Prediction of Pharmacoresistance in Drug-Naïve Temporal Lobe Epilepsy Using Ictal EEGs Based on Convolutional Neural Network
Yiwei Gong1,2 · Zheng Zhang3 · Yuanzhi Yang1 · Shuo Zhang1 · Ruifeng Zheng3,4 · Xin Li3 · Xiaoyun Qiu1 · Yang Zheng5 · Shuang Wang6 · Wenyu Liu7 · Fan Fei1 · Heming Cheng1 · Yi Wang1,2 · Dong Zhou7 · Kejie Huang3 · Zhong Chen1,2,6 · Cenglin Xu1,51 Key Laboratory of Neuropharmacology and Translational Medicine of Zhejiang Province, School of Pharmaceutical Sciences, College of Pharmaceutical Sciences, The Second Afliated Hospital of Zhejiang Chinese Medical University (Xinhua Hospital), Zhejiang Chinese Medical University, Hangzhou 310053, China
2 Institute of Pharmacology & Toxicology, College of Pharmaceutical Sciences, Zhejiang University, Hangzhou 310058, China
3 College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310058, China
4 School of Cyberspace, Hangzhou Dianzi University, Hangzhou 310018, China
5 Department of Neurology, The First Afliated Hospital of Zhejiang Chinese Medical University, Hangzhou 310006, China
6 Epilepsy Center, School of Medicine, Second Afliated Hospital, Zhejiang University, Hangzhou 310009, China
7 Department of Neurology, West China Hospital of Sichuan University, Chengdu 610041, China
Abstract
Approximately 30%–40% of epilepsy patients do not respond well to adequate anti-seizure medications (ASMs), a condition known as pharmacoresistant epilepsy. The management of pharmacoresistant epilepsy remains an intractable issue in the clinic. Its early prediction is important for prevention and diagnosis. However, it still lacks effective predictors and approaches. Here, a classical model of pharmacoresistant temporal lobe epilepsy (TLE) was established to screen pharmacoresistant and pharmaco-responsive individuals by applying phenytoin to amygdaloid-kindled rats. Ictal electroencephalograms (EEGs) recorded before phenytoin treatment were analyzed. Based on ictal EEGs from pharmacoresistant and pharmaco-responsive rats, a convolutional neural network predictive model was constructed to predict pharmacoresistance, and achieved 78% prediction accuracy. We further found the ictal EEGs from pharmacoresistant rats have a lower gamma-band power, which was verified in seizure EEGs from pharmacoresistant TLE patients. Prospectively, therapies targeting the subiculum in those predicted as “pharmacoresistant” individual rats significantly reduced the subsequent occurrence of pharmacoresistance. These results demonstrate a new methodology to predict whether TLE individuals become resistant to ASMs in a classic pharmacoresistant TLE model. This may be of translational importance for the precise management of pharmacoresistant TLE.
Keywords
Pharmacoresistance; Temporal lobe epilepsy; EEG; Prediction; Precision medicine
