首页|期刊导航|康复学报|基于人工智能的强化言语-语言治疗对脑卒中后失语症患者语言功能的影响

基于人工智能的强化言语-语言治疗对脑卒中后失语症患者语言功能的影响OA

Effects of Artificial Intelligence-Based Intensive Speech-Language Therapy on Language Function in Post-Stroke Aphasic Patients

中文摘要英文摘要

目的 探讨人工智能(AI)的强化言语-语言治疗(AISLT)对脑卒中后失语症患者语言功能的影响.方法 选取2024年1-12月在复旦大学附属华东医院康复医学科住院的脑卒中后失语症患者36例,采用随机数字表法分为对照组、常规强化言语-语言治疗(TISLT)组和AISLT组3组,每组12例.TISLT组脱落1例,AISLT组脱落1例,最终完成试验者34例,其中对照组12例,TISLT组11例,AISLT组11例.对照组接受常规言语语言康复治疗及健康教育,30 min/次;TISLT组接受言语治疗师介导的TISLT,120 min/次;AIS-LT组接受基于AISLT,120 min/次;3组均1次/d,5 d/周,持续治疗2周.分别于治疗前、治疗2周后和治疗后1个月采用西方失语症成套测验(WAB)对失语症的语言功能(听、说、读、写)进行全面评估并进行分类诊断;采用波士顿命名测验(BNT)对患者命名能力进行评估;采用日常生活交流能力评定(CADL)对患者交流能力进行评估;采用脑卒中失语症生活质量量表-39条目通用版(SAQOL-39 g)对患者生活质量进行评估.结果 与治疗前比较,对照组命名评分升高(P<0.05),TISLT组治疗后WAB-AQ、自发语言、听理解、复述和命名评分均升高(P<0.05),TISLT组治疗后1个月和AISLT组治疗后、治疗后1个月WAB-AQ、听理解、复述和命名评分均升高(P<0.05);与对照组比较,TISLT组和AISLT组治疗后WAB-AQ、听理解、复述和命名评分均升高(P<0.05),TISLT组治疗后1个月WAB-AQ、听理解、复述评分均升高(P<0.05),AISLT组治疗后1个月WAB-AQ、听理解和命名评分均升高(P<0.05);与TISLT组比较,AISLT组在治疗后WAB-AQ和复述评分均升高(P<0.05).与治疗前比较,3组治疗后BNT评分均升高(P<0.05),TISLT组和AISLT组治疗1个月后BNT评分均升高(P<0.05).与对照组比较,AISLT组治疗后和治疗后1个月BNT评分均升高(P<0.05).与治疗前比较,AISLT组治疗后、治疗后1个月的CADL评分均升高(P<0.05);与对照组比较,AISLT组治疗后、治疗后1个月CADL评分均升高(P<0.05);与TISLT组比较,AISLT组治疗后、治疗后1个月CADL评分均升高(P<0.05).与治疗前比较,TISLT组和AISLT组治疗后、治疗后1个月SAQOL-39g评分均升高(P<0.05);与对照组比较,TISLT组治疗后和AISLT组治疗后、治疗后1个月评分均升高(P<0.05).结论 TISLT和AISLT均能有效改善脑卒中后亚急性期失语症患者语言功能,但AISLT疗效更显著.

