A method, device, electronic device and storage medium for generating a diagnosis and treatment path

Abstract

The present invention discloses a method, device, electronic device and storage medium for generating a diagnosis and treatment path, and relates to the field of medical informationization. The method comprises: obtaining multimodal diagnosis and treatment data of a diagnosis and treatment object, extracting multimodal feature data of the multimodal diagnosis and treatment data, and performing feature fusion on the multimodal feature data to obtain fused feature data; inputting the fused feature data into an intelligent analysis module for analysis to obtain an intelligent analysis result; obtaining an initial diagnosis and treatment path, and based on the intelligent analysis result, using at least one preset optimization algorithm to perform multi-objective optimization on the initial diagnosis and treatment path to obtain a target diagnosis and treatment path for the diagnosis and treatment object. The present invention performs intelligent analysis on the multimodal diagnosis and treatment data of the diagnosis and treatment object, optimizes the diagnosis and treatment path based on the intelligent analysis result, generates a customized diagnosis and treatment path for the diagnosis and treatment object, optimizes the medical treatment process, and improves the diagnosis and treatment efficiency.

Description

Technical Field

[0001] The present invention relates to the field of medical informationization, and in particular to a diagnosis and treatment path generation method, device, electronic equipment and storage medium. Background Art

[0002] Traditional diagnosis and treatment pathways are usually based on medical guidelines and expert experience. However, this approach is difficult to meet the needs of personalized medicine, and cannot adapt to medical development and changes in patient conditions in a timely manner, lacking the ability to be personalized and dynamically adjusted. Summary of the invention

[0003] The present invention provides a diagnosis and treatment path generation method, device, electronic device and storage medium to solve the above technical problems.

[0004] According to one aspect of the present invention, a method for generating a diagnosis and treatment pathway is provided, comprising:

[0005] Acquiring multimodal diagnosis and treatment data of a diagnosis and treatment object, extracting multimodal feature data of the multimodal diagnosis and treatment data, and performing feature fusion on the multimodal feature data to obtain fused feature data;

[0006] Inputting the fused feature data into an intelligent analysis module for analysis to obtain an intelligent analysis result;

[0007] An initial diagnosis and treatment pathway is obtained, and based on the intelligent analysis result, at least one preset optimization algorithm is used to perform multi-objective optimization on the initial diagnosis and treatment pathway to obtain a target diagnosis and treatment pathway for the diagnosis and treatment object.

[0008] According to another aspect of the present invention, there is provided a diagnosis and treatment path generating device, comprising:

[0009] A multimodal diagnosis and treatment data fusion module, used to obtain multimodal diagnosis and treatment data of a diagnosis and treatment object, extract multimodal feature data of the multimodal diagnosis and treatment data, and perform feature fusion on the multimodal feature data to obtain fused feature data;

[0010] An intelligent analysis module, used for inputting the fused feature data into the intelligent analysis module for analysis to obtain an intelligent analysis result;

[0011] The diagnosis and treatment pathway optimization module is used to obtain an initial diagnosis and treatment pathway, and based on the intelligent analysis results, use at least one preset optimization algorithm to perform multi-objective optimization on the initial diagnosis and treatment pathway to obtain a target diagnosis and treatment pathway for the diagnosis and treatment object.

[0012] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0013] at least one processor; and

[0014] a memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the diagnosis and treatment path generation method described in any embodiment of the present invention.

[0016] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the diagnosis and treatment pathway generation method described in any embodiment of the present invention when executed.

[0017] The technical solution of the embodiment of the present invention obtains the multimodal diagnosis and treatment data of the diagnosis and treatment object, extracts the multimodal feature data of the multimodal diagnosis and treatment data, and performs feature fusion on the multimodal feature data to obtain fused feature data; inputs the fused feature data into the intelligent analysis module for analysis to obtain the intelligent analysis result; obtains the initial diagnosis and treatment path, and based on the intelligent analysis result, uses at least one preset optimization algorithm to perform multi-objective optimization on the initial diagnosis and treatment path to obtain the target diagnosis and treatment path of the diagnosis and treatment object. By performing intelligent analysis on the multimodal diagnosis and treatment data of the diagnosis and treatment object, the diagnosis and treatment path is optimized based on the intelligent analysis result, and a customized diagnosis and treatment path for the diagnosis and treatment object is generated, the medical process is optimized, and the diagnosis and treatment efficiency is improved.

