Study to Provide Real World Data for A.I. Seizure Forecasts

Embrace2 smart watch (Andres Babic, Empatica Inc.)

30 Nov. 2023. A developer of medical monitoring systems with data from wearable devices will soon start asking individuals with epilepsy to offer data for an algorithm that predicts seizures. Empatica Inc. in Boston today announced the study that aims to capture evidence from people with epilepsy in the general population, for an algorithm demonstrated earlier with a few participants.

Empatica designs and markets wearable devices for research, clinical trials, and monitoring health conditions. One of its devices is a smart watch called Embrace2 for people with epilepsy that detects possible convulsive seizures, including during sleep. Epilepsy is a neurological condition affecting more than one percent of the U.S. population causing recurring seizures. Those seizures can range from brief loss of awareness to muscular twitching or shaking and loss of consciousness called convulsions. The Empatica seizure-detector watch, cleared by FDA, detects changes in electrical activity on the skin, and connects wirelessly to a phone app that alerts parents or caregivers of a seizure, as well as another diary-style app that shows patterns of seizures.

Until recently, forecasting rather than detecting the onset of seizures required capturing data with an electroencephalogram or EEG that requires wearing a helmet-like device with electrodes attached to the scalp. In Nov. 2021, a team from the Mayo Clinic and other institutions reported on an algorithm that processes data captured from a clinical-grade Empatica watch. This device detects changes in several physiological indicators in addition to electrical activity on the skin. The team developed a recurrent neural network, a machine-learning algorithm that processes data in sequence over time, which accurately predicted seizures in five of six subjects with epilepsy.

Physiological and lifestyle indicators, and seizure experiences

The new initiative seeks to validate the earlier proof-of-concept findings, as well as further develop the algorithm with data and experiences from the general population of people with epilepsy. The study expects to capture data from participating individuals with an Empatica watch and app, measuring several physiological and lifestyle indicators, along with their seizure experiences. Empatica expects to begin enrolling participants in Jan. 2024, but interested individuals can pre-register on the company web site.

“Seizure forecasting has long been among the most-requested features for people with epilepsy,” says Rosalind Picard, Empatica co-founder and chief scientist in a company statement released through BusinessWire. Picard adds, “Patients with epilepsy understand the toll that uncertainty around seizures takes and we hope that this study will help give them better control over their life, reducing stress and perhaps also enabling early interventions that ultimately reduce or prevent seizures from happening.”

Empatica is spun-off from the Media Lab at Massachusetts Institute of Technology, where Picard is director of affective computing research. Science & Enterprise reported most recently on the company in Nov. 2022, when FDA cleared Empatica’s continuous patient monitoring system that collects real-time data on four digital health indicators for home patient care.

More from Science & Enterprise:

• Home V.R. Therapy Shown to Reduce Back Pain
• FDA Funds Study of Real World Data for Biosimilars
• Small Biz Award Funds Wearable Drug Delivery Patch
• Challenge Seeks Point-of-Care, Home Maternal Health Technologies
• NIH Challenge Seeks Early Alzheimer’s Indicators with A.I.

We designed Science & Enterprise for busy readers including investors, researchers, entrepreneurs, and students. Except for a narrow cookies and privacy strip for first-time visitors, we have no pop-ups blocking the entire page, nor distracting animated GIF graphics. If you want to subscribe for daily email alerts, you can do that here, or find the link in the upper left-hand corner of the desktop page. The site is free, with no paywall. But, of course, donations are gratefully accepted.

*     *     *