Real-Time Patient Interaction With an Epilepsy Monitor Enhances Seizure Detection

October 13, 2026

National Institutes Of Health

Objective:

Evaluate a conversational SI interface that lets patients ask questions about EEG data in real time and provide feedback on detected events.

Approach:

Study design: Over several days, researchers evaluated the interface in 13 patients in the epilepsy monitoring unit at the Hospital of the University of Pennsylvania.
Interface: Participants used a smartphone app to receive EEG-based alerts, surveys, and other notifications, ask questions about their brain activity, and classify detected events as true, false, or uncertain.

Key Findings:

Participants and the system exchanged more than 1,300 messages.
The system provided accurate, relevant information for more than 92% of patient inquiries.
Participants responded to more than 67% of real-time seizure detections.
Patient feedback helped refine the seizure-detection model and reduced its false-alarm rate by approximately 80% over the study period. Expert feedback produced an additional 2% reduction in alarms.

Interpretation:

The authors described the study as a proof of concept for real-time, two-way communication between patients and monitoring devices. They considered the findings promising enough to justify further research and development, including planned large-scale outpatient studies.

Limitations:

The study included 13 participants.
The study took place over several days in a hospital epilepsy monitoring unit.

Conclusion:

The researchers are planning large-scale outpatient studies and are preparing to study the platform for communication between deep brain stimulation devices and people with Parkinson’s disease.