首页|期刊导航|机器人外科学杂志(中英文)|机器人辅助训练对脑卒中患者康复效果的影响因素分析

机器人辅助训练对脑卒中患者康复效果的影响因素分析OA

Analysis of influencing factors on the rehabilitation effect of robot-assisted training in stroke patients

中文摘要英文摘要

目的:探讨机器人辅助训练对脑卒中患者康复效果的影响因素.方法:前瞻性选取 2023 年 3 月—2025 年 3 月于北京大学第三医院秦皇岛医院接受机器人辅助康复训练的 103 例脑卒中患者,统计并比较不同特征脑卒中患者的机器人辅助康复训练效果,采用多元线性回归识别患者康复效果的独立影响因素及其量化关系,构建决策树模型以辨识关键预测变量,绘制受试者工作特征(ROC)曲线并计算曲线下面积(AUC),以检验决策树模型的预测价值.结果:训练 3 个月时,脑卒中患者平均Fugl-Meyer运动功能评定量表(FMA)评分高于训练前,且年龄<50 岁、病程<3 个月、抑郁自评量表(SDS)评分<53 分、蒙特利尔认知评估量表(MoCA)评分≥26 分、改良Barthel 指数(mBI)>40 的脑卒中患者进行机器人辅助康复训练的效果较好(P<0.05);经多元线性回归分析显示,年龄、病程、SDS评分、MoCA评分、mBI评分是脑卒中患者机器人辅助康复训练效果的影响因素(P<0.05);由决策树模型可得,病程、SDS评分、MoCA评分、mBI评分是脑卒中患者机器人辅助康复训练效果的独立影响因素,其中病程的影响最为显著,其AUC值为 0.811(95%CI:0.727~0.895),敏感度、特异度、约登指数分别为0.833、0.745、0.578.结论:机器人辅助康复训练能显著改善脑卒中患者的运动功能,其效果受年龄、病程、抑郁、认知及日常生活活动能力五大因素共同影响,其中病程影响最为显著,临床可依据患者是否伴有上述风险因素,进行针对性干预,从而提高机器人辅助康复训练效果.

Objective:To investigate the influencing factors on the rehabilitation effect of robot-assisted training in stroke patients.Methods:A prospective study was conducted to select 103 stroke patients who received robot-assisted rehabilitation training at Qinhuangdao Hospital of Peking University Third Hospital from March 2023 to March 2025.The effects of robot-assisted rehabilitation training on stroke patients with different characteristics were collected and compared.Multiple linear regressions was used to identify independent factors affecting the rehabilitation and quantify their relationships with rehabilitation outcomes.A decision tree model was constructed to identify key predictive variables,and receiver operating characteristic(ROC)curves were plotted with area under the curve(AUC)calculated to evaluate the predictive value of the model.Results:At 3 months of training,the mean Fugl-Meyer assessment(FMA)score of stroke patients significantly increased compared to that before training.Patients aged<50 years,with disease duration<3 months,self-rating depression scale(SDS)score<53,Montreal cognitive assessment(MoCA)score≥26,and modified Barthel index(mBI)>40 showed significantly better rehabilitation outcomes(P<0.05).Multiple linear regression indicated that age,disease duration,SDS score,MoCA score,and mBI score were significant influencing factors(P<0.05).The decision tree model identified disease duration,SDS,MoCA,and mBI scores as independent predictors of the effect of robot-assisted rehabilitation training in stroke patients,with disease duration being the most significant.The AUC for disease duration in predicting rehabilitation outcome was 0.811(95%CI:0.727-0.895),with sensitivity,specificity,and Youden index of 0.833,0.745,and 0.578,respectively.Conclusion:Robot-assisted rehabilitation training can significantly improve the motor function of stroke patients.Its effectiveness is jointly influenced by age,disease duration,depression,cognition,and activities of daily living.Among them,the disease duration has the most significant impact.Clinically,targeted interventions based on these risk factors can enhance the efficacy of robot-assisted rehabilitation.

安月;姜美玉;陈静霞

北京大学第三医院秦皇岛医院康复医学科 河北 秦皇岛 066000北京大学第三医院秦皇岛医院康复医学科 河北 秦皇岛 066000北京大学第三医院秦皇岛医院康复医学科 河北 秦皇岛 066000

医药卫生

脑卒中机器人辅助训练康复影响因素

StrokeRobot-assisted TrainingRehabilitationInfluencing Factors

《机器人外科学杂志(中英文)》 2026 (3)

517-523,7

河北省卫生健康委医学科学研究计划(20250263) Hebei Provincial Health Commission Medical Research Program(20250263)

10.12180/j.issn.2096-7721.2026.03.025

评论