patients with a history of cure were grouped to obtain multiple patient groups;
[0023] Based on the medical record data of each patient group under each dimension within the corresponding treatment time period, a linear fit is performed to obtain the change curve of each patient group under each dimension;
[0024] The step of generating the disease course data model corresponding to each dimension based on the marked correspondence includes:
[0025] Based on the change curves of each patient group under each dimension, a disease course data model corresponding to each patient group in each dimension is generated.
[0026] Optionally, the process of grouping the multiple historically cured patients into multiple patient groups includes:
[0027] Based on patient attributes or the stage of the patient's disease at the start of treatment, the multiple historically cured patients are grouped to obtain the multiple patient groups.
[0028] Optionally, the step of predicting the course of disease using a preset disease course data model based on the medical record data from the multiple dimensions to obtain the target disease progression direction for the patient to be predicted further includes:
[0029] Based on the medical record data from the multiple dimensions, a pre-set multi-dimensional disease course data model is used to predict the disease course and obtain the target disease course development direction.
[0030] Optionally, before predicting the disease course using a preset multi-dimensional disease course data model based on the multi-dimensional medical record data to obtain the target disease course development direction, the method further includes:
[0031] Multiple linear regression is performed on the change curves of the disease course data model corresponding to the multiple dimensions to obtain the multidimensional disease course data model.
[0032] Optionally, the multi-dimensional medical record data includes medical record data in at least two of the following dimensions:
[0033] Nursing examination data, blood test data, imaging diagnostic data, and treatment order data.
[0034] Secondly, another embodiment of this application provides a disease course prediction device, the device comprising: an acquisition module, a prediction module, and a calculation module, wherein:
[0035] The acquisition module is used to acquire medical record data of the patient to be predicted in multiple dimensions;
[0036] The prediction module is used to predict the course of a patient's disease based on the medical record data from the multiple dimensions using a preset disease course data model, thereby obtaining the target disease course development direction of the patient to be predicted.
[0037] The calculation module is used to calculate the predicted cure time for the patient to be predicted based on the target disease progression direction.
[0038] Optionally, the prediction module is specifically used to predict the course of the disease based on the medical record data of the multiple dimensions, using the disease course data model corresponding to the multiple dimensions respectively, to obtain the disease course development direction corresponding to the multiple dimensions; and to obtain the target disease course development direction of the patient to be predicted based on the disease course development direction corresponding to the multiple dimensions.
[0039] Optionally, the device further includes a fitting module, used to perform fitting analysis on the disease progression direction corresponding to the multiple dimensions using a preset regression analysis algorithm, so as to obtain the target disease progression direction.
[0040] Optionally, the apparatus further includes: a marking module and a generation module, wherein;
[0041] The acquisition module is specifically used to acquire sample data in multiple dimensions. The sample data in each dimension includes: medical record data of multiple historical cured patients in each dimension within the corresponding treatment time period.
[0042] The fitting module is specifically used to perform linear fitting based on the medical record data of the multiple historical cured patients under each dimension within the corresponding treatment time period, to obtain the change curve of the medical record data under each dimension within the corresponding treatment time period.
[0043] The marking module is used to mark the correspondence between medical record data and disease progression direction in each dimension according to the change curve;
[0044] The generation module is used to generate a disease course data model corresponding to each dimension based on the correspondence of the markings.
[0045] Optionally, the device further includes: a grouping module for grouping the plurality of historically cured patients into multiple patient groups;
[0046] The fitting module is specifically used to perform linear fitting based on the medical record data of each patient group in each dimension within the corresponding treatment time period, so as to obtain the change curve of each patient group in each dimension.
[0047] The generation module is specifically used to generate a disease course data model for each patient group in each dimension based on the change curve of each patient group in each dimension.
[0048] Optionally, the grouping module is specifically used to group the multiple historically cured patients according to patient attributes or the stage of the patient's disease at the start of treatment, thereby obtaining the multiple patient groups.
[0049] Optionally, the prediction module is specifically used to predict the course of the disease based on the medical record data of the multiple dimensions using a preset multi-dimensional disease course data model, so as to obtain the development direction of the target disease course.
