基于深度神经网络的颞叶癫痫^(18)F-FDG PET术后复发预测研究  

Deep neural networks analysis of ^(18)F-FDG PET imaging in postoperative patients with temporal lobe epilepsy

作　　者：吴环华 陈少波 尚靖杰[1] 周海玲 吴彪 弓健[1] 凌雪英[1] 郭强 徐浩[1] Wu Huanhua;Chen Shaobo;Shang Jingjie;Zhou Hailing;Wu Biao;Gong Jian;Ling Xueying;Guo Qiang;Xu Hao(Department of Nuclear Medicine,the First Affiliated Hospital of Jinan University,Guangzhou 510630,China;Central Laboratory,the Affiliated Shunde Hospital of Jinan University,Foshan 528305,China;College of Information Science and Technology,Jinan University,Guangzhou 510632,China;Radiology Department,Zhanjiang Central People′s Hospital,Zhanjiang 524045,China;Epilepsy Center,Guangdong 999 Brain Hospital,Guangzhou 510510,China)

机构地区：[1]暨南大学附属第一医院核医学科,广州510630 [2]暨南大学附属顺德医院中心实验室,佛山528305 [3]暨南大学信息科学技术学院,广州510632 [4]湛江中心人民医院放射科,湛江524045 [5]广东三九脑科医院癫痫中心,广州510510

出　　处：《中华核医学与分子影像杂志》2024年第4期220-224,共5页Chinese Journal of Nuclear Medicine and Molecular Imaging

基　　金：国家自然科学基金(82371998);广州市科技计划项目(2023A03J1035);广州市科技计划项目-市校联合资助项目(SL2022A03J01222)。

摘　　要：目的基于深度残差神经网络(ResNet)分析术前^(18)F-FDG PET影像及患者临床特征,预测难治性颞叶癫痫(TLE)患者术后复发状况。方法回顾性分析2014年1月至2020年6月期间暨南大学附属第一医院诊治的220例难治性TLE患者[男132例、女88例,年龄23.0(20.0,30.2)岁]的术前^(18)F-FDG PET影像及临床资料。采用ResNet对预处理好的PET图像及临床特征进行高通量特征提取,并进行区分TLE患者的术后复发预测任务。评估模型的预测性能,并将其ROC曲线分析所得AUC与经典的生存分析Cox比例风险模型的AUC进行比较(Delong检验)。结果基于PET影像联合临床特征,ResNet预测难治性TLE患者术后12、24、36个月复发的AUC分别为0.895±0.073、0.861±0.058和0.754±0.111,Cox比例风险回归模型相应AUC依次为0.717±0.093、0.697±0.081和0.645±0.087(z值:-3.00、-2.98、-1.09,P值:0.011、0.018、0.310),其中ResNet对术后12个月内复发事件的预测效果最佳。结论ResNet模型有望在临床实践中用于TLE患者术后随访,帮助对术后患者进行风险分层个体化管理。Objective To predict the short-term postoperative recurrence status of patients with refractory temporal lobe epilepsy(TLE)by analyzing preoperative ^(18)F-FDG PET images and patients'clinical characteristics based on deep residual neural network(ResNet).Methods Retrospective analysis was conducted on preoperative I8F-FDG PET images and clinical data of 220 patients with refractory TLE(132 males and 88 females,age 23.0(20.0,30.2)years))in the First Affiliated Hospital of Jinan University between January 2014 and June 2020.ResNet was used to perform high-throughput feature extraction on preprocessed PET images and clinical features,and to perform a postoperative recurrence prediction task for differentiating patients with TLE.The predictive performance of ResNet model was evaluated by ROC curve analysis,and the AUC was compared with that of classical Cox proportional risk model using Delong test.Results Based on PET images combined with clinical feature training,AUCs of the ResNet in predicting 12-,24-,and 36-month postoperative recurrence were 0.895±0.073,0.861±0.058 and 0.754±0.111,respectively,which were 0.717±0.093,0.697±0.081 and 0.645±0.087 for Cox proportional hazards model respectively(z values:-3.00,-2.98,-1.09,P values:0.011,0.018,0.310).The ResNet showed best predictive effect for recurrence events within 12 months after surgery.Conclusion The ResNet model is expected to be used in clinical practice for postoperative follow-up of patients with TLE,helping for risk stratification and individualized management of postoperative patients.

关 键 词：癫痫 颞叶 复发 神经网络(计算机) 正电子发射断层显像术 氟脱氧葡萄糖F18 预测 

分 类 号：R742.1[医药卫生—神经病学与精神病学]