[0018] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 is a flow chart of a method for generating a diagnosis and treatment pathway provided in Embodiment 1 of the present invention;

[0021] Figure 2 This is an example diagram of a visual display of a diagnosis and treatment pathway provided by the first embodiment of the present invention;

[0022] Figure 3 It is a structural schematic diagram of a diagnosis and treatment path generating device provided in Embodiment 2 of the present invention;

[0023] Figure 4 It is a structural schematic diagram of an electronic device provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first target diagnosis and treatment pathway", "second target diagnosis and treatment pathway", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention 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 inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] Embodiment 1

[0027] Figure 1 This is a flow chart of a method for generating a diagnosis and treatment path provided by the first embodiment of the present invention. This embodiment is applicable to the case of generating a personalized diagnosis and treatment path for a patient. The method can be executed by a diagnosis and treatment path generating device. The diagnosis and treatment path generating device can be implemented in the form of hardware and / or software. The diagnosis and treatment path generating device can be configured in electronic devices such as computers and servers. Figure 1 As shown, the method includes:

[0028] S110, acquiring multimodal diagnosis and treatment data of a diagnosis and treatment object, extracting multimodal feature data of the multimodal diagnosis and treatment data, and performing feature fusion on the multimodal feature data to obtain fused feature data.

[0029] Among them, multimodal diagnosis and treatment data include but are not limited to imaging data (such as X-ray, CT, MRI scans), physiological parameters (such as heart rate, blood pressure), biochemical indicators (such as blood test results), medical history records, etc. In an embodiment of the present invention, multimodal diagnosis and treatment data of the diagnosis and treatment object can be collected from data sources such as medical imaging equipment, physiological monitoring equipment, and electronic medical record systems. For diagnosis and treatment data of different modalities, feature data can be extracted through corresponding feature extraction models. Specifically, pre-trained image encoders can be used to extract image features. The image encoder can be ResNet, EfficientNet, VisionTransformer, etc. A pre-trained text encoder can be used to extract text features. The text encoder can be BERT (Bidirectional Encoder Representations from Transformers), RoBERTa (RobustlyOptimized BERT Pretraining Approach), GPT (Generative Pre-trained Transformer) series, etc. A time series data processing model can be used to extract time series features. The time series data processing model can be a long short-term memory network (STM), a gated recurrent unit (GRU), a temporal convolutional network (TCN), etc. After extracting the multimodal feature data, the multimodal feature data is subjected to feature fusion to obtain fused feature data.

[0030] In some embodiments, after acquiring the multimodal diagnosis and treatment data, the multimodal diagnosis and treatment data is preprocessed. Specifically, the multimodal diagnosis and treatment data can be subjected to data cleaning, data standardization, data structuring and other processing operations. Data cleaning refers to the use of natural language processing technology to process unstructured text data, and to screen, remove redundancy, and correct errors (duplicate, redundant, and missing data) of the multimodal diagnosis and treatment data. Data standardization refers to the use of unified medical terminology standards to standardize multimodal diagnosis and treatment data, wherein the medical terminology standards may be SNOMED CT (Systematized Nomenclature of Medicine-Clinical Terms), ICD-10 (International Classification of Diseases, Tenth Revision), etc. Data structuring refers to converting the processed diagnosis and treatment data into a structured format based on a preset medical interoperability standard. Exemplarily, the preset medical interoperability standard may be the FHIR (Fast Healthcare Interoperability Resources) standard.

[0031] On the basis of the above embodiment, optionally, the multimodal feature data is subjected to feature fusion to obtain fused feature data, including: inputting the multimodal feature data into a multi-head attention layer for fusion to obtain fused feature data.

[0032] Specifically, in the feature fusion process, feature data of different modalities can be used as input to the multi-head attention layer. Specifically, the feature representation of feature data of different modalities is first converted into corresponding query, key, and value vectors, and then the attention weight is calculated through the interaction between these vectors. Finally, the value vector is weighted and summed according to the attention weight to obtain the feature representation of the fused feature data. It can be understood that the multi-head attention mechanism of the multi-head attention layer allows the model to focus on different parts of the input data at the same time, thereby capturing richer information.

