Analysis of Conjoint Response in Post-stroke Patients Using the Attention-coupled Weighting Method  

作　　者：Jingyao Chen Chen Wang Ningcun Xu Zeng-Guang Hou Liang Peng Pu Zhang 

机构地区：[1]Faculty of Innovation Engineering,CASIA-MUST Joint Laboratory of Intelligence Science and Technology,Macao University of Science and Technology,Macao 999078,China [2]State Key Laboratory of Multimodal Artificial Intelligence Systems,Institute of Automation,Chinese Academy of Sciences,Beijing 100190,China [3]Institute of Analysis and Testing,Beijing Academy of Science and Technology,Beijing 100089,China [4]Department of Rehabilitation Evaluation,China Rehabilitation Research Center,Beijing 100068,China

出　　处：《Machine Intelligence Research》2025年第2期352-367,共16页机器智能研究(英文版)

基　　金：supported in part by the National Key Research and Development Program of China(No.2022YFC3601200);in part by the National Natural Science Foundation of China(Nos.U1913601,U21A20479 and 62203441);in part by the Beijing Sci&Tech Program,China(No.Z211100007921021);in part by the Beijing Natural Science Foundation,China(No.Z170003).

摘　　要：Rehabilitation assessment plays a vital role in the recovery process of post-stroke patients.Currently,many studies have focused on the implementation of automated assessment scales.However,the impact of stroke on the synergistic pattern between different human joints is not well understood.In this study,we developed an attention-coupled weighting based quantitative analysis model to identify the pathological alteration in the conjoint response structure between stroke patients and healthy participants,thus providing more intrinsic suggestions for the post-stroke rehabilitation training.In collaboration with the China Rehabilitation Research Center and the Macao University of Science and Technology,we recruited 15 post-stroke patients and 15 healthy participants to complete several actions selected from the Fugl-Meyer assessment(FMA)scale,which is a commonly used post-stroke rehabilitation assessment scale in clinical practice.Our study proposed an attentional coupling weight(ACW)extraction method.By filtering the weights in the attentional coupling network using a priori matrix,we can analyse the effect of stroke on the alteration of synergistic structures and the coordination patterns of the upper and lower extremities.Moreover,we used a commonly accepted synergy analysis method,the nonnegative matrix decomposition(NMF)method,for comparison and validation.The experimental results demonstrated the effectiveness of our method in quantifying the synergistic structure pattern alterations in post-stroke patients,which provides a new targeted option for the rehabilitation process.

关 键 词：Intelligent assessment conjoint response attentional coupling network spatio-temporal weight extraction. 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TP391[自动化与计算机技术—计算机科学与技术]