[0050] Optionally, the generation module is specifically used to perform multiple linear regression on the change curves of the disease course data model corresponding to the multiple dimensions to obtain the multidimensional disease course data model.
[0051] Optionally, the multi-dimensional medical record data includes medical record data in at least two of the following dimensions:
[0052] Nursing examination data, blood test data, imaging diagnostic data, and treatment order data.
[0053] Thirdly, another embodiment of this application provides a disease course prediction device, including: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the disease course prediction device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of any of the methods described in the first aspect above.
[0054] Fourthly, another embodiment of this application provides a storage medium storing a computer program, which, when executed by a processor, performs the steps of any of the methods described in the first aspect above.
[0055] The beneficial effects of this application are: by using the disease course prediction method provided in this application, a preset disease course data model can be used to predict the target disease course development direction of the patient to be predicted based on the medical record data of the patient in multiple dimensions. This not only improves the prediction efficiency, but also provides data support for generating corresponding treatment plans for the patient to be predicted in the future. Attached Figure Description
[0056] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0057] Figure 1 A schematic flowchart of a disease course prediction method provided in an embodiment of this application;
[0058] Figure 2 A flowchart illustrating a disease course prediction method provided in another embodiment of this application;
[0059] Figure 3 A flowchart illustrating a disease course prediction method provided in another embodiment of this application;
[0060] Figure 4 This is a schematic diagram of the structure of a disease course prediction device provided in an embodiment of this application;
[0061] Figure 5 This is a schematic diagram of the structure of a disease course prediction device provided in another embodiment of this application;
[0062] Figure 6 This is a schematic diagram of the structure of a disease course prediction device provided in an embodiment of this application. Detailed Implementation
[0063] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of this application, but not all embodiments.
[0064] The components of the embodiments of this application described and illustrated in the accompanying drawings can be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of this application provided in the drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0065] Furthermore, the flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed in order or performed simultaneously. Moreover, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.
[0066] The following explanation, using several specific application examples, illustrates a disease course prediction method provided in the embodiments of this application. Figure 1 This is a flowchart illustrating a disease course prediction method provided in one embodiment of this application, as shown below. Figure 1 As shown, the method includes:
[0067] S101: Obtain medical record data of the patient to be predicted across multiple dimensions.
[0068] In one embodiment of this application, the multi-dimensional medical record data includes at least two of the following dimensions: nursing examination data, blood sample test data, imaging diagnostic data, and treatment order data.
[0069] The patient to be predicted can be, for example, a currently hospitalized patient. The method of obtaining the medical record data of the patient to be predicted in multiple dimensions can be, for example, by extracting the medical record data of the patient to be predicted from the systems corresponding to the multiple dimensions. The extraction method can be, for example, extraction based on the identifier of the patient to be predicted. The identifier of the patient to be predicted can be, for example, the hospitalization number of the patient to be predicted, or a document number that can uniquely indicate the identity information of the patient to be predicted. It should be understood that the above embodiments are only illustrative examples. The specific method of extracting the medical record data of the patient to be predicted in multiple dimensions can be flexibly adjusted according to the user's needs and is not limited to the above embodiments.
[0070] S102: Based on medical record data from multiple dimensions, a pre-set disease course data model is used to predict the disease course and obtain the target disease course development direction for the pre-set patient.
[0071] By importing medical record data from multiple dimensions into a pre-defined disease progression data model, the target disease progression direction can be obtained.
[0072] In one embodiment of this application, the method for determining the target disease progression direction may be, for example, by using a preset regression analysis algorithm to perform fitting analysis on the disease progression directions corresponding to multiple dimensions to obtain the target disease progression direction.
[0073] Among them, the preset regression analysis algorithm can be, for example, the maximum likelihood algorithm for preset regression analysis.
[0074] S103: Calculate the predicted cure time for the target patient based on the target disease progression direction.
[0075] Once the target disease progression and predicted cure time for the pre-defined patient are obtained, data support can be provided for subsequent adjustments to the pre-defined patient's treatment plan.