[0033] S120: Input the fused feature data into an intelligent analysis module for analysis to obtain an intelligent analysis result.

[0034] In an embodiment of the present invention, the fused feature data can be input into an intelligent analysis module for analysis to obtain an intelligent analysis result. The intelligent analysis module can be a machine learning model, a deep learning network, etc., which is used to identify disease types, predict disease risks, evaluate treatment effects, etc. Specifically, the intelligent analysis module includes but is not limited to a recommendation model based on a graph neural network, a causal reasoning model, a reasoning model based on a knowledge graph, etc.

[0035] S130, obtaining an initial diagnosis and treatment pathway, and based on the intelligent analysis result, using at least one preset optimization algorithm to perform multi-objective optimization on the initial diagnosis and treatment pathway to obtain a target diagnosis and treatment pathway for the diagnosis and treatment object.

[0036] Among them, the initial diagnosis and treatment path refers to a preliminary diagnosis and treatment path developed based on medical guidelines and expert experience. The initial diagnosis and treatment path includes multiple diagnosis and treatment nodes. Specifically, the diagnosis and treatment nodes include but are not limited to appointment registration, outpatient reporting, initial diagnosis, outpatient payment, blood sampling, follow-up visits and other nodes. In an embodiment of the present invention, the initial diagnosis and treatment path is obtained, and the initial diagnosis and treatment path is optimized for multiple objectives using at least one preset optimization algorithm to obtain a target diagnosis and treatment path. The intelligent analysis results are considered in the optimization process to ensure that the target diagnosis and treatment path obtained not only meets the actual needs of the diagnosis and treatment object, but also achieves the optimal balance in multiple aspects. Among them, the optimization goals include but are not limited to improving the treatment effect, reducing the side effects of treatment, reducing medical costs, shortening the treatment time, etc. The optimization algorithms include but are not limited to A* algorithm, genetic algorithm, particle swarm optimization algorithm, simulated annealing algorithm, etc.

[0037] In some embodiments, the target diagnosis and treatment pathway obtained can be visualized in the form of a diagnosis and treatment timeline to guide the diagnosis and treatment object to see a doctor, and a corresponding service entrance can be linked to each diagnosis and treatment node of the target diagnosis and treatment pathway to facilitate user access. Figure 2 This is an example diagram of a visual display of a diagnosis and treatment pathway provided in Example 1 of the present invention.

[0038] Based on the above embodiment, optionally, obtaining the initial diagnosis and treatment pathway includes: obtaining the initial diagnosis and treatment pathway in response to a configuration operation on a diagnosis and treatment pathway configuration interface; wherein the configuration operation includes a diagnosis and treatment institution selection operation, a diagnosis and treatment scope selection operation, and a diagnosis and treatment node selection operation.

[0039] In an embodiment of the present invention, in the diagnosis and treatment path configuration interface, multiple diagnosis and treatment institutions are first displayed, and in response to the diagnosis and treatment institution selection operation, the diagnosis and treatment scope corresponding to the selected diagnosis and treatment institution is displayed; wherein, the diagnosis and treatment institutions include hospitals, clinics or other medical facilities. Further, in response to the diagnosis and treatment scope selection operation, the diagnosis and treatment nodes under the diagnosis and treatment scope are displayed; wherein, the diagnosis and treatment scope refers to the scope of diagnosis and treatment services that the diagnosis and treatment institution can provide. It can be understood that for general hospitals, this scope may be very broad, covering the treatment of various diseases and symptoms, while for specialized hospitals, the diagnosis and treatment scope is more focused on a specific field, such as cardiology, neurology or pediatrics. The user needs to select an appropriate diagnosis and treatment scope based on the specific needs of the diagnosis and treatment object and the expertise of the selected diagnosis and treatment institution. Further, in response to the diagnosis and treatment node selection operation, an initial diagnosis and treatment path is obtained. Wherein, the initial diagnosis and treatment path is composed of the selected diagnosis and treatment nodes.

