EEG-Based Machine Learning Models for Predicting Ketogenic Diet Outcomes in Pediatric Drug-Resistant Epilepsy

Ketogenic diet therapy (KDT) is an established treatment for drug-resistant epilepsy (DRE); however, methods for predicting its effectiveness remain underdeveloped. This study evaluated various machine learning models in predicting responses to KDT among DRE patients based on electroencephalography (EEG) data. Using leave-one-out cross-validation, this study evaluated 19 machine learning classifiers in predicting the outcomes of 90 DRE patients based on absolute and relative power, as well as functional connectivity measures (phase locking value, phase lag index, and weighted phase lag index) across standard frequency bands. KDT significantly reduced seizure frequency at three and six months after initiation. The most effective classifier at three months was a Coarse Tree classifier trained on absolute power (recall = 0.933, precision = 0.767, F2 = 0.894, AUC = 0.607). The most effective classifier at six months was a Gaussian Naive Bayes classifier trained on weighted phase lag index + relative power (recall = 0.759, precision = 0.854, F2 = 0.776, AUC = 0.603). This study identified the most effective machine-learning models for predicting KDT outcomes in DRE patients. The results highlight the potential of EEG-based machine learning tools for guiding KDT treatment in clinical practice.