Objective This study aimed to investigate the effects of artificial intelligence-based intensive speech-language therapy(AISLT)on language function among patients with post-stroke aphasia.Methods Thirty-six patients with post-stroke apha-sia who were hospitalized in the Department of Rehabilitation Medicine,Huadong Hospital Affiliated to Fudan University from Jan-uary to December 2024 were enrolled.The participants were randomly allocated into a control group,a therapist-led traditional inten-sive speech-language therapy(TISLT)group and an AISLT group via a random number table,with 12 patients in each group.One patient in the TISLT group and one in the AISLT group withdrew from the study.Finally,34 patients completed the trial,including 12 in the control group,11 in the TISLT group and 11 in the AISLT group.The control group received conventional speech-language rehabilitation combined with health education for 30 min per session.The TISLT group received therapist-led intensive speech-language therapy for 120 min per session,while the AISLT group received artificial intelligence-based intensive speech-lan-guage therapy for 120 min per session.All interventions were administered once daily,5 days per week,for 2 consecutive weeks.As-sessments were performed before treatment,immediately after treatment and one month after treatment.Language functions(listen-ing,speaking,reading and writing)were comprehensively evaluated and aphasia was classified using the Western Aphasia Battery(WAB).Naming ability was assessed via the Boston naming test(BNT),functional communication was evaluated using the commu-nicative activities of daily living(CADL),and quality of life was assessed using the Stroke and Aphasia Quality of Life Scale-39 generic version(SAQOL-39g).Results Compared with baseline,the naming score significantly increased in the control group(P<0.05).In the TISLT group,the WAB Aphasia Quotient(WAB-AQ),spontaneous speech-language,auditory comprehension,repeti-tion and naming scores significantly increased immediately after treatment(P<0.05),while the WAB-AQ,auditory comprehension,repetition and naming scores remained significantly improved at one month after treatment(P<0.05).In the AISLT group,the WAB-AQ,auditory comprehension,repetition and naming scores significantly increased both immediately after treatment and at one month after treatment(P<0.05).Compared with the control group,both the TISLT and AISLT groups showed significantly higher WAB-AQ,auditory comprehension,repetition and naming scores immediately after treatment(P<0.05).At one month after treat-ment,the TISLT group demonstrated significantly higher WAB-AQ,auditory comprehension and repetition scores(P<0.05),where-as the AISLT group exhibited significantly higher WAB-AQ,auditory comprehension and naming scores(P<0.05).Compared with the TISLT group,the AISLT group achieved significantly higher WAB-AQ and repetition scores immediately after treatment(P<0.05).Compared with baseline,BNT scores significantly increased after treatment in all three groups(P<0.05),and remained signifi-cantly improved at one month after treatment in both the TISLT and AISLT groups(P<0.05).Compared with the control group,the AISLT group showed significantly higher BNT scores both immediately after treatment and at one month after treatment(P<0.05).CADL scores significantly increased immediately after treatment and at one month after treatment in the AISLT group(P<0.05),and were significantly higher than those in both the control group and the TISLT group(P<0.05).SAQOL-39g scores significantly increased immediately after treatment and at one month after treatment in both the TISLT and AISLT groups(P<0.05).Compared with the control group,the TISLT group showed significantly higher SAQOL-39g score immediately after treatment,whereas the AISLT group demonstrated significantly higher scores both immediately after treatment and at one month after treatment(P<0.05).Conclusion Both therapist-led intensive speech-language therapy and artificial intelligence-based intensive speech-language thera-py effectively improve language function in patients with subacute post-stroke aphasia.However,AISLT producing more pronounced therapeutic effects.

陈秀恩;江紫辉;郑洁皎;郑拥军;宁志超;石文慧;段林茹;朱婷;张云云

上海中医药大学附属岳阳中西医结合医院,上海 200437||复旦大学附属华东医院,上海 200040||上海体育大学运动健康学院,上海 200438上海体育大学运动健康学院,上海 200438||上海市浦东新区周浦医院,上海 201318复旦大学附属华东医院,上海 200040复旦大学附属华东医院,上海 200040复旦大学附属华东医院,上海 200040||上海体育大学运动健康学院,上海 200438复旦大学附属华东医院,上海 200040||上海体育大学运动健康学院,上海 200438复旦大学附属华东医院,上海 200040复旦大学附属华东医院,上海 200040上海中医药大学附属岳阳中西医结合医院,上海 200437

脑卒中后失语症语言功能基于人工智能的强化言语-语言治疗常规强化言语-语言治疗人工智能

post-stroke aphasialanguage functionartificial intelligence-based intensive speech-language therapytraditional intensive speech-language therapyartificial intelligence

《康复学报》 2026 (7)

500-507,516,9

国家重点研发计划项目(2024YFF0507501)上海市卫健委面上项目(202240256)转化医学研究基金项目(2024YJJB13)复旦大学附属华东医院横向课题(H1573)

10.3724/SP.J.1329.2026.07005

评论