Abstract
AI-based telemonitoring in epilepsy is an emerging field focused on the
theoretical exploration of seizure-related signatures and the identification
of reliable biomarkers using wearable sensor data to accompany people
with epilepsy during and beyond doctor’s visits. Current research
investigates how multimodal physiological signals—such as heart rate,
movement, and electrodermal activity—can be analyzed with machine
learning to detect and characterize epileptic seizures. While promising
approaches are being developed, the field remains in an early phase, with
many questions around signal specificity, individual variability, and clinical
relevance still open. In this talk, I will provide insights into ongoing work
on biomarker discovery and discuss the necessary steps toward clinical
translation of these technologies.