[0040] On the basis of the above embodiment, optionally, the initial diagnosis and treatment path includes multiple diagnosis and treatment nodes; based on the intelligent analysis results, the initial diagnosis and treatment path is optimized using at least one preset optimization algorithm to obtain a target diagnosis and treatment path, including: constructing a search space based on the multiple diagnosis and treatment nodes of the initial diagnosis and treatment path; in the search space, iteratively expanding the diagnosis and treatment nodes based on the intelligent analysis results and a preset cost function, and obtaining a first target diagnosis and treatment path when the target diagnosis and treatment node is searched; wherein the preset cost function is constructed based on multiple optimization objectives.

[0041] In an embodiment of the present invention, a search space is constructed based on multiple diagnosis and treatment nodes of the initial diagnosis and treatment path. Among them, the search space can be a graph or tree structure, the diagnosis and treatment nodes represent the diagnosis and treatment steps, and the edges represent the transfer or dependency between the steps. It should be noted that each diagnosis and treatment node in the search space contains information related to the diagnosis and treatment steps corresponding to the diagnosis and treatment node, such as the severity of the disease, treatment priority, department congestion, doctor expertise, etc. Furthermore, in the search space, taking the starting diagnosis and treatment node of the initial diagnosis and treatment path as the starting point, the diagnosis and treatment nodes are iteratively expanded based on the intelligent analysis results and the preset cost function until the target diagnosis and treatment node is searched. If the target diagnosis and treatment node is searched, the parent node pointer is traced back from the target diagnosis and treatment node to the starting point diagnosis and treatment node to form the first target diagnosis and treatment path. Among them, the first target diagnosis and treatment node can be the node where the diagnosis and treatment object is discharged from the hospital after recovery.

[0042] The preset cost function includes an actual cost function, a heuristic function, and a node travel cost coefficient. The actual cost function is used to reflect the actual cost from the starting diagnosis and treatment node to the current diagnosis and treatment node, and the heuristic function is used to reflect the potential cost from the current diagnosis and treatment node to the target diagnosis and treatment node. Exemplarily, the preset cost function is as follows:

[0043] f(n)=g(n)+h(n)+W*c(n);

[0044] h(n)=w1*h1(n)+w2*h2(n)+...+wk*hk(n);

[0045] Among them, f(n) represents the preset cost function, g(n) represents the heuristic function, c(n) is the node travel cost coefficient, which indicates the difficulty of passing through the diagnosis and treatment node. c(n) can be a constant or a variable related to the node status, such as department congestion, doctor expertise, etc. In this way, when selecting a node, not only the distance is considered, but also the convenience of travel. W represents the weight of the node travel cost coefficient, which is used to adjust the influence of the node travel cost coefficient in the preset cost function. h1(n), h2(n), ..., hk(n) are heuristic functions of different optimization objectives, and w1, w2, ..., wk are the weights corresponding to the heuristic functions of each optimization objective.

[0046] In an embodiment of the present invention, optionally, based on the intelligent analysis result, at least one preset optimization algorithm is used to optimize the initial treatment pathway to obtain a target treatment pathway, including: generating an initial treatment pathway population based on the initial treatment pathway or the first target treatment pathway, the initial treatment pathway population including multiple treatment pathways; for any of the treatment pathways, performing a quality assessment on the treatment pathway based on a preset fitness function to obtain the quality of the treatment pathway; wherein the preset fitness function is constructed based on multiple optimization objectives; performing selection operations, crossover operations, and mutation operations on the treatment pathways in the initial treatment pathway population based on the quality of the treatment pathway to obtain an updated treatment pathway population; and obtaining a second target treatment pathway when preset optimization conditions are met.

[0047] Among them, the second target diagnosis and treatment path can be a diagnosis and treatment path after optimizing the initial diagnosis and treatment path, or it can be a diagnosis and treatment path after further optimizing the first target diagnosis and treatment path. In an embodiment of the present invention, the initial diagnosis and treatment path can be optimized, and the first target diagnosis and treatment path can be further optimized to obtain a second target diagnosis and treatment path. Specifically, a random generation method, a method based on historical data generation, or a combination of random generation and historical data generation can be used to generate an initial diagnosis and treatment path population based on the initial diagnosis and treatment path or the first target diagnosis and treatment path; for each diagnosis and treatment path in the population, a preset fitness function is used to perform quality assessment. Among them, the fitness function is constructed based on multiple optimization objectives, and the optimization objectives include but are not limited to improving diagnosis and treatment effects, reducing medical costs, reducing diagnosis and treatment time, etc. Exemplarily, the preset fitness function is:

[0048]

[0049] Among them, F(P) is the value of the fitness function, and C(P) represents the total cost of path P, including multiple optimization objectives.

