特发性全面性癫痫患者大脑结构异常改变的MRI研究  

MRI study of abnormal changes of brain structure in patients with idiopathic generalized epilepsy

作　　者：夏新建 李宁 陈金鸽 于乔文 侯中煜[1] Xia Xinjian;Li Ning;Chen Jinge;Yu Qiaowen;Hou Zhongyu(Department of Medical Imaging,Shandong Provincial Hospital Affiliated to Shandong First Medical University,Shandong Provincial Hospital Affiliated to Shandong University,Jinan 250021,China;Department of Medical Imaging,Shouguang People's Hospital,Shouguang 262700,China)

机构地区：[1]山东第一医科大学附属省立医院/山东省立医院/山东大学附属省立医院医学影像科,济南250021 [2]山东省寿光市人民医院影像中心,寿光262700

出　　处：《中华解剖与临床杂志》2023年第3期141-146,共6页Chinese Journal of Anatomy and Clinics

基　　金：国家自然科学基金(81001223);山东省重点科研项目(2017GSF218077);山东省医药卫生科技发展计划(2106WS0435);山东省自然科学基金(ZR2021MH063);山东第一医科大学学术提升计划(2019QL023);潍坊市卫健委科研项目计划(WFWSJK-2021-038)。

摘　　要：目的基于深度学习脑龄预测方法和FreeSurfer图像分析软件, 探讨特发性全面性癫痫(IGE)患者大脑形态结构的异常改变。方法回顾性病例对照研究。纳入2020年1月—2021年12月山东省立医院及寿光市人民医院神经内科确诊的144例IGE患者。其中男84例、女60例, 年龄10~70岁。基于公共脑成像数据库(人类连接组项目、中国人类连接组项目)建立正常人脑T1加权像(T1WI)MRI训练深度学习脑龄预测模型, 并输入脑皮质及脑白质脑图。将IGE患者T1WI MRI输入脑龄预测模型, 获取预测脑龄。将27例未经治疗的早期IGE患者设为观察组, 男18例、女9例, 年龄(16.1±3.2)岁;招募与观察组年龄、性别匹配的29名健康志愿者作为对照组, 男19例、女10例, 年龄(16.4±4.5)岁, 均行MR检查。观察项目:(1)比较IGE患者不同年龄段(10~20岁、>20~30岁、>30~40岁、>40~50岁、>50~60岁、>60~70岁)大脑的预测年龄与实际年龄的差异。(2)利用FreeSurfer软件, 分析和比较观察组与对照组大脑皮质体积、厚度和表面积, 以及局部形态的差异。结果 (1)IGE患者的不同年龄段中, 除>30~40岁外(t=-1.58, P=0.138), 脑龄预测值均大于实际年龄, 差异均有统计学意义(P值均<0.05)。(2)观察组与对照组的年龄、性别、教育程度等基线资料差异均无统计学意义(P值均>0.05)。与对照组比较, 观察组大脑左、右半球皮质厚度略变薄, 差异有统计学意义(t=3.91、3.40, P值均<0.01);2组大脑半球皮质体积、表面积无明显变化, 差异均无统计学意义(P值均>0.05)。与对照组相比, 观察组脑区中右侧眶额皮质体积增大, 无脑区皮质体积减小;右侧中央后回、左侧中央沟附近中央后回、右侧舌回后部分、左侧舌回及楔叶的皮质厚度均增加, 右侧中央前回、双侧额上回(背侧)、右侧内嗅皮质的皮质厚度均减小;右侧海马旁回的皮质表面积增大, 无脑区的皮质表面积减小。观察Objective Abnormal changes in brain morphology in patients with idiopathic generalized epilepsy(IGE)were discussed using deep learning brain age prediction and FreeSurfer image analysis software.Methods This retrospective case-control study recruited 144 patients with IGE diagnosed in the Department of Neurology of Shandong Provincial Hospital and Shouguang People's Hospital from January 2020 to December 2021.The participants consisted of 84 males and 60 females and aged 10-70 years.Based on the public brain imaging database(human connectivity group project,Chinese human connectivity group project),the T1-weighted(T1WI)magnetic resonance imaging(MRI)of the normal human brain was trained to deeply learn the brain age prediction model,and the brain cortex and white matter maps were inputted.The T1WI MRI of patients with IGE was incorporated into the brain age prediction model to obtain the predicted age of the brain.Twenty-seven untreated patients with early IGE were divided into the observation group and comprised 18 males and 9 females aged(16.1±3.2)years.Twenty-nine healthy volunteers matched with the age and sex of the observation group were recruited as the control group.These volunteers consisted of 19 males and 10 females aged(16.4±4.5)years who underwent MR examination.(1)The predicted age and the actual age of patients with IGE were compared among different age groups(10-20 years old,>20-30 years old,>30-40 years old,>40-50 years old,>50-60 years old,>60-70 years old).(2)Freesurfer software was used to analyze and compare the volume,thickness,surface area,and local morphology of the cerebral cortex between the observation and control groups.Results(1)In different age groups of IGE patients,except for>30-40 years old(t=-1.58,P=0.138),the predicted value of brain age was greater than the actual age,and the difference was statistically significant(all P values<0.05).(2)No significant difference in age,sex,education,and other baseline data was found between the observation and control groups(all P values>0.0

关 键 词：癫痫 特发性全面性癫痫 核磁共振成像 深度学习 脑皮质 核团 

分 类 号：R742.1[医药卫生—神经病学与精神病学] R445.2[医药卫生—临床医学]