CN-122000005-A - Information planning crowd health record management system

CN122000005ACN 122000005 ACN122000005 ACN 122000005ACN-122000005-A

Abstract

The invention relates to the technical field of medical data processing and health information management, in particular to an informationized planning crowd health archive management system which comprises an archive data aggregation module, a physiological benchmark reconstruction module, a pathological disturbance simulation module, a double-track differential extraction module, a coupling research judgment module and a dynamic intervention module, wherein the archive data aggregation module is used for acquiring multidimensional health records and retrieving knowledge maps, the physiological benchmark reconstruction module is used for constructing ideal physiological benchmarks, the pathological disturbance simulation module is used for generating pathological simulation states, the double-track differential extraction module is used for respectively calculating time sequence sign data, deviation of the pathological simulation states and the ideal benchmarks, acquiring actual and theoretical residual sequences, the coupling research judgment module is used for comparing waveform morphology similarity of the residual sequences to judge abnormal properties, the dynamic intervention module is used for generating a grading intervention instruction if judging effective health risks, and executing data filtering if judging environmental noise, and the system also comprises an adaptive correction module which utilizes non-pathological samples to finely tune construction parameters of the ideal physiological benchmarks.

Inventors

PAN JIANBO
LIU FAPENG
ZHANG WEIBIN
HOU YING
YAN KUN

Assignees

北京啄木鸟云健康科技有限公司
重庆市卫生健康委员会

Dates

Publication Date: 20260508
Application Date: 20260202

Claims (9)

1. The utility model provides an informationized planning crowd health record management system which characterized in that includes: The archive data aggregation module is used for carrying out multidimensional collection on the historical health record of the target object, acquiring time sequence sign data and static physiological attribute data, and calling a preset medical knowledge graph; The physiological reference reconstruction module is used for constructing a dynamic health track of the target object in an ideal compliance state based on the static physiological attribute data and the medical knowledge graph to serve as an ideal physiological reference; The pathological disturbance simulation module is used for defining a pathological disturbance function based on a medical knowledge graph, superposing the pathological disturbance function to an ideal physiological reference and generating a pathological simulation state containing theoretical pathological features; the double-track differential extraction module is used for calculating the deviation between the time sequence sign data and the ideal physiological reference, obtaining a real residual sequence, calculating the deviation between the pathological simulation state and the ideal physiological reference and obtaining a theoretical residual sequence; The coupling judging module is used for comparing the waveform form of the actual residual sequence with that of the theoretical residual sequence, calculating the form similarity and judging the abnormal property of the time sequence sign data based on the form similarity; and the dynamic intervention module is used for generating a hierarchical intervention instruction if the abnormal property is judged to be the effective health risk, and executing data filtering operation if the abnormal property is judged to be the environmental noise.
2. The system of claim 1, wherein the physiological baseline reconstruction module is specifically configured to: analyzing the static physiological attribute data, and determining the age parameter and the sex parameter of the target object; Based on the age parameter and the sex parameter, matching a corresponding natural physiological decay curve from the medical knowledge graph; Establishing a physiological steady-state differential equation containing a metabolic rate attenuation variable and a natural physiological drift rate variable based on a natural physiological decay curve and a steady-state rhythm function in a medical knowledge graph; And solving a physiological steady-state differential equation to generate a time sequence curve which does not contain pathological mutation characteristics and taking the time sequence curve as an ideal physiological reference.
3. The system of claim 1, wherein the pathological disturbance simulation module is specifically configured to: Extracting pathological causes and patterns of illegal behaviors from the medical knowledge graph, and converting the pathological causes and patterns of illegal behaviors into computable disturbance factors; Constructing a pathological disturbance function set, wherein the pathological disturbance function set comprises a concentration falling function for simulating drug compliance failure and a peak attenuation coefficient for simulating metabolic function deterioration; And respectively injecting the pathological disturbance functions in the pathological disturbance function set into ideal physiological references to generate a plurality of corresponding hypothetical pathological curves, and taking each hypothetical pathological curve as a pathological simulation state.
4. The system for managing health archives of an information-based planning crowd according to claim 3, wherein the dual-rail differential extraction module is specifically configured to: performing differential operation with an ideal physiological reference on each pathological simulation state to generate a pure pathological deviation waveform; And extracting morphological characteristics of the pure pathological deviation waveform, removing baseline drift influence, and taking the processed waveform data as a theoretical residual sequence, wherein the theoretical residual sequence represents the theoretical deviation morphology caused by specific pathological reasons.
5. The system for managing health records of an information-based planning crowd according to claim 1, wherein the coupling research module is specifically configured to: performing time axis alignment processing on the real residual sequence and the theoretical residual sequence; calculating the waveform optimal matching path of the real residual sequence and the theoretical residual sequence on time sequence by adopting a dynamic time warping algorithm; And calculating Euclidean distance or cosine similarity between the two sequences based on the optimal matching path, and normalizing the calculation result to obtain the morphological similarity.
6. The system of claim 5, wherein the coupled research module is configured to, when determining the abnormal nature of the time series sign data: presetting a judgment threshold set, wherein the judgment threshold set comprises a risk confirmation threshold; comparing the morphological similarity with a risk confirmation threshold; if the morphological similarity is greater than or equal to a risk confirmation threshold, judging that the reality residual sequence has pathological evolution characteristics, determining that the abnormal property is effective health risk, and triggering a grading intervention instruction; if the morphological similarity is smaller than the risk confirmation threshold, judging that the actual residual sequence is sporadic fluctuation, and determining that the abnormal property is environmental noise.
7. The system of claim 6, wherein the dynamic intervention module is specifically configured to: Acquiring a pathological simulation state corresponding to the effective health risk, and identifying a pathological disturbance function type associated with the pathological simulation state; based on the pathological disturbance function type, retrieving a corresponding clinical intervention path from the medical knowledge graph; And determining a risk level according to the deviation amplitude of the time sequence sign data, and generating a corresponding grading intervention instruction by combining a clinical intervention path.
8. The system for managing a health archive of an information-based planning crowd according to claim 1, wherein the archive data aggregation module is specifically configured to: receiving raw sign signals from a plurality of heterogeneous monitoring devices; Filling the missing value and smoothing the abnormal value of the original sign signal to generate standardized time sequence sign data; Periodically decomposing the time sequence sign data, separating a long-term trend component and a seasonal fluctuation component, and taking the normalized time sequence sign data after noise removal as main input data of subsequent processing.
9. The system for managing an information-based planning crowd health profile of claim 1, further comprising an adaptive correction module configured to: If the abnormal property is judged to be environmental noise, marking the corresponding time sequence sign data as a non-pathological sample; and fine-tuning the construction parameters of the ideal physiological reference by using the non-pathological sample so as to update the personalized physiological steady-state model of the target object.