[0076] When the data includes treatment orders across multiple dimensions, to predict the time to recovery, the treatment orders can be modified to compare the predicted times for different treatment orders. Then, based on the patient's current physical condition, the treatment orders that are more suitable for the patient can be selected from the different treatment orders.
[0077] The disease course prediction method provided in this application can be used to predict the target disease course development direction of the patient based on the patient's medical record data in multiple dimensions using a preset disease course data model. This not only has high prediction efficiency, but also provides data support for generating corresponding treatment plans for the patient in the future.
[0078] Optionally, based on the above embodiments, this application embodiment may also provide a disease course prediction method. The following is an example illustration of the process of obtaining the target disease course development direction of the patient to be predicted in the above method, with reference to the accompanying drawings. Figure 2 A flowchart illustrating a disease course prediction method provided in another embodiment of this application is shown below. Figure 2 As shown, S102 may include:
[0079] S111: Based on medical record data from multiple dimensions, disease course prediction is performed using disease course data models corresponding to each dimension, resulting in the disease course development direction corresponding to each dimension.
[0080] S112: Based on the disease progression direction corresponding to multiple dimensions, the target disease progression direction of the patient to be predicted is obtained.
[0081] In one embodiment of this application, the target disease progression direction for the patient to be predicted can be obtained based on the disease progression direction corresponding to multiple dimensions; in another embodiment of this application, the method for determining the target disease progression direction can be, for example, to use a preset multi-dimensional disease progression data model to predict the disease progression based on medical record data of multiple dimensions, thereby obtaining the target disease progression direction.
[0082] Before predicting the course of disease based on the multi-dimensional disease course data model, it is necessary to perform multiple linear regression on the change curves of the disease course data model corresponding to multiple dimensions to obtain the multi-dimensional disease course data model. It should be understood that the above embodiments are merely illustrative examples, and the specific method for predicting the development direction of the target name can be flexibly adjusted according to user needs and is not limited to the methods given in the above embodiments.
[0083] Optionally, based on the above embodiments, this application embodiment may also provide a disease course prediction method. The following is an example illustration of the process of obtaining the target disease course development direction of the patient to be predicted in the above method, with reference to the accompanying drawings. Figure 3 A flowchart illustrating a disease course prediction method provided in another embodiment of this application is shown below. Figure 3 As shown, before S102, the method may further include:
[0084] S121: Obtain sample data from multiple dimensions.
[0085] The sample data for each dimension includes: medical records of multiple historically cured patients within the corresponding treatment period for each dimension.
[0086] S122: Linear fitting is performed on the medical record data of multiple historical cured patients under each dimension within the corresponding treatment time period to obtain the change curve of medical record data under each dimension within the corresponding treatment time period.
[0087] S123: Based on the change curve, mark the correspondence between medical record data and disease progression direction in each dimension.
[0088] S124: Generate the disease course data model for each dimension based on the correspondence of the labels.
[0089] In some possible embodiments, multiple historically cured patients can be grouped to obtain multiple patient groups; linear fitting can be performed on the medical record data of each patient group in each dimension within the corresponding treatment time period to obtain the change curve of each patient group in each dimension; then, based on the change curve of each patient group in each dimension, a disease course data model corresponding to each patient group in each dimension can be generated.
[0090] In the embodiments of this application, the grouping method for grouping multiple historically cured patients can be, for example, grouping multiple historically cured patients according to patient attributes or the stage of the patient's disease at the start of treatment to obtain multiple patient groups.
[0091] For example, in the embodiments of this application, each dimension may also include disease course data corresponding to at least one sub-dimension. If the predicted disease course is different, the corresponding sub-dimension data under each dimension will be different. For example, taking the disease course to be predicted as the disease course corresponding to COVID-19 as an example, the sub-dimension data corresponding to each dimension can be, for example, as follows:
[0092] For data in the nursing examination dimension, body temperature data of multiple historically cured patients can be extracted from the nursing record system. Combined with the patient group and disease stage of each patient in the multiple historically cured patients, the least squares method of linear regression equation is used to form a variation curve that fits each patient group. The relationship between body temperature data and subsequent disease development is marked, and the most likely direction of disease development for each patient group under the single body temperature index dimension is calculated to generate a disease data model for each patient group under the body temperature dimension.
