Predicting the therapeutic response to valproic acid in childhood absence epilepsy through electroencephalogram analysis using machine learning

• Published in: Epilepsy & behavior 2024-02, Vol.151, p.109647-109647, Article 109647
• Main Authors: Li, Sheng-Ping, Lin, Lung-Chang, Yang, Rei-Cheng, Ouyang, Chen-Sen, Chiu, Yi-Hung, Wu, Mu-Han, Tu, Yi-Fang, Chang, Tung-Ming, Wu, Rong-Ching
• Format: Article
• Language: English
• Subjects: Childhood absence epilepsy; Quantitative electroencephalogram; Therapeutic response; Valproic acid
• ISSN: 1525-5050; 1525-5069
• DOI: 10.1016/j.yebeh.2024.109647

•Approximately 25% of patients with childhood absence seizure respond poorly to first-line antiseizure medication.•A reliable method for predicting first-line medication responsiveness is lacking in patients with childhood absence seizure.•In this study, we used the quantitative electroencephalogram features along with machine learning to predict the therapeutic effects of valproic acid in this population.•The responders had significantly higher alpha band power and lower delta band power than the nonresponders.•Using KNN classification with theta band power in the temporal lobe had best performance, with sensitivity of 92.31%, specificity of 76.92%, accuracy of 84.62%, and area under the curve of 88.46%. Childhood absence epilepsy (CAE) is a common type of idiopathic generalized epilepsy, manifesting as daily multiple absence seizures. Although seizures in most patients can be adequately controlled with first-line antiseizure medication (ASM), approximately 25 % of patients respond poorly to first-line ASM. In addition, an accurate method for predicting first-line medication responsiveness is lacking. We used the quantitative electroencephalogram (QEEG) features of patients with CAE along with machine learning to predict the therapeutic effects of valproic acid in this population. We enrolled 25 patients with CAE from multiple medical centers. Twelve patients who required additional medication for seizure control or who were shifted to another ASM and 13 patients who achieved seizure freedom with valproic acid within 6 months served as the nonresponder and responder groups. Using machine learning, we analyzed the interictal background EEG data without epileptiform discharge before ASM. The following features were analyzed: EEG frequency bands, Hjorth parameters, detrended fluctuation analysis, Higuchi fractal dimension, Lempel–Ziv complexity (LZC), Petrosian fractal dimension, and sample entropy (SE). We applied leave-one-out cross-validation with support vector machine, K-nearest neighbor (KNN), random forest, decision tree, Ada boost, and extreme gradient boosting, and we tested the performance of these models. The responders had significantly higher alpha band power and lower delta band power than the nonresponders. The Hjorth mobility, LZC, and SE values in the temporal, parietal, and occipital lobes were higher in the responders than in the nonresponders. Hjorth complexity was higher in the nonresponders than in the responders in almost all the brain