CN-122000043-A - Dementia cognition assessment method, device, equipment and storage medium

CN122000043ACN 122000043 ACN122000043 ACN 122000043ACN-122000043-A

Abstract

The application provides a dementia cognition evaluation method, device, equipment and storage medium, wherein the method comprises the steps of constructing original interaction data of a patient cognition execution whole process, determining attribution relation of the original interaction data in different cognition domains based on preset cognition domain division rules, constructing individual cognition baselines based on historical interaction data through sliding time windows, calculating offset vectors of different domain feature subsets relative to corresponding individual cognition baselines, acquiring function damage quantification results of different cognition domains according to the offset vectors, calculating damage dispersion among different cognition domains through variance, analyzing correlation among different cognition domains, acquiring cognition function distribution states, determining a function imbalance dominant domain and a relative cooperative domain, generating a compensation care strategy corresponding to the imbalance states based on the function imbalance dominant domain and the relative cooperative domain.

Inventors

ZHANG JIE
WANG XU
DUAN JIANPING
MA SEN
GUO YANG
WANG XIAOCHUN
GAO BEI
LIU XUERU

Assignees

河南中医药大学第一附属医院

Dates

Publication Date: 20260508
Application Date: 20251224

Claims (10)

1. A cognitive assessment method for dementia, comprising: Based on a preset cognition assessment task, acquiring operation behaviors, input contents and response time of a patient in the task execution process in real time through an interaction terminal, marking the operation behaviors and the input contents by combining a time stamp, and constructing original interaction data of the whole patient cognition execution process; based on a preset cognitive domain division rule, carrying out feature analysis on operation behaviors, input contents and response time, and determining attribution relations of the original interaction data in different cognitive domains by combining task types, interaction phases and time relations corresponding to the original interaction data to generate corresponding domain feature subsets; based on historical interaction data, constructing individual cognitive baselines through a sliding time window, calculating offset vectors of different domain feature subsets relative to corresponding individual cognitive baselines, and acquiring functional damage quantification results of different cognitive domains according to the offset vectors; based on the function damage quantification result of each cognitive domain, calculating damage dispersion among different cognitive domains through variance, analyzing correlation among different cognitive domains, obtaining cognitive function distribution states, and determining a function imbalance dominant domain and a relative cooperative domain; Based on the functional imbalance dominant domain and the relative synergy domain, a compensatory care strategy corresponding to the imbalance state is generated.
2. The dementia cognitive assessment method as claimed in claim 1, wherein the acquisition is based on a preset cognitive assessment task, comprising: Constructing an initial state parameter vector based on the state information of the patient; configuring task parameters based on a preset cognition evaluation task parameter space; Determining a task parameter combination matched with the state of the patient according to the mapping relation between the initial state parameter vector and the task parameter space; Generating an instance set adapted to the patient cognitive assessment task based on the determined task parameter combination; In the executing process of the cognitive evaluation task instance, the interactive data of the patient is collected, and the task parameters of the cognitive evaluation task instance are dynamically adjusted according to the interactive data, so that the self-adaptive evaluation of the cognitive state of the patient is realized.
3. The dementia cognitive assessment method of claim 1, wherein determining the home relation of the raw interaction data in different cognitive domains comprises: based on the input content and response time, the language expression capability of the patient is estimated through semantic density extraction and voice fluency analysis, and language domain attribution information of the patient is obtained; Analyzing the execution control and planning capacity through the operation behavior sequence and the logic switching response time, and acquiring the execution domain attribution information of the patient; Comprehensively analyzing by combining recall accuracy of input content and information checking behaviors in the task execution process, and quantitatively evaluating instantaneous memory of a patient to acquire memory domain attribution information of the patient; And analyzing the visual space construction capacity according to the touch track and the coordinate precision of the operation behavior, and acquiring the time-space perception domain attribution information of the patient.
4. The dementia cognitive assessment method of claim 1, wherein constructing an individual cognitive baseline through a sliding time window based on historical interaction data of the patient comprises: the method comprises the steps of obtaining historical interaction data of a patient, carrying out time sequence segmentation on the historical interaction data, constructing a plurality of continuous evaluation time windows through a sliding time window algorithm, identifying and reflecting optimal cognitive function characteristics of the patient in each time window, and carrying out weighted fusion on the optimal cognitive function characteristics in different time windows to form a dynamic individual cognitive baseline updated along with time.
5. The dementia cognitive assessment method of claim 1, wherein the offset vectors of the different subsets of domain features relative to the corresponding individual cognitive baselines are calculated as Wherein The offset vector is represented as such, As the characteristic value of the current cognitive domain, Cognitive baselines for individuals corresponding to the cognitive domains.
6. The dementia cognitive assessment method of claim 1, wherein calculating the impairment dispersion between different cognitive domains by variance based on the functional impairment quantification results for each cognitive domain, comprises: in the same evaluation period, respectively obtaining the function damage quantification results of different cognitive domains; Based on a plurality of continuous evaluation periods, acquiring a function damage quantification sequence corresponding to each cognitive domain; according to the adjustment weight coefficients of different cognitive domains of different patients, calculating the variance among all components in the functional impairment quantification sequence, and taking the variance value as the impairment dispersion among different cognitive domains.
7. The dementia cognitive assessment method as claimed in claim 6, wherein before analyzing the correlation between different cognitive domains, comprising: judging the damage dispersion, and judging that the function damage degree of each cognitive domain is relatively consistent and the cognitive structure is in an equilibrium state when the damage dispersion is smaller than a preset threshold; And when the damage dispersion is greater than or equal to a preset threshold, judging that the structural imbalance among the cognitive domains exists.
8. A dementia cognitive assessment device for implementing the method of claims 1-7, comprising: The original interaction data construction module is used for acquiring the operation behaviors, the input contents and the response time of the patient in the task execution process in real time through the interaction terminal based on a preset cognition assessment task, marking the operation behaviors and the input contents by combining the time stamp, and constructing the original interaction data of the whole patient cognition execution process; The cognitive domain attribution module is used for carrying out characteristic analysis on the operation behaviors, the input contents and the response time based on preset cognitive domain division rules, and determining attribution relations of the original interaction data in different cognitive domains by combining task types, interaction phases and time relations corresponding to the original interaction data to generate corresponding domain feature subsets; The function damage result quantification module is used for constructing an individual cognitive baseline through a sliding time window based on historical interaction data of a patient, calculating offset vectors of different domain feature subsets relative to corresponding individual cognitive baselines, and acquiring function damage quantification results of different cognitive domains according to the offset vectors; The imbalance relation construction module is used for calculating the damage dispersion degree among different cognitive domains through variance based on the function damage quantification result of each cognitive domain, analyzing the correlation among different cognitive domains, acquiring the cognitive function distribution state of a patient, and determining a function imbalance dominant domain and a relative cooperative domain; And the compensation care strategy generation module is used for generating a compensation care strategy corresponding to the unbalanced state based on the functional unbalance main domain and the relative cooperative domain.
9. An electronic device, comprising: One or more processors; a memory for storing one or more programs; The one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 1-7.
10. A computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the steps of the method of any of claims 1-7.