[0093] It can also extract the intake and output data of multiple historically cured patients from the nursing record system, and combine the grouping (patient group), disease stage, and patient weight information of each patient in multiple historically cured patients. According to the least squares method of linear regression equation, a change curve fitting each patient group is formed, marking the relationship between intake and output results and subsequent disease development. The most likely direction of disease development for each patient group under a single intake and output index dimension is calculated, and a disease data model corresponding to each patient group under the intake and output dimension is generated.
[0094] For data in the blood sample testing dimension, the absolute lymphocyte count data of multiple historically cured patients can be extracted from the Laboratory Information Management System (LIS). Combined with the patient group and disease stage of each patient in the multiple historically cured patients, the least squares method of the linear regression equation is used to form a curve fitting the change of each patient group. The relationship between the absolute lymphocyte count result and the subsequent disease development is marked. The most likely direction of disease development for each patient group under the single absolute lymphocyte count index dimension is calculated, and the disease development data model corresponding to each patient group under the absolute lymphocyte count data dimension is generated.
[0095] It is also possible to extract liver enzymes, lactate dehydrogenase, muscle enzymes, myoglobin, troponin, ferritin, C-reactive protein (CRP), and procalcitonin (PCT) data from multiple historically cured patients from the LIS system, and combine the patient groups (patient groups) and disease stages of multiple historically cured patients to form a fitted change curve for each patient group using the least squares method of linear regression equations. The relationship between the results of liver enzymes, lactate dehydrogenase, muscle enzymes, myoglobin, troponin, ferritin, C-reactive protein (CRP), and procalcitonin (PCT) and subsequent disease development is marked, and the most likely direction of disease development for each patient group under a single indicator dimension is calculated, generating disease data models for each patient group under each single dimension.
[0096] It is also possible to extract clinical classification data of multiple historically cured patients into severe and critical types, such as D-dimer, peripheral blood lymphocytes, and inflammatory factors, from the LIS system. Combined with the grouping (patient group) and disease stage of each patient in multiple historically cured patients, the least squares method of linear regression equation is used to form a curve fitting the change of each patient group. The relationship between the results of D-dimer, peripheral blood lymphocytes, and inflammatory factors and the subsequent disease development is marked. The most likely direction of disease development for each patient group under a single indicator dimension is calculated, and the disease data model corresponding to each patient group under each single dimension is generated.
[0097] It can also extract novel coronavirus N gene detection data and RT-PCR detection CT value data from multiple historically cured patients from the LIS system, and combine the patient group (patient group) and disease stage of each patient in multiple historically cured patients, and form a variation curve fitting each patient group according to the least squares method of linear regression equation, mark the relationship between novel coronavirus N gene detection data and RT-PCR detection CT value results and subsequent disease development, calculate the most likely direction of disease development for each patient group under the novel coronavirus N gene detection data and RT-PCR detection CT value index dimension, and generate disease data models for each patient group under each single dimension.
[0098] It can also extract novel coronavirus E gene detection data and RT-PCR detection CT value data from multiple historically cured patients from the LIS system, and combine the patient group (patient group) and disease stage of each patient in multiple historically cured patients, and form a variation curve fitting each patient group according to the least squares method of linear regression equation, mark the relationship between novel coronavirus E gene detection data and RT-PCR detection CT value results and subsequent disease development, calculate the most likely direction of disease development for each patient group under the novel coronavirus E gene detection data and RT-PCR detection CT value index dimension, and generate disease data models for each patient group under each single dimension.
[0099] It can also extract the novel coronavirus RdRP (ORF1ab) gene detection data and RT-PCR detection technology CT value data of multiple historically cured patients from the LIS system, and combine the grouping (patient group) and disease stage of each patient in multiple historically cured patients. According to the least squares method of linear regression equation, a change curve fitting each patient group is formed, marking the relationship between the novel coronavirus RdRP (ORF1ab) gene detection data and RT-PCR detection technology CT value results and subsequent disease development. It calculates the most likely direction of disease development for each patient group under the dimension of novel coronavirus RdRP (ORF1ab) gene detection data and RT-PCR detection technology CT value index, and generates disease data models for each patient group under each single dimension.
