Smart Shirt Uses AI to Detect Epileptic Seizures in Real Time

A research team at Université de Montréal has developed a wearable smart shirt that can detect epileptic seizures in real time using artificial intelligence and integrated biometric sensors. Designed for comfort, discretion, and continuous monitoring, the shirt offers a promising alternative to traditional seizure detection systems, which often rely on bulky equipment or invasive procedures.

The project’s goal was to create a textile-based solution that could monitor physiological signals associated with seizures and alert caregivers or medical personnel when an episode occurs. The result is a sensor-embedded shirt that tracks respiration, movement, and heart rate, all while maintaining the look and feel of everyday clothing.

At the core of the system is a machine learning model trained to recognize the physiological signatures of epileptic seizures. The team collected data from patients during clinical evaluations and used this information to train the algorithm to distinguish between normal activity and seizure-related patterns. Once trained, the model was embedded into a microcontroller that processes incoming signals from the shirt’s sensors in real time.

In laboratory tests, the smart shirt demonstrated high sensitivity and specificity in detecting seizures. The researchers emphasized that the system is designed to minimize false alarms while ensuring timely alerts. When a seizure is detected, the shirt can send a notification to a connected device, such as a smartphone or tablet, allowing caregivers to respond quickly and appropriately.

One of the key advantages of the shirt is its wearability. Unlike traditional monitoring systems that require electrodes, wires, or hospital-based equipment, the smart shirt can be worn comfortably throughout the day. This makes it suitable for long-term use in real-world settings, including at home, at school, or during sleep. The team believes that this level of comfort and discretion could improve adherence and reduce the stigma often associated with epilepsy monitoring.