Description

Information planning crowd health record management system Technical Field The invention relates to the technical field of medical data processing and health information management, in particular to an informationized planning crowd health record management system. Background Along with the rapid development of medical informatization technology, the dimension of physiological data covered by the crowd health record shows explosive growth, and the complexity of the multi-source heterogeneous data brings serious challenges to the continuous tracking and accurate evaluation of health status, especially in the aspect of automatic analysis of massive time sequence sign data; at present, the traditional health data management and monitoring method mainly depends on a preset static numerical value threshold to alarm or only files physical examination records at a single moment, however, the prior art lacks of modeling the first principle of the natural physiological decay track of an individual, and is difficult to effectively strip the interference of environmental noise generated by factors such as measurement posture difference, short-term emotion fluctuation and the like; therefore, how to accurately filter noise and identify effective health risks in complex dynamic physiological data is a problem to be solved in the art. Disclosure of Invention In order to solve the technical problems, the invention provides an informationized planning crowd health record management system, and specifically, the technical scheme of the invention comprises the following steps: The archive data aggregation module is used for carrying out multidimensional collection on the historical health record of the target object, acquiring time sequence sign data and static physiological attribute data, and calling a preset medical knowledge graph; The physiological reference reconstruction module is used for constructing a dynamic health track of the target object in an ideal compliance state based on the static physiological attribute data and the medical knowledge graph to serve as an ideal physiological reference; The pathological disturbance simulation module is used for defining a pathological disturbance function based on a medical knowledge graph, superposing the pathological disturbance function to an ideal physiological reference and generating a pathological simulation state containing theoretical pathological features; the double-track differential extraction module is used for calculating the deviation between the time sequence sign data and the ideal physiological reference, obtaining a real residual sequence, calculating the deviation between the pathological simulation state and the ideal physiological reference and obtaining a theoretical residual sequence; The coupling judging module is used for comparing the waveform form of the actual residual sequence with that of the theoretical residual sequence, calculating the form similarity and judging the abnormal property of the time sequence sign data based on the form similarity; and the dynamic intervention module is used for generating a hierarchical intervention instruction if the abnormal property is judged to be the effective health risk, and executing data filtering operation if the abnormal property is judged to be the environmental noise. Preferably, the physiological reference reconstruction module is specifically configured to: analyzing the static physiological attribute data, and determining the age parameter and the sex parameter of the target object; Based on the age parameter and the sex parameter, matching a corresponding natural physiological decay curve from the medical knowledge graph; Establishing a physiological steady-state differential equation containing a metabolic rate attenuation variable and a natural physiological drift rate variable based on a natural physiological decay curve and a steady-state rhythm function in a medical knowledge graph; And solving a physiological steady-state differential equation to generate a time sequence curve which does not contain pathological mutation characteristics and taking the time sequence curve as an ideal physiological reference. Preferably, the pathological disturbance simulation module is specifically configured to: Extracting pathological causes and patterns of illegal behaviors from the medical knowledge graph, and converting the pathological causes and patterns of illegal behaviors into computable disturbance factors; Constructing a pathological disturbance function set, wherein the pathological disturbance function set comprises a concentration falling function for simulating drug compliance failure and a peak attenuation coefficient for simulating metabolic function deterioration; And respectively injecting the pathological disturbance functions in the pathological disturbance function set into ideal physiological references to generate a plurality of corresponding hypothetical pathological curves, an