[0100] It can also extract arterial oxygen partial pressure (PaO2) and the ratio of arterial oxygen partial pressure to inhaled oxygen concentration (PaO2 / FiO2) data from the LIS system of cured patients, group them according to the preset judgment range, and combine the patient's condition and the stage of the disease course. Based on the subsequent development of the disease course, it calculates the most likely direction of the disease course corresponding to each group and generates the disease course data model corresponding to each patient group under each single dimension.
[0101] For imaging diagnostic data, imaging diagnostic data of multiple historical cured patients can be extracted from the Picture Archiving and Communication System (PACS). Combining patient grouping and disease stage, the imaging diagnostic conditions of patients are grouped according to severity based on the diagnostic description text. Based on the subsequent disease development, the most likely direction of disease development for each group is calculated, generating a disease course data model for each patient group under the imaging diagnostic data dimension.
[0102] For treatment order data, historical medication order data of cured patients can be extracted from the information management CIS system. Based on the types of drugs given in the preset treatment items, and combined with the patient's condition and the stage of the disease, the data can be grouped. After the patient's medication is determined, the most likely direction of disease development for each group can be calculated, and a disease data model corresponding to each patient group under the treatment order data dimension can be generated.
[0103] It should be understood that the above embodiments are merely illustrative examples, and the specific sub-dimensional data included in each dimension are not limited to the above embodiments, but are determined according to the type of disease course to be predicted and the disease course situation.
[0104] The disease progression prediction method provided in this application uses a data extraction and aggregation engine to extract medical record data from multiple dimensions from data sources in different systems. The extracted data is then standardized to ensure uniformity in data format and scope for subsequent steps. Based on the standardized data from multiple dimensions, a data analysis and fitting engine is used to statistically fit single-dimensional data and disease progression and cure time data for historically cured patients. Using the single-dimensional data fitting results as input, the data analysis and fitting engine integrates multiple dimensions and uses maximum likelihood regression analysis to derive the most likely disease progression direction and predicted cure time. Furthermore, when predicting cure time, this application allows for horizontal comparison of the estimated disease progression and cure time of various treatment plans by modifying treatment orders and other methods.
[0105] The disease course prediction device provided in this application will be explained below with reference to the accompanying drawings. This disease course prediction device can perform the above-described... Figures 1-3 The specific implementation and beneficial effects of any disease course prediction method are as described above, and will not be repeated below.
[0106] Figure 4 This is a schematic diagram of the structure of a disease course prediction device provided in an embodiment of this application, as shown below. Figure 4 As shown, the device includes: an acquisition module 201, a prediction module 202, and a calculation module 203, wherein:
[0107] The acquisition module 201 is used to acquire medical record data of the patient to be predicted in multiple dimensions;
[0108] The prediction module 202 is used to predict the course of a patient's disease based on medical record data from multiple dimensions using a preset disease course data model, thereby obtaining the target disease course development direction for the patient to be predicted.
[0109] The calculation module 203 is used to calculate the predicted cure time for the patient to be predicted based on the direction of the target disease progression.
[0110] Optionally, the prediction module 202 is specifically used to predict the course of the disease based on medical record data from multiple dimensions, using disease course data models corresponding to the multiple dimensions respectively, to obtain the disease course development direction corresponding to the multiple dimensions; and to obtain the target disease course development direction of the patient to be predicted based on the disease course development direction corresponding to the multiple dimensions.
[0111] Optionally, based on the above embodiments, this application may also provide a disease course prediction device, as described below with reference to the accompanying drawings. Figure 4 The implementation process of the given device is illustrated with examples. Figure 5 This is a schematic diagram of the structure of a disease course prediction device provided in another embodiment of this application, as shown below. Figure 5 As shown, the device also includes a fitting module 204, which uses a preset regression analysis algorithm to perform fitting analysis on the disease progression direction corresponding to multiple dimensions to obtain the target disease progression direction.
