多模态磁共振成像联合estroke人工智能评估在晚时间窗急性脑梗死动脉取栓术的临床研究  

A Clinical Study of Multimodal MRI Combined with Estroke Artificial Inte ligence to Evaluate Arterial Thrombectlomy in Late Time Window AIS

作　　者：李绍发[1] 黄晓[2] 李登星 黄志志[1] 廖宝 杨再职 黄宇基 巫颖[3] LI Shao-fa;HUANG Xiao;LI Deng-xing;HUANG Zhi-zhi;LIAO Bao;YANG Zai-zhi;HUANG Yu-ji;WU Ying(Department of Neurology,Baise People's Hospital,Baise 533000,Guangxi Zhuang Autonomous Region,China;Department of Radiology,Baise People's Hospital,Baise 533000,Guangxi Zhuang Autonomous Region,China;Department of Cardiovascular Medicine,Baise People's Hospital,Baise 533000,Guangxi Zhuang Autonomous Region,China)

机构地区：[1]百色市人民医院神经内科 [2]百色市人民医院放射科 [3]百色市人民医院心血管内科,广西百色533000

出　　处：《中国CT和MRI杂志》2024年第7期14-17,共4页Chinese Journal of CT and MRI

基　　金：广西壮族自治区卫生健康委员会科研课(Z20201451)。

摘　　要：目的分析多模态磁共振成像(MRI)联合estroke人工智能评估在晚时间窗急性脑梗死(AIS)动脉取栓术中的临床价值。方法选取2021年7月至2023年6月本院收治的104例晚时间窗AIS患者,所有患者经多模态MRI联合estroke人工智能评估后均可进行动脉取栓,根据eStroke人工智评估结果,将患者按照梗死体积分为低梗死组(梗死体积<21ml)、中等梗死组(21mL<梗死体积<31mL)及较大梗死组(31mL<梗死体积<51mL)。另同期选取于本院进行急诊取栓的时间窗内AIS患者30例,纳入对照组。术后采用改良脑梗死溶栓(mTICI)分级评估血管再通情况。统计患者取栓次数、下床活动时间及住院时间,统计患者住院期间并发症发生率。出院后所有患者均随访3个月,统计患者90 d死亡情况及脑卒中再发情况。采用改良Rankin量表(mRS)及美国国立卫生院神经功能缺损评分(NIHSS)评估患者神经功能预后。结果与低梗死组比,较大梗死组血管再通率降低(P<0.05),与较大梗死组比较,对照组血管再通率升高(P<0.05);与低梗死组比较,对照组血取栓次数增加(P<0.05);与低梗死组比较,对照组并发症发生率升高(P<0.05);与术前比较,4组患者术后3个月mRS评分、NIHSS评分均降低(P<0.05);与低梗死组比较,中等梗死组及较大梗死组mRS评分、NIHSS评分升高(P<0.05),与中等梗死组比较,较大梗死组mRS评分、NIHSS评分升高(P<0.05),与较大梗死组比较,对照组mRS评分、NIHSS评分降低(P<0.05),低梗死组与对照组比较,mRS评分、NIHSS评分差异无统计学意义(P>0.05)。结论多模态MRI联合estroke人工智能评估晚时间窗AIS动脉取栓术中具有良好的血管再通率,有利于减少取栓次级术后并发症的发生,对于神经功能改善具有积极作用。Objective To analyze the clinical study of multimodal magnetic resonance imaging(MRI)combined with estroke artificial intelligence to evaluate arterial thrombectomy in late time window acute cerebral infarction(AIS).Methods 104 patients with late time window AIS admitted to our hospital from July 2021 to June 2023 were selected.Arterial thrombectomy was performed in all patients after multimodal MRI combined with estroke artificial intelligence assessment.According to the results of eStroke artificial intelligence evaluation,patients were divided into low infarct group(infarct volume<21mL),medium infarct group(21mL<31mL)and large infarct group(31mL<51mL)according to the infarct volume.In the same period,30 patients with AIS were selected in the time window of emergency thrombectomy in our hospital and included in the control group.Modified cerebral infarction thrombolysis(mTICI)grade was used to evaluate vascular recanalization.The number of thrombectomy,the time of getting out of bed and the time of hospitalization were counted,and the incidence of complications during hospitalization were counted.All patients were followed up for 3 months after discharge,and the death and recurrence of stroke were counted at 90 days.The modified Rankin Scale(mRS)and the National Institutes of Health Neurological Impairment Score(NIHSS)were used to evaluate the neurological prognosis of the patients.Results Compared with the low-infarct group,the vascular revascularization rate in the large-infarct group was lower(P<0.05),while the vascular revascularization rate in the control group was higher(P<0.05).Compared with the low infarction group,the number of thrombectomy increased in the control group(P<0.05).Compared with low infarction group,the complication rate of control group was higher(P<0.05).Compared with preoperative results,mRS Scores and NIHSS scores in 4 groups were decreased 3 months after operation(P<0.05)..Compared with low infarction group,mRS Score and NIHSS score were increased in medium infarction group and large

关 键 词：急性脑梗死 多模态磁共振成像 estroke人工智能 晚时间窗 动脉取栓 

分 类 号：R743.3[医药卫生—神经病学与精神病学] R445.2[医药卫生—临床医学]