CN-122000070-A - Cognitive trend analysis method, system, device, medium and program product

CN122000070ACN 122000070 ACN122000070 ACN 122000070ACN-122000070-A

Abstract

The disclosure provides a cognitive trend analysis method, a system, equipment, a medium and a program product, wherein the cognitive trend analysis method comprises the steps of obtaining historical test data of a test object, obtaining cognitive characteristic vectors of the test object under different time windows based on the historical test data to obtain cognitive development slopes of the test object, obtaining cognitive development slope distribution data of a crowd to which the test object belongs based on group characteristics of the test object, and generating a cognitive trend analysis result of the test object based on the cognitive development slope distribution data and the cognitive development slopes of the test object. By acquiring the cognitive development slope distribution data of the crowd to which the test object belongs, the distribution situation of the cognitive development slope of the test object in the crowd to which the test object belongs can be comprehensively considered when the cognitive trend analysis result is generated, and the result is more reliable.

Inventors

XIA MINGYUE
TANG WEI
YANG CHUAN
LI WENBO

Assignees

华院计算技术(上海)股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260407

Claims (10)

1. A method of cognitive trend analysis comprising: Acquiring historical test data of a test object, wherein the historical test data comprises multi-modal characteristics generated by the test object in different time windows for performing cognitive tests, and the multi-modal characteristics are associated with the cognitive states of the test object; Acquiring cognitive characteristic vectors of the test object under different time windows based on the historical test data so as to acquire a cognitive development slope of the test object; acquiring cognitive development slope distribution data of a crowd to which the test object belongs based on the group characteristics of the test object; and generating a cognitive trend analysis result of the test subject based on the cognitive development slope distribution data and the cognitive development slope of the test subject.
2. The cognitive trend analysis method of claim 1, wherein the step of obtaining the cognitive feature vector of the test subject over different time windows based on the historical test data comprises: acquiring a multi-modal feature mean value and a multi-modal feature standard deviation of a crowd to which the test object belongs based on the group features of the test object; normalizing the multi-modal characteristics generated by the cognitive test of the test object under different time windows based on the multi-modal characteristic mean value and the multi-modal characteristic standard deviation to obtain normalized multi-modal characteristics of the test object; acquiring cognitive feature vectors of the test object under different time windows based on the normalized multi-modal features and the multi-modal fusion model; and/or the number of the groups of groups, The cognitive development slope is obtained based on Theil-Sen method or linear regression method.
3. The cognitive trend analysis method of claim 1, wherein the step of obtaining a cognitive development slope of the test subject comprises: Acquiring a cognitive trend smooth curve of the test object based on the cognitive feature vectors of the test object under different time windows and a lower method; and acquiring the cognitive development slope of the test object based on the cognitive trend smooth curve, theil-Sen method or linear regression method.
4. The cognitive trend analysis method of claim 1, wherein the cognitive feature vector includes cognitive features in different dimensions, the method further comprising: acquiring a plurality of cognitive feature sampling values in at least one dimension based on a Monte Carlo sub-sampling method and the multi-modal feature; Acquiring cognitive feature variances of corresponding dimensions based on the plurality of cognitive feature sampling values; and acquiring confidence intervals of the cognitive features of the corresponding dimension based on the cognitive feature variance.
5. The cognitive trend analysis method of claim 1, wherein the multimodal features include at least one of speech behavior features, visual behavior features, motion trajectories, reaction time, operational behavior features, scale records, event stream features; and/or the number of the groups of groups, The cognitive feature vector comprises cognitive features of at least one dimension of memory, attention, executive function, language ability, visual space ability; and/or the number of the groups of groups, The population characteristics include at least one of age group, educational background, gender, stage of disease progression.
6. The cognitive trend analysis method of any one of claims 1-5, wherein the step of generating a cognitive trend analysis result for the test subject based on the cognitive development slope distribution data and the cognitive development slope of the test subject comprises: Acquiring crowd scores of cognitive trends of the test object in the affiliated crowd based on the cognitive development slope distribution data and the cognitive development slope of the test object; generating a cognitive trend analysis result of the test object based on the crowd score; and/or the number of the groups of groups, The cognitive trend analysis result comprises a crowd score of the cognitive trend of the test object in the crowd to which the test object belongs and the health degree of the cognitive trend of the test object, wherein the health degree of the cognitive trend is related to the crowd score; and/or the number of the groups of groups, The step of generating the cognitive trend analysis result of the test subject based on the cognitive development slope distribution data further comprises the following steps: early warning is carried out based on the cognitive trend analysis result, or early warning is carried out based on the cognitive trend analysis result and the crowd quantile of the cognitive feature vector of the crowd to which the test object belongs in the crowd to which the test object belongs; and/or the number of the groups of groups, The step of generating the cognitive trend analysis result of the test subject based on the cognitive development slope distribution data further comprises the following steps: Generating a cognitive analysis report, wherein the cognitive analysis report comprises at least one of a cognitive development slope of the test object, cognitive development slope distribution data of a crowd to which the test object belongs, a cognitive feature vector of the test object, cognitive feature vector distribution data of the crowd to which the test object belongs, crowd score of a cognitive trend of the test object in the crowd to which the test object belongs, crowd score of the cognitive feature vector of the crowd to which the test object belongs, health degree of the cognitive trend of the test object, rehabilitation advice, multi-modal features, confidence intervals of the cognitive features and improvement rate of the cognitive features.
7. A cognitive trend analysis system, comprising: The data acquisition module is used for acquiring historical test data of a test object, wherein the historical test data comprises multi-modal characteristics generated by the test object in different time windows for performing cognitive tests, and the multi-modal characteristics are associated with the cognitive states of the test object; The cognitive development slope calculation module is used for acquiring cognitive characteristic vectors of the test object under different time windows based on the historical test data so as to acquire the cognitive development slope of the test object; The distribution data acquisition module is used for acquiring cognitive development slope distribution data of the crowd to which the test object belongs based on the group characteristics of the test object; and the result generation module is used for generating a cognitive trend analysis result of the test object based on the cognitive development slope distribution data and the cognitive development slope of the test object.
8. An electronic device comprising a memory, a processor and a computer program stored on the memory for execution on the processor, wherein the processor implements the cognitive trend analysis method of any one of claims 1 to 6 when the computer program is executed.
9. A computer readable storage medium having stored thereon a computer program, characterized in that the computer program, when executed by a processor, implements the cognitive trend analysis method of any one of claims 1 to 6.
10. A computer program product comprising a computer program which, when executed by a processor, implements the cognitive trend analysis method of any one of claims 1-6.