[0112] Optionally, the device further includes: a marking module 205 and a generation module 206, wherein;
[0113] The acquisition module 201 is specifically used to acquire sample data in multiple dimensions. The sample data in each dimension includes: medical record data of multiple historical cured patients in each dimension within the corresponding treatment time period;
[0114] The fitting module 204 is specifically used to perform linear fitting based on the medical record data of multiple historical cured patients in each dimension within the corresponding treatment time period, so as to obtain the change curve of the medical record data in each dimension within the corresponding treatment time period.
[0115] The labeling module 205 is used to label the correspondence between medical record data and disease progression direction in each dimension according to the change curve;
[0116] The generation module 206 is used to generate a disease course data model for each dimension based on the correspondence of the labels.
[0117] Optionally, the device further includes: a grouping module 207 for grouping multiple historically cured patients to obtain multiple patient groups;
[0118] The fitting module 204 is specifically used to perform linear fitting based on the medical record data of each patient group in each dimension during the corresponding treatment time period, so as to obtain the change curve of each patient group in each dimension.
[0119] The generation module 206 is specifically used to generate a disease course data model for each patient group in each dimension based on the change curve of each patient group in each dimension.
[0120] Optionally, the grouping module 207 is specifically used to group multiple historically cured patients according to patient attributes or the stage of the patient's disease at the start of treatment, thereby obtaining multiple patient groups.
[0121] Optionally, the prediction module 202 is specifically used to predict the course of a disease based on medical record data from multiple dimensions using a preset multi-dimensional disease course data model, so as to obtain the target disease course development direction.
[0122] Optionally, the generation module 206 is specifically used to perform multiple linear regression on the change curves of the disease course data model corresponding to multiple dimensions to obtain a multidimensional disease course data model.
[0123] Optionally, multi-dimensional medical record data includes at least two of the following dimensions:
[0124] Nursing examination data, blood test data, imaging diagnostic data, and treatment order data.
[0125] The above-described device is used to execute the method provided in the foregoing embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.
[0126] These modules can be one or more integrated circuits configured to implement the above methods, such as one or more Application Specific Integrated Circuits (ASICs), one or more microprocessors, or one or more Field Programmable Gate Arrays (FPGAs). Alternatively, when a module is implemented using processing element scheduler code, the processing element can be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. Furthermore, these modules can be integrated together as a system-on-a-chip (SOC).
[0127] Figure 6 This is a schematic diagram of the structure of a disease course prediction device provided in an embodiment of this application. The disease course prediction device can be integrated into a terminal device or a chip of a terminal device.
[0128] like Figure 6 As shown, the disease course prediction device includes: processor 501, bus 502 and storage medium 503.
[0129] Processor 501 is used to store programs, and processor 501 calls the programs stored in storage medium 503 to execute the above-mentioned programs. Figures 1-4 The corresponding method implementation is similar in both implementation and technical effect, and will not be described in detail here.
[0130] Optionally, this application also provides a program product, such as a storage medium storing a computer program, including a program that executes the embodiments corresponding to the above-described methods when run by a processor.
[0131] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0132] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0133] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in a combination of hardware and software functional units.
[0134] The integrated units implemented as software functional units described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
Claims
1. A method for predicting the course of a disease, characterized in that, The method includes: Obtain medical record data of the patient to be predicted across multiple dimensions; Obtain sample data from multiple dimensions, each dimension including: medical records of multiple historically cured patients within the corresponding treatment time period under each dimension; Linear fitting is performed on the medical record data of multiple historical cured patients under each dimension within the corresponding treatment time period to obtain the change curve of the medical record data under each dimension within the corresponding treatment time period. Based on the change curve, mark the correspondence between medical record data and disease progression direction under each dimension; Based on the correspondence described in the labels, generate a disease course data model corresponding to each dimension; Based on the medical record data from the multiple dimensions, a preset disease course data model is used to predict the disease course and obtain the target disease course development direction of the patient to be predicted. The step of predicting the disease course using a preset disease course data model based on the multiple dimensions of medical record data to obtain the target disease course development direction of the patient to be predicted includes: Based on the medical record data of the multiple dimensions, the disease course is predicted using the disease course data model corresponding to the multiple dimensions, and the disease course development direction corresponding to the multiple dimensions is obtained. Based on the disease progression direction corresponding to the multiple dimensions, the target disease progression direction of the patient to be predicted is obtained; Based on the target disease progression direction, calculate the predicted cure time for the patient to be predicted.
