CN-122025011-A - Operation simulation method, equipment and storage medium based on electronic medical record

CN122025011ACN 122025011 ACN122025011 ACN 122025011ACN-122025011-A

Abstract

The application discloses an operation simulation method, equipment and a storage medium based on an electronic medical record, and belongs to the technical field of data processing. Generating an initial operation scheme according to operation information and three-dimensional electronic medical record and/or clinical detection data of a user through an operation planning model, defining at least two optimization targets according to the operation information, constructing a multi-target optimization problem containing the optimization targets, solving the multi-target optimization problem through a co-evolution algorithm based on a preset search direction and the initial operation scheme to obtain a candidate scheme population, simulating the candidate scheme population, calculating the fitness of the candidate scheme population according to a simulation result, updating the search direction of the co-evolution algorithm according to the candidate scheme population and the fitness, or generating a target operation scheme based on the candidate scheme population. According to the collaborative evolution algorithm, dynamic balance is realized in a plurality of optimization targets of the operation simulation, and the reliability of the operation simulation decision is improved.

Inventors

SHI WENJUAN
JIN LONG
ZHAN JUN

Assignees

深圳市迈加生物医疗有限公司

Dates

Publication Date: 20260512
Application Date: 20260105

Claims (10)

1. An operation simulation method based on electronic medical records is characterized by comprising the following steps: generating an initial operation scheme according to the operation information and the three-dimensional electronic medical record and/or clinical detection data of the user through an operation planning model; Defining at least two optimization targets according to the operation information, and constructing a multi-target optimization problem containing the optimization targets; Solving the multi-objective optimization problem based on a preset searching direction and the initial operation scheme through a co-evolution algorithm to obtain a candidate scheme population; Simulating the candidate scheme population through the three-dimensional electronic medical record, and calculating the adaptability of the candidate scheme population according to a simulation result; updating the search direction of the co-evolution algorithm according to the candidate solution population and the fitness, or generating a target surgical solution based on the candidate solution population.
2. The method of claim 1, wherein the step of generating an initial surgical plan from the surgical information, and the user's three-dimensional electronic medical record and/or clinical test data via the surgical planning model comprises: Converting anatomical structure data in the three-dimensional electronic medical record into graph structure data; Inputting the graph structure data and the operation information into the operation planning model; And calculating and outputting at least one initial operation scheme according to the operation target corresponding to the operation information and the graph structure data through a graph neural network in the operation planning model.
3. The method of claim 1, wherein the step of solving the multi-objective optimization problem by a co-evolution algorithm based on a preset search direction and the initial surgical plan to obtain a candidate plan population comprises: decomposing the multi-objective optimization problem into at least one sub-problem based on a preset decomposition strategy, and constructing a scheme sub-population for the sub-problem; The evolution information of the scheme sub-population is shared through individual migration operation, and the scheme sub-population is evolved based on the preset searching direction and the evolution information; and obtaining the candidate scheme population according to the evolution result.
4. The method of claim 1, wherein the step of modeling the candidate solution population via the three-dimensional electronic medical record and calculating the fitness of the candidate solution population based on the modeling result comprises: Mapping the candidate surgical schemes in the candidate scheme population into a three-dimensional structure model of the three-dimensional electronic medical record, and obtaining a simulation result of the candidate surgical schemes according to deformation information and physiological index change information of the three-dimensional structure model; acquiring performance indexes of the initial operation scheme corresponding to the optimization targets according to the simulation results; sorting the candidate scheme population according to the performance index, and determining the pareto grade of the candidate surgical scheme according to the sorting result; And calculating the crowding degree distance of the candidate surgical scheme in the pareto grade, and determining the fitness according to the crowding degree distance.
5. The method of claim 1, wherein the step of updating the search direction of the co-evolution algorithm or generating a target surgical plan based on the candidate plan population according to the candidate plan population and the fitness comprises: Calculating the convergence of the candidate scheme population according to the fitness and a preset convergence condition corresponding to the operation information; If the convergence is higher than a convergence threshold, generating the target surgical plan according to the candidate plan population; and if the convergence is lower than the convergence threshold, updating the searching direction according to the adaptability distribution information of the candidate surgical schemes in the candidate scheme population.
6. The method of claim 5, wherein the step of generating the target surgical plan from the candidate plan population if the convergence is above a convergence threshold comprises: receiving weight configurations of the at least two optimization targets; Weighting and scoring the candidate surgical schemes in the candidate scheme population according to the weight configuration; And generating the target surgical scheme according to the candidate surgical scheme according to the weighted scoring result.
7. The method of claim 5, wherein if the convergence is below the convergence threshold, updating the search direction based on fitness distribution information of the candidate surgical plan in the candidate plan population comprises: Selecting a target scheme individual from the candidate scheme population according to the fitness; Analyzing the distribution characteristics and performance trends of the target scheme individuals; based on the distribution characteristics and the performance trend, adjusting search parameters of the co-evolution algorithm; And skipping to execute the step of solving the multi-objective optimization problem based on a preset searching direction and the initial operation scheme by the co-evolution algorithm to obtain a candidate scheme population.
8. The method of claim 1, wherein the step of defining at least two optimization objectives from the surgical information and constructing a multi-objective optimization problem that includes the optimization objectives comprises: selecting at least two optimization targets related to the operation targets and the operation risk from a predefined optimization target library according to the operation type in the operation information; quantifying the optimized objective into a mathematical objective function and determining a constraint relationship between the mathematical objective functions; and constructing the multi-objective optimization problem according to the constraint relation.
9. An electronic medical record based surgical simulation device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the electronic medical record based surgical simulation method of any one of claims 1 to 8.
10. A storage medium, characterized in that the storage medium is a computer-readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the steps of the electronic medical record based surgical simulation method according to any one of claims 1 to 8.