Description

Cognitive trend analysis method, system, device, medium and program product Technical Field The invention relates to the technical field of electric digital data processing, in particular to a cognitive trend analysis method, a system, equipment, a medium and a program product. Background The cognitive evaluation has important disease diagnosis, treatment and rehabilitation significance, and can be used for identifying cognitive dysfunction, guiding clinical diagnosis and treatment, guiding clinical medication and the like. The single cognitive evaluation can only reflect the cognitive ability of the test object under a single time window, has instability, cannot obtain the development trend of the cognitive ability of the test object, and is difficult to propose the diagnosis and treatment suggestion of the next step, so that the analysis of the cognitive trend of the test object is particularly important in the scenes of clinical diagnosis and treatment and the like. In addition, the conventional cognitive trend analysis method only judges the cognitive trend of the test object according to a fixed threshold value when generating an analysis result, and is difficult to comprehensively consider the characteristics of the crowd to which the test object belongs, so that the generated analysis result lacks reliability. Disclosure of Invention The technical problem to be solved by the present disclosure is to overcome the defects of insufficient utilization of objective data and lack of reliability of the generated result in the cognitive trend analysis method in the prior art, and provide a cognitive trend analysis method, a system, a device, a medium and a program product. The technical problems are solved by the following technical scheme: The present disclosure provides a cognitive trend analysis method comprising: Acquiring historical test data of a test object, wherein the historical test data comprises multi-modal characteristics generated by the test object in different time windows for performing cognitive tests, and the multi-modal characteristics are associated with the cognitive states of the test object; Acquiring cognitive characteristic vectors of the test object under different time windows based on the historical test data so as to acquire a cognitive development slope of the test object; acquiring cognitive development slope distribution data of a crowd to which the test object belongs based on the group characteristics of the test object; and generating a cognitive trend analysis result of the test subject based on the cognitive development slope distribution data and the cognitive development slope of the test subject. Optionally, the step of acquiring the cognitive feature vector of the test object under different time windows based on the historical test data includes: acquiring a multi-modal feature mean value and a multi-modal feature standard deviation of a crowd to which the test object belongs based on the group features of the test object; normalizing the multi-modal characteristics generated by the cognitive test of the test object under different time windows based on the multi-modal characteristic mean value and the multi-modal characteristic standard deviation to obtain normalized multi-modal characteristics of the test object; And acquiring cognitive characteristic vectors of the test object under different time windows based on the normalized multi-modal characteristics and the multi-modal fusion model. Optionally, the cognitive development slope is obtained based on Theil-Sen method or linear regression method. Optionally, the step of obtaining the cognitive development slope of the test subject includes: Acquiring a cognitive trend smooth curve of the test object based on the cognitive feature vectors of the test object under different time windows and a lower method; and acquiring the cognitive development slope of the test object based on the cognitive trend smooth curve, theil-Sen method or linear regression method. Optionally, the cognitive feature vector includes cognitive features in different dimensions, and the method further includes: acquiring a plurality of cognitive feature sampling values in at least one dimension based on a Monte Carlo sub-sampling method and the multi-modal feature; Acquiring cognitive feature variances of corresponding dimensions based on the plurality of cognitive feature sampling values; and acquiring confidence intervals of the cognitive features of the corresponding dimension based on the cognitive feature variance. Optionally, the multi-modal features include at least one of a voice behavior feature, a visual behavior feature, a motion trail, a reaction time, an operational behavior feature, a scale record, an event stream feature; Optionally, the cognitive feature vector includes cognitive features of at least one dimension of memory, attention, executive function, linguistic ability, visual space ability; Optionally, the population cha