2. The method according to claim 1, characterized in that, The step of predicting the disease course using a preset disease course data model based on the multiple dimensions of medical record data to obtain the target disease course development direction of the patient to be predicted includes: A preset regression analysis algorithm is used to fit and analyze the disease progression direction corresponding to the multiple dimensions to obtain the target disease progression direction.
3. The method according to claim 1, characterized in that, The step of performing linear fitting on the medical record data of multiple historical cured patients under each dimension within the corresponding treatment time period to obtain the change curve of the medical record data under each dimension within the corresponding treatment time period includes: The aforementioned patients with a history of cure were grouped to obtain multiple patient groups; Based on the medical record data of each patient group under each dimension within the corresponding treatment time period, a linear fit is performed to obtain the change curve of each patient group under each dimension; The step of generating the disease course data model corresponding to each dimension based on the marked correspondence includes: Based on the change curves of each patient group under each dimension, a disease course data model corresponding to each patient group in each dimension is generated.
4. The method according to claim 3, characterized in that, The process of grouping the multiple historically cured patients into multiple patient groups includes: Based on patient attributes or the stage of the patient's disease at the start of treatment, the multiple historically cured patients are grouped to obtain the multiple patient groups.
5. The method according to claim 1, characterized in that, The step of predicting the course of a patient's disease using a pre-defined disease progression data model based on the multi-dimensional medical record data to obtain the target disease progression direction for the patient to be predicted also includes: Based on the medical record data from the multiple dimensions, a pre-set multi-dimensional disease course data model is used to predict the disease course and obtain the target disease course development direction.
6. The method according to claim 5, characterized in that, Before predicting the disease course using a preset multi-dimensional disease course data model based on the multi-dimensional medical record data to obtain the target disease course development direction, the method further includes: Multiple linear regression is performed on the change curves of the disease course data model corresponding to the multiple dimensions to obtain the multidimensional disease course data model.
7. The method according to any one of claims 1-6, characterized in that, The multi-dimensional medical record data includes medical record data in at least two of the following dimensions: Nursing examination data, blood test data, imaging diagnostic data, and treatment order data.
8. A disease course prediction device, characterized in that, The device includes: an acquisition module, a prediction module, and a calculation module, wherein: The acquisition module is used to acquire medical record data of the patient to be predicted in multiple dimensions; The acquisition module is specifically used to acquire sample data in multiple dimensions. The sample data in each dimension includes: medical record data of multiple historical cured patients in each dimension within the corresponding treatment time period. The disease course prediction device further includes: a fitting module, a labeling module, and a generation module; The fitting module is used to perform linear fitting based on the medical record data of the multiple historical cured patients under each dimension within the corresponding treatment time period, to obtain the change curve of the medical record data under each dimension within the corresponding treatment time period. The marking module is used to mark the correspondence between medical record data and disease progression direction in each dimension according to the change curve; The generation module is used to generate a disease course data model corresponding to each dimension based on the marked correspondence. The prediction module is used to predict the course of a patient's disease based on the medical record data from the multiple dimensions using a preset disease course data model, thereby obtaining the target disease course development direction of the patient to be predicted. The prediction module is specifically used to predict the course of a patient's illness based on the medical record data of the multiple dimensions, using the disease course data models corresponding to the multiple dimensions respectively, to obtain the disease course development direction corresponding to the multiple dimensions; and to obtain the target disease course development direction of the patient to be predicted based on the disease course development direction corresponding to the multiple dimensions. The calculation module is used to calculate the predicted cure time for the patient to be predicted based on the target disease progression direction.
9. A disease course prediction device, characterized in that, The device includes a processor, a storage medium, and a bus. The storage medium stores machine-readable instructions executable by the processor. When the disease course prediction device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the method described in any one of claims 1-7.
10. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, performs the method described in any one of claims 1-7.