Video-based detection of tonic–clonic seizures using a three-dimensional convolutional neural network
Abstract
Objective
Seizure detection in epilepsy monitoring units (EMUs) is essential for the clinical assessment of drug-resistant epilepsy. Automated video analysis using machine learning provides a promising aid for seizure detection, with resultant reduction in the resources required for diagnostic monitoring. We employ a three-dimensional (3D) convolutional neural network with fully fine-tuned backbone layers to identify seizures from EMU videos.
Methods
A two-stream inflated 3D-ConvNet architecture (I3D) classified video clips as a seizure or not a seizure. A pretrained action classifier was fine-tuned on 11 h of video containing 49 tonic–clonic seizures from 25 patients monitored at a large academic hospital (site A) using leave-one-patient-out cross-validation. Performance was evaluated by comparing model predictions to ground-truth annotations obtained from video-electroencephalographic review by an epileptologist on videos from site A and a separate dataset from a second large academic hospital (site B).
Results
The model achieved a leave-one-patient-out cross-validation F1-score of .960 ± .007 (mean ± SD) and area under the receiver operating curve score of .988 ± .004 at site A. Evaluation on full videos detected all seizures (95% binomial exact confidence interval = 94.1%–100%), with median detection latency of 0.0 s (interquartile range = 0.0–3.0) from seizure onset. The site A model had an average false alarm rate of 1.81 alarms per hour, although 36 of the 49 videos (73.5%) had no false alarms. Evaluation at site B demonstrated generalizability of the architecture and training strategy, although cross-site evaluation (site A model tested on site B data and vice versa) resulted in diminished performance.
Significance
Our model demonstrates high performance in the detection of epileptic seizures from video data using a fine-tuned I3D model and outperforms similar models identified in the literature. This study provides a foundation for future work in real-time EMU seizure monitoring and possibly for reliable, cost-effective at-home detection of tonic–clonic seizures.
CONFLICT OF INTEREST STATEMENT
Z.H. receives research support from Neuro Events Lab and the Department of Defense, CDMRP Virtual Post-Traumatic Epilepsy Research Center Faculty Award HT9425-24-1-0355. R.J.C. receives funding from the Research Accelerator Program of the Shirley Ryan AbilityLab. No commercial conflicts of interest are reported by the authors. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
Open Research
DATA AVAILABILITY STATEMENT
Code and model weights used to generate the results in this study are publicly available at https://github.com/aidanboyne/CVEpilepsy under the Apache 2.0 license. At the time of submission, the video data are not available due to patient privacy issues.
