Conference Coverage

Chest-worn seizure detection device shows promise


 

AT AES 2016

HOUSTON – An investigative, chest-worn device shows promise for detecting a wide range of seizure types in children with epilepsy, results from a small single-center study showed.

“Our goal is for parents to have more peace of mind, not feeling like they have to watch their kids all night long,” one of the study authors, Kristin H. Gilchrist, PhD, said in an interview at the annual meeting of the American Epilepsy Society. “There are currently many wearable devices marketed for fitness [purposes], but if we can leverage some of these heart rate monitors and use them to detect seizures with specialized algorithms, that would be ideal.”

Dr. Kristin H. Gilchrist stands in front of her poster presentation. Doug Brunk/Frontline Medical News

Dr. Kristin H. Gilchrist

In a study funded by a grant from the National Institutes of Health, Dr. Gilchrist, a research scientist at RTI International, a not-for-profit research and development organization based in Research Triangle Park, N.C., and her associates collected data on 50 children undergoing video EEG evaluation at Children’s National Medical Center (CNMC) in Washington by attaching a Zephyr BioPatch sensor to the chest of patients to continuously record electrocardiogram and acceleration. They developed novel methods for accurate extraction of cardiac metrics in real time from mobile subjects and condensed heart and accelerometer data into multiple parameters that differentiate seizure and nonseizure activity. Next, they developed a detection algorithm that was implemented in a compact unit which can be clipped to a belt or placed on a nearby table. A microcontroller performs algorithm computations on data streamed from a Bluetooth-enabled ECG sensor to enable real-time seizure detection and alert.

Approximately 60% of patients had a seizure during the monitoring period. Seizures without any clinical response or those lasting less than 10 seconds (such as single myoclonic jerks or clusters) were excluded from analysis, as were subjects with multiple seizures per hour because the autonomic signals often did not return to baseline, and this seizure frequency is outside of the intended use of the monitor. After exclusions, the algorithm was evaluated on 10 children with a mean age of 12 years. For additional validation, the algorithm was also evaluated with the Massachusetts Institute of Technology, Boston, PhysioNet database with ECG from five adults with partial seizures (Neurology. 1999;53:1590-2).The algorithm without motion parameters detected all seizures classified as tonic-clonic (3/3) or atonic/clonic/tonic (4/4), and 3/9 and 7/10 of focal seizures from the CNMC and MIT subjects, respectively. In the CNMC dataset, the motion algorithm detected all seizures classified as tonic-clonic (3/3), half of those categorized as atonic/clonic/tonic (2/4), and one of nine classified as focal seizures.

In the CNMC dataset, false positives averaged one per 14 hours, however, the majority of false positives occurred in a few patients with poor sensor data quality. More than half of the subjects (70%) had no false positives. One false positive occurred in the 16.8 hours of MIT data.

“In addition to an alert application, we have a technology that can be beneficial to clinic-based studies to quantify how many seizures people are having,” Dr. Gilchrist said. “Hopefully someday it will reach the commercial market.” She reported having no financial disclosures.

Next Article: