People living with epilepsy could gain a better understanding of their condition through a new artificial intelligence system that allows them to ask questions about their brain activity and receive answers in real time.

The findings, published in Nature Medicine, show how the technology could help patients take a more active role in managing epilepsy. The study, conducted by researchers at the University of Pennsylvania in the United States, found that feedback from patients helped improve seizure detection and reduced false alarms by about 80 per cent.

Epilepsy is a neurological condition that causes repeated seizures, which occur when there is a sudden burst of unusual electrical activity in the brain. Seizures can affect a person's movement, awareness, behaviour and senses, with symptoms varying from one individual to another.

People with epilepsy may undergo monitoring to help doctors understand their seizures and determine the best way to manage the condition. However, the information collected by monitoring devices is often reviewed by medical specialists, meaning patients may not immediately understand what is happening in their brains.

The new system seeks to change this by allowing patients to communicate with their monitoring devices through a smartphone application.

The technology connects to electroencephalography (EEG) devices, which record electrical activity in the brain through electrodes placed on the scalp or, in some cases, within the skull. The system uses artificial intelligence to analyse the information and respond to patients' questions about their brain activity.

For example, patients can ask about their brain activity during sleep or whether a seizure may have occurred. The application can also send alerts and ask patients to provide information about symptoms or events they have experienced.

Researchers tested the system on 13 patients admitted to the epilepsy monitoring unit at the Hospital of the University of Pennsylvania. The participants used the application over several days while their brain activity was being monitored.

During the study, the system answered more than 92 per cent of questions about brain activity with information considered accurate and relevant. Patients and the system exchanged more than 1,300 messages, showing that participants were able to use the technology to seek information about their condition.

“The system also allowed patients to respond to seizure alerts by indicating whether they believed an event was a genuine seizure, a false alarm or an event they were unsure about.” According to researchers

This feedback helped the system improve its ability to distinguish possible seizures from other brain activity. Researchers found that the rate of false alarms fell by about 80 per cent as the system incorporated patient feedback. Using feedback from medical experts reduced false alarms further.

False alarms can be a challenge in seizure monitoring because devices may sometimes mistake other patterns of brain activity for seizures. Repeated alerts can cause unnecessary concern for patients and require specialists to spend additional time reviewing recordings.

The study suggests that patients' own observations could help improve how monitoring systems recognise seizure-related activity.

“This proof-of-concept showcases a technology with the exciting potential to shift management of epilepsy and possibly other diseases in a more adaptive direction that puts patients in the driver’s seat,” said Amy Bany Adams, acting director of the National Institute of Neurological Disorders and Stroke, which is part of the US National Institutes of Health.

The findings point to a possible shift in how medical devices are used, moving beyond collecting information for doctors to systems that also communicate with patients and learn from their responses.

However, the researchers cautioned that the technology is still at an early stage. The study involved only 13 patients in a hospital setting, where medical teams were available to monitor them. More research is needed to establish whether the system will work safely and effectively for people managing epilepsy at home.

The team is planning larger studies involving patients outside hospital settings to assess how the technology performs during everyday activities.

Researchers are also exploring whether a similar approach could be used with deep brain stimulation devices for people with Parkinson's disease. These devices deliver electrical impulses to specific parts of the brain to help manage symptoms.

The system is not yet a replacement for medical care, and patients should continue to rely on their healthcare teams for diagnosis and treatment decisions.

Its potential lies in helping patients better understand the information collected by their devices, report their experiences, and contribute to improving how seizure alerts are generated.

If further studies confirm its safety and effectiveness, the technology could help make epilepsy monitoring more interactive and responsive to patients' needs.