Seeing Seizures: A Large-Scale Empirical Study of Video and EEG Deep Learning for Automated Rodent Seizure Detection and Severity Grading

A large-scale empirical study of deep learning for automated rodent seizure detection and severity grading, using 24,131 paired video and EEG clips from 601 sessions across 20 rats in a soman-induced epilepsy model, found that R(2+1)D-18 video and GRU EEG models reached detection macro-F1 scores of 0.968 and 0.974 respectively in session-exclusive comparison, but five-class severity grading fell to 0.657 for video and 0.560 for EEG. Under subject-exclusive evaluation, five-class macro-F1 dropped to 0.475 for video and 0.404 for EEG, which the authors say highlights the difficulty of grading unseen animals. A separate video-only binary-detection evaluation on RodEpil achieved a mean seizure-class F1 of 0.979 after uniform re-encoding to reduce identified class-correlated encoding cues, and the authors released their code and GitHub page to support expert-assisted annotation workflows.

arXiv:2610.22566v1 Announce Type: new Abstract: Manual annotation of synchronized video and EEG recordings is a bottleneck in preclinical seizure research. We present an empirical study of seizure detection and behavioral severity grading using 24,131 paired clips from 601 sessions across 20 rats in a soman-induced epilepsy model. Ten video architectures and seven single-channel EEG models are evaluated independently on binary detection and three- and five-class severity tasks. In the session-exclusive reference comparison, R 2+1 D-18 and GRU achieve detection macro-F1 scores of 0.968 and 0.974, respectively; for five-class grading, the corresponding scores are 0.657 and 0.560. Subject-exclusive evaluation reduces five-class macro-F1 to 0.475 for video and 0.404 for EEG, highlighting the difficulty of grading unseen animals. Temporal-sampling experiments demonstrate the value of multiple frames, while detection performance varies relatively little across the evaluated video architectures. A separate video-only binary-detection evaluation on RodEpil achieves a mean seizure-class F1 of 0.979 after uniform re-encoding to reduce identified class-correlated encoding cues. The study characterizes existing models for behavior-defined severity labels under the evaluated protocol, motivating further investigation of expert-assisted annotation workflows. Our code and github page is available here.