[0050] It is understandable that the output value of the preset fitness function (i.e., the quality of the treatment pathway) will be used to guide subsequent genetic operations such as selection, crossover, and mutation.

[0051] The genetic operations are as follows:

[0052] Selection operation: According to the quality of the diagnosis and treatment pathways, some excellent diagnosis and treatment pathways are selected as parents for subsequent crossover and mutation operations. The selection operation includes but is not limited to roulette selection, tournament selection and other strategies.

[0053] Exemplarily, the probability calculation formula for each diagnosis and treatment pathway being selected is:

[0054]

[0055] Among them, Pi is the probability that the i-th path is selected, and N is the population size.

[0056] Crossover operation: Pair the selected parent diagnosis and treatment paths, and exchange some information through a preset crossover method to generate a new child diagnosis and treatment path. The crossover operation aims to combine the advantages of the parent diagnosis and treatment path to produce a better child. The preset crossover methods include but are not limited to single-point crossover, double-point crossover, uniform crossover, etc. For example, suppose we have two paths P1 and P2, and the crossover operation generates new paths P′1 and P′2. Single-point crossover can be expressed as:

[0057] P'1=(P1[1:k], P2[k+1:end]);

[0058] P'2=(P2[1:k], P1[k+1:end]);

[0059] Here, k is the location of the intersection point.

[0060] Mutation operation: modify the offspring diagnosis and treatment pathway randomly or based on a preset strategy to increase the diversity of the population and explore new solutions. Mutation operations include but are not limited to randomly changing any diagnosis and treatment node in the diagnosis and treatment pathway, adjusting the order of diagnosis and treatment nodes, etc. For example, mutate the diagnosis and treatment pathway P, assuming that the mutation point is m, and the mutated path P′ is expressed as:

[0061] P′=(P[1:m-1],new state , P[m+1:end]);

[0062] Among them, new state is a randomly generated new state.

[0063] Through fork selection and mutation operations, an updated treatment path population is obtained. The updated treatment path population is used as the input for the next round of iteration to continue optimization. After each iteration, check whether the preset optimization conditions are met. If the preset optimization conditions are met, the iteration is stopped, and the current optimal treatment path is output as the second target treatment path. Among them, the preset optimization conditions include but are not limited to reaching the maximum number of iterations, finding a treatment path that meets specific quality standards, and the quality of the population is no longer significantly improved.

[0064] In some embodiments, optionally, the method also includes: acquiring the diagnosis and treatment data of the diagnosis and treatment object during the diagnosis and treatment process in real time; dynamically adjusting the target diagnosis and treatment path based on the diagnosis and treatment data and preset diagnosis and treatment path adjustment rules; or, inputting the diagnosis and treatment data into a diagnosis and treatment data prediction model for prediction to obtain predicted diagnosis and treatment data; and dynamically adjusting the target diagnosis and treatment path based on the predicted diagnosis and treatment data.

[0065] Among them, the diagnosis and treatment data of the diagnosis and treatment object include but are not limited to vital signs (such as heart rate, blood pressure), laboratory test results, imaging data, patient feedback, etc. In an embodiment of the present invention, during the diagnosis and treatment process, the diagnosis and treatment data of the diagnosis and treatment object are collected and recorded in real time, and the diagnosis and treatment data of the diagnosis and treatment object can be matched with the diagnosis and treatment pathway adjustment rules. If the trigger conditions in the diagnosis and treatment pathway adjustment rules are met, the target diagnosis and treatment pathway is adjusted. Among them, the preset diagnosis and treatment pathway adjustment rules can be based on clinical guidelines, expert consensus or historical data statistical analysis. If the data deviates from the preset normal range or expected trend, the system will automatically or prompt the doctor to adjust the diagnosis and treatment pathway. Exemplarily, for a diagnosis and treatment object with diabetes, if the fasting blood sugar is higher than 11.1mmol / L for three consecutive times, the insulin dose is increased; if symptoms of hypoglycemia occur, the insulin dose is reduced.

