AI platform allows real-time conversation between patients and their medical devices

A new artificial intelligence (AI) platform is enabling real-time conversation between patients and their medical devices, allowing individuals to interact with their devices through a smartphone application and receive feedback promptly. This innovative approach marks a significant departure from traditional medical devices that primarily gather data reviewed periodically by healthcare professionals.

In a proof-of-concept study published in Nature Medicine, researchers from the Perelman School of Medicine at the University of Pennsylvania and the Center for Neuroengineering and Therapeutics tested the platform on 13 adults undergoing continuous brain monitoring using electroencephalogram (EEG) in Penn's Epilepsy Monitoring Unit. The participants engaged with the AI-driven system to ask questions about their brain activity, request visualizations, report symptoms and behaviors, and receive personalized alerts.

The AI platform demonstrated its potential by helping patients better understand their health patterns and allowing both patients and devices to learn from each other over time. The device learned which factors were most relevant to patients, while patients gained insights into how their behaviors affected their health. The system analyzed 1,307 messages exchanged between patients and the platform, primarily discussing seizure events, sleep, mood, energy levels, and other factors that could influence seizure activity.

Patients reported that the platform helped clarify patterns they had previously observed but struggled to understand. Some found the system valuable for tracking sleep, providing warnings before major seizures, and solidifying their understanding of the relationship between their episodes and emotions. Similarly, patients could teach the system about their condition, noting which seizures were correctly identified and which were false alarms, allowing the AI to improve its accuracy.

The researchers envision the platform having broad applications for various conditions, such as cardiac devices, insulin pumps, and treatments for neuropsychiatric disorders like Parkinson's disease, depression, obsessive-compulsive disorder, addiction, and cognitive disorders. In the future, these AI-powered devices could automatically adjust treatments while communicating directly with patients about symptoms and treatment goals.

This two-way communication between patients and medical devices could fundamentally change the relationship between the two, transforming passive monitoring tools into active partners in care. By integrating patient feedback and AI-driven analysis with physiologic signals, future devices could deliver more personalized and responsive care, potentially reducing seizure frequency and severity, improving quality of life, decreasing emergency department visits, and lowering healthcare utilization.