KOREAN NET

Panic disorder can cause recurring attacks of