Researchers at the University of Pennsylvania's Perelman School of Medicine and Center for Neuroengineering and Therapeutics have developed an artificial intelligence platform that allows patients to interact directly with their medical devices in real time via smartphone text messaging.
To test the platform, researchers enrolled 13 adults undergoing continuous electroencephalogram (EEG) monitoring in Penn’s Epilepsy Monitoring Unit. Built by a team of Penn Bioengineering PhD students—including Zack Goldblum, Haoer Shi, and Zhongchuan Xu—the platform allowed patients to use a phone app to review data visualizations, report daily symptoms like mood and sleep, and receive timely feedback. The analysis examined 1,307 messages exchanged during the study, showing that participants were able to help the system identify true seizure events from false alarms.
Senior author Brian Litt, MD, described the system as functioning like a personal guide that allows the patient and device to learn together over time. Clinicians can also review and add notations through an expert-in-the-loop algorithm designed to reduce the need for constant supervision.
While the initial study focused on epilepsy, researchers believe the technology could eventually be applied to cardiac devices, insulin pumps, and treatments for neuropsychiatric disorders such as Parkinson's disease, depression, and addiction. Ongoing research supported by an NIH Pioneer Award will evaluate outpatient applications, wearable technologies, and whether the platform can help improve quality of life and reduce healthcare utilization.
The study was conducted by researchers at the University of Pennsylvania and published in Nature Medicine.