[0066] Optionally, the diagnosis and treatment data can also be input into the prediction model to obtain the predicted diagnosis and treatment data, and the target diagnosis and treatment pathway can be dynamically adjusted based on the predicted diagnosis and treatment data. The prediction model can be based on machine learning or deep learning algorithms, obtained through training with historical data, and used to predict the future health status or diagnosis and treatment needs of the diagnosis and treatment object. The adjustment content of the target diagnosis and treatment pathway includes, but is not limited to, adjusting drug dosage, treatment plan, examination items, etc.

[0067] In some embodiments, optionally, the method further includes: adjusting the target diagnosis and treatment pathway in response to a diagnosis and treatment pathway adjustment operation at the review end.

[0068] Among them, the diagnosis and treatment pathway adjustment operation refers to the manual intervention operation of the auditor at the audit end. Specifically, the diagnosis and treatment pathway adjustment operation includes but is not limited to the order adjustment operation, deletion operation, editing operation, etc. of the diagnosis and treatment nodes, which are not limited here. It is understandable that the auditor can be a professional medical staff. In the embodiment of the present invention, the target diagnosis and treatment pathway can be adjusted through the manual intervention mechanism at the audit end. Specifically, the target diagnosis and treatment pathway is adjusted in response to the diagnosis and treatment pathway adjustment operation at the audit end.

[0069] The technical solution of this embodiment obtains the multimodal diagnosis and treatment data of the diagnosis and treatment object, extracts the multimodal feature data of the multimodal diagnosis and treatment data, and performs feature fusion on the multimodal feature data to obtain fused feature data; inputs the fused feature data into the intelligent analysis module for analysis to obtain intelligent analysis results; obtains the initial diagnosis and treatment path, and based on the intelligent analysis results, uses at least one preset optimization algorithm to perform multi-objective optimization on the initial diagnosis and treatment path to obtain the target diagnosis and treatment path of the diagnosis and treatment object. By performing intelligent analysis on the multimodal diagnosis and treatment data of the diagnosis and treatment object, the diagnosis and treatment path is optimized based on the intelligent analysis results, and a customized diagnosis and treatment path for the diagnosis and treatment object is generated, the medical process is optimized, and the diagnosis and treatment efficiency is improved.

[0070] Embodiment 2

[0071] Figure 3 Schematic diagram of the structure of a diagnosis and treatment path generation device provided by Embodiment 2 of the present invention. Figure 3 As shown, the device comprises:

[0072] The multimodal diagnosis and treatment data fusion module 310 is used to obtain the multimodal diagnosis and treatment data of the diagnosis and treatment object, extract the multimodal feature data of the multimodal diagnosis and treatment data, and perform feature fusion on the multimodal feature data to obtain fused feature data;

[0073] An intelligent analysis module 320 is used to input the fused feature data into the intelligent analysis module for analysis to obtain an intelligent analysis result;

[0074] The diagnosis and treatment pathway optimization module 330 is used to obtain an initial diagnosis and treatment pathway, and based on the intelligent analysis results, use at least one preset optimization algorithm to perform multi-objective optimization on the initial diagnosis and treatment pathway to obtain a target diagnosis and treatment pathway for the diagnosis and treatment object.

[0075] The technical solution of this embodiment obtains the multimodal diagnosis and treatment data of the diagnosis and treatment object, extracts the multimodal feature data of the multimodal diagnosis and treatment data, and performs feature fusion on the multimodal feature data to obtain fused feature data; inputs the fused feature data into the intelligent analysis module for analysis to obtain intelligent analysis results; obtains the initial diagnosis and treatment path, and based on the intelligent analysis results, uses at least one preset optimization algorithm to perform multi-objective optimization on the initial diagnosis and treatment path to obtain the target diagnosis and treatment path of the diagnosis and treatment object. By performing intelligent analysis on the multimodal diagnosis and treatment data of the diagnosis and treatment object, the diagnosis and treatment path is optimized based on the intelligent analysis results, and a customized diagnosis and treatment path for the diagnosis and treatment object is generated, the medical process is optimized, and the diagnosis and treatment efficiency is improved.

[0076] On the basis of the above embodiment, optionally, the multimodal diagnosis and treatment data fusion module 310 includes a feature fusion unit for inputting the multimodal feature data into a multi-head attention layer for fusion to obtain fused feature data.