Description

Dementia cognition assessment method, device, equipment and storage medium Technical Field The application relates to the technical field of dementia cognitive assessment, in particular to a dementia cognitive assessment method, device, equipment and storage medium. Background Dementia is a group of disorders characterized mainly by a decline in acquired, sustained cognitive function, often manifested by a decline in memory, attention deficit, impaired executive function, speech understanding, and expression. Dementia cognitive impairment severely affects the patient's daily life and social abilities. Clinically, alzheimer's disease, vascular dementia and the like are the most common dementia types, and the core features are all the systematic decline of multidimensional cognitive ability. Based on dementia cognition problems, the method mainly focuses on cognition evaluation, risk screening, intervention assistance and the like at present, common evaluation modes comprise a scale evaluation method, such as a simple mental state examination scale, a Montreal cognition evaluation scale and the like, the memory, the attention, the language and the execution capacity of a patient are scored through standardized questions, and a plurality of cognitive states are comprehensively judged by combining imaging examination, biochemical index detection or neuropsychological test results. Although these techniques have certain applications in practice, most scale evaluations rely on manual administration and subjective judgment, and the results are easily affected by experience of an evaluator, coordination degree of a patient and emotional state, and have poor stability and continuity, and it is difficult to characterize dynamic characteristics of cognitive function changing with time based on evaluation results at single or discrete time points. The partial informatization means only realize data acquisition or display, cannot perform deep correlation analysis on the multisource cognitive data, has limited technical effects, and is difficult to provide reliable support for fine management and individual intervention. Therefore, how to overcome the above-mentioned technical problems and drawbacks becomes a major problem to be solved. Disclosure of Invention In order to overcome the problems in the prior art, the application provides a method, a device, equipment and a storage medium, which adopt the following technical scheme: In a first aspect, the present application provides a dementia cognitive assessment method comprising: Based on a preset cognition assessment task, acquiring operation behaviors, input contents and response time of a patient in the task execution process in real time through an interaction terminal, marking the operation behaviors and the input contents by combining a time stamp, and constructing original interaction data of the whole patient cognition execution process; based on a preset cognitive domain division rule, carrying out feature analysis on operation behaviors, input contents and response time, and determining attribution relations of the original interaction data in different cognitive domains by combining task types, interaction phases and time relations corresponding to the original interaction data to generate corresponding domain feature subsets; based on historical interaction data of a patient, constructing an individual cognitive baseline through a sliding time window, calculating offset vectors of different domain feature subsets relative to corresponding individual cognitive baselines, and acquiring functional impairment quantification results of different cognitive domains according to the offset vectors; Based on the function damage quantification result of each cognitive domain, calculating damage dispersion among different cognitive domains through variance, analyzing correlation among different cognitive domains, acquiring cognitive function distribution state of a patient, and determining a function imbalance dominant domain and a relative cooperative domain; Based on the functional imbalance dominant domain and the relative synergy domain, a compensatory care strategy corresponding to the imbalance state is generated. Further, constructing the original interaction data in the cognitive executing process, including: The method comprises the steps of constructing an initial state parameter vector based on state information of a patient, configuring task parameters based on a preset cognition evaluation task parameter space, determining task parameter combinations matched with the state of the patient according to the mapping relation between the initial state parameter vector and the task parameter space, generating an instance set suitable for the cognition evaluation task of the patient based on the determined task parameter combinations, collecting interaction data of the patient in the execution process of the cognition evaluation task instance, and dynamically adj