Description

Operation simulation method, equipment and storage medium based on electronic medical record Technical Field The present application relates to the field of data processing technologies, and in particular, to a surgical simulation method, apparatus, and storage medium based on electronic medical records. Background Surgical simulation methods typically rely on physician experience or simple computer aided planning, such as generating a surgical plan based on electronic medical record data and fine tuning through fixed rules or single objective optimization, lacking systematic coordination of multiple clinical objectives. However, due to different user requirements and actual circumstances, surgery may have different performance requirements, such as surgery time, trauma minimization and prognostic effects. In such cases, the user's surgical simulation is difficult to consider in practice, resulting in the generated surgical plan possibly being underrepresented in some critical respect, affecting the overall surgical effect. The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present application and is not intended to represent an admission that the foregoing is prior art. Disclosure of Invention The application mainly aims to provide an operation simulation method, equipment and storage medium based on an electronic medical record, and aims to solve the technical problem that the generated operation scheme possibly has insufficient performance in some key aspects and influences the overall operation effect because the actual situation is difficult to consider in operation simulation. In order to achieve the above object, the present application provides a surgical simulation method based on electronic medical records, the method comprising the following steps: generating an initial operation scheme according to the operation information and the three-dimensional electronic medical record and/or clinical detection data of the user through an operation planning model; Defining at least two optimization targets according to the operation information, and constructing a multi-target optimization problem containing the optimization targets; Solving the multi-objective optimization problem based on a preset searching direction and the initial operation scheme through a co-evolution algorithm to obtain a candidate scheme population; Simulating the candidate scheme population through the three-dimensional electronic medical record, and calculating the adaptability of the candidate scheme population according to a simulation result; updating the search direction of the co-evolution algorithm according to the candidate solution population and the fitness, or generating a target surgical solution based on the candidate solution population. In one embodiment, the step of generating the initial surgical plan based on the surgical information and the three-dimensional electronic medical record and/or clinical test data of the user by the surgical planning model includes: Converting anatomical structure data in the three-dimensional electronic medical record into graph structure data; Inputting the graph structure data and the operation information into the operation planning model; And calculating and outputting at least one initial operation scheme according to the operation target corresponding to the operation information and the graph structure data through a graph neural network in the operation planning model. In an embodiment, the step of solving the multi-objective optimization problem by the co-evolution algorithm based on a preset search direction and the initial surgical plan to obtain a candidate plan population includes: decomposing the multi-objective optimization problem into at least one sub-problem based on a preset decomposition strategy, and constructing a scheme sub-population for the sub-problem; The evolution information of the scheme sub-population is shared through individual migration operation, and the scheme sub-population is evolved based on the preset searching direction and the evolution information; and obtaining the candidate scheme population according to the evolution result. In an embodiment, the step of simulating the candidate solution population through the three-dimensional electronic medical record, and calculating the fitness of the candidate solution population according to the simulation result includes: Mapping the candidate surgical schemes in the candidate scheme population into a three-dimensional structure model of the three-dimensional electronic medical record, and obtaining a simulation result of the candidate surgical schemes according to deformation information and physiological index change information of the three-dimensional structure model; acquiring performance indexes of the initial operation scheme corresponding to the optimization targets according to the simulation results; sorting the candidate sche