[0077] On the basis of the above embodiment, optionally, the diagnosis and treatment pathway optimization module 330 includes an initial diagnosis and treatment pathway acquisition unit, which is used to obtain an initial diagnosis and treatment pathway in response to a configuration operation on a diagnosis and treatment pathway configuration interface; wherein the configuration operation includes a diagnosis and treatment institution selection operation, a diagnosis and treatment scope selection operation, and a diagnosis and treatment node selection operation.

[0078] On the basis of the above embodiment, optionally, the initial diagnosis and treatment path includes multiple diagnosis and treatment nodes; the diagnosis and treatment path optimization module 330 is specifically used to construct a search space based on the multiple diagnosis and treatment nodes of the initial diagnosis and treatment path; in the search space, the diagnosis and treatment nodes are iteratively expanded based on the intelligent analysis results and the preset cost function, and when the target diagnosis and treatment node is searched, a first target diagnosis and treatment path is obtained; wherein, the preset cost function is constructed based on multiple optimization objectives.

[0079] On the basis of the above embodiments, optionally, the diagnosis and treatment path optimization module 330 is also used to generate an initial diagnosis and treatment path population based on the initial diagnosis and treatment path or the first target diagnosis and treatment path, and the initial diagnosis and treatment path population includes multiple diagnosis and treatment paths; for any of the diagnosis and treatment paths, the quality of the diagnosis and treatment path is evaluated based on a preset fitness function to obtain the quality of the diagnosis and treatment path; wherein the preset fitness function is constructed based on multiple optimization objectives; based on the quality of the diagnosis and treatment path, the diagnosis and treatment paths in the initial diagnosis and treatment path population are selected, crossed and mutated to obtain an updated diagnosis and treatment path population; when the preset optimization conditions are met, a second target diagnosis and treatment path is obtained.

[0080] On the basis of the above-mentioned embodiment, optionally, the device also includes a dynamic adjustment module, which is used to obtain the diagnosis and treatment data of the diagnosis and treatment object during the diagnosis and treatment process in real time; dynamically adjust the target diagnosis and treatment path based on the diagnosis and treatment data and preset diagnosis and treatment path adjustment rules; or, input the diagnosis and treatment data into a diagnosis and treatment data prediction model for prediction to obtain predicted diagnosis and treatment data; and dynamically adjust the target diagnosis and treatment path based on the predicted diagnosis and treatment data.

[0081] On the basis of the above embodiment, optionally, the device further includes a manual adjustment module for adjusting the target diagnosis and treatment pathway in response to a diagnosis and treatment pathway adjustment operation at the review end.

[0082] The diagnosis and treatment path generation device provided in the embodiment of the present invention can execute the diagnosis and treatment path generation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0083] Embodiment 3

[0084] Figure 41 is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0085] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0086] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0087] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 executes the various methods and processes described above, such as a diagnosis and treatment path generation method.

[0088] In some embodiments, the diagnosis and treatment path generation method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the diagnosis and treatment path generation method described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the diagnosis and treatment path generation method in any other appropriate manner (for example, by means of firmware).

[0089] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0090] The computer program for implementing the diagnosis and treatment path generation method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0091] Embodiment 4

[0092] Embodiment 4 of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to cause a processor to execute a diagnosis and treatment pathway generation method, the method comprising:

[0093] Acquiring multimodal diagnosis and treatment data of a diagnosis and treatment object, extracting multimodal feature data of the multimodal diagnosis and treatment data, and performing feature fusion on the multimodal feature data to obtain fused feature data;

[0094] Inputting the fused feature data into an intelligent analysis module for analysis to obtain an intelligent analysis result;

[0095] An initial diagnosis and treatment pathway is obtained, and based on the intelligent analysis result, at least one preset optimization algorithm is used to perform multi-objective optimization on the initial diagnosis and treatment pathway to obtain a target diagnosis and treatment pathway for the diagnosis and treatment object.

