Interdisciplinary Team Needed to Apply Machine Learning in Epilepsy Surgery: Lara Jehi, MD, MHCDS

WATCH TIME: 4 minutes

"This project wouldn’t have happened without a close partnership between clinical epilepsy expertise and biomedical engineering. The success lies in both sides—clinicians and data scientists—working as equal partners, respecting each other's contributions, and adapting along the way."

Surgical brain resection has been an option for patients with drug resistant epilepsy (DRE) for many years; however, only half of patients who opt for this approach achieve sustained seizure freedom. After years of research, members at Cleveland Clinic designed a machine learning tool that uses 5 minutes of peri-ictal scalp EEG to predict seizure outcomes in patients undergoing epilepsy surgery. Consistent with recent trends in machine learning applications, the study investigators implemented an AutoML workflow to model building and used the Oral Data Science platform.

Published in Nature, the tool was tested on 294 patients who underwent temporal lobe resection for seizures. Overall, investigators showed that machine learning classifiers can make accurate predictions of postoperative seizure outcome, demonstrated by area under the receiver curves of 0.98. A decision curve analysis also revealed that compared with the prevalent clinical-variable based nomogram, use of the EEG-augmented approach could decrease the rate of unsuccessful brain resections by 20%.

Following the publication, lead investigator Lara Jehi, MD, MHCDS, sat down to discuss the team needed to implement such a tool in clinical settings. Jehi, an epilepsy specialist and Cleveland Clinic’s Chief Research and Information Officer, emphasized the importance of a collaborative approach, involving clinical epilepsy expertise, neurosurgery, biomedical engineering, and computer science. By highlighting the contributions of each team member, Jehi underscored how equal partnership and mutual respect between clinicians and data scientists are essential for developing impactful research and meaningful clinical applications.

REFERENCE
1. Sheikh SR, McKee ZA, Ghosn S, et al. Machine learning algorithm for predicting seizure control after temporal lobe resection using peri-ictal. Nature. 2024(14):21771. doi:10.1038/s41598-024-72249-7