[0096] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0097] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0098] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0099] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0100] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0101] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for generating a diagnosis and treatment pathway, characterized in that: include: Acquiring multimodal diagnosis and treatment data of a diagnosis and treatment object, extracting multimodal feature data of the multimodal diagnosis and treatment data, and performing feature fusion on the multimodal feature data to obtain fused feature data; Inputting the fused feature data into an intelligent analysis module for analysis to obtain an intelligent analysis result; An initial diagnosis and treatment pathway is obtained, and based on the intelligent analysis result, at least one preset optimization algorithm is used to perform multi-objective optimization on the initial diagnosis and treatment pathway to obtain a target diagnosis and treatment pathway for the diagnosis and treatment object.

2. The method according to claim 1, characterized in that: The step of fusing the multimodal feature data to obtain fused feature data includes: The multimodal feature data is input into the multi-head attention layer for fusion to obtain fused feature data.

3. The method according to claim 1, characterized in that The obtaining of the initial diagnosis and treatment pathway includes: In response to a configuration operation on a diagnosis and treatment pathway configuration interface, an initial diagnosis and treatment pathway is obtained; Among them, the configuration operation includes a diagnosis and treatment institution selection operation, a diagnosis and treatment scope selection operation and a diagnosis and treatment node selection operation.

4. The method according to claim 1, characterized in that The initial diagnosis and treatment pathway includes a plurality of diagnosis and treatment nodes; based on the intelligent analysis result, the initial diagnosis and treatment pathway is optimized using at least one preset optimization algorithm to obtain a target diagnosis and treatment pathway, including: Constructing a search space based on multiple diagnosis and treatment nodes of the initial diagnosis and treatment path; In the search space, the diagnosis and treatment nodes are iteratively expanded based on the intelligent analysis results and the preset cost function, and when the target diagnosis and treatment node is searched, the first target diagnosis and treatment path is obtained; wherein the preset cost function is constructed based on multiple optimization objectives.

5. The method according to claim 1 or 4, characterized in that: Based on the intelligent analysis result, the initial diagnosis and treatment pathway is optimized using at least one preset optimization algorithm to obtain a target diagnosis and treatment pathway, including: Generate an initial diagnosis and treatment path population based on the initial diagnosis and treatment path or the first target diagnosis and treatment path, wherein the initial diagnosis and treatment path population includes multiple diagnosis and treatment paths; For any of the diagnosis and treatment pathways, a quality assessment is performed on the diagnosis and treatment pathway based on a preset fitness function to obtain the quality of the diagnosis and treatment pathway; wherein the preset fitness function is constructed based on multiple optimization objectives; Based on the quality of the diagnosis and treatment pathways, selection operations, crossover operations, and mutation operations are performed on the diagnosis and treatment pathways in the initial diagnosis and treatment pathway population to obtain an updated diagnosis and treatment pathway population; When the preset optimization conditions are met, the second target diagnosis and treatment pathway is obtained.

6. The method according to claim 5, characterized in that The method further comprises: Obtain the diagnosis and treatment data of the diagnosis and treatment object in real time during the diagnosis and treatment process; Dynamically adjusting the target diagnosis and treatment pathway based on the diagnosis and treatment data and preset diagnosis and treatment pathway adjustment rules; Alternatively, the diagnosis and treatment data are input into a diagnosis and treatment data prediction model for prediction to obtain predicted diagnosis and treatment data; and the target diagnosis and treatment pathway is dynamically adjusted based on the predicted diagnosis and treatment data.

7. The method according to claim 1, characterized in that The method further comprises: In response to the diagnosis and treatment pathway adjustment operation at the review end, the target diagnosis and treatment pathway is adjusted.

8. A diagnosis and treatment pathway generation device, characterized in that: include: A multimodal diagnosis and treatment data fusion module, used to obtain multimodal diagnosis and treatment data of a diagnosis and treatment object, extract multimodal feature data of the multimodal diagnosis and treatment data, and perform feature fusion on the multimodal feature data to obtain fused feature data; An intelligent analysis module, used for inputting the fused feature data into the intelligent analysis module for analysis to obtain an intelligent analysis result; The diagnosis and treatment pathway optimization module is used to obtain an initial diagnosis and treatment pathway, and based on the intelligent analysis results, use at least one preset optimization algorithm to perform multi-objective optimization on the initial diagnosis and treatment pathway to obtain a target diagnosis and treatment pathway for the diagnosis and treatment object.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the diagnosis and treatment pathway generation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the diagnosis and treatment pathway generation method according to any one of claims 1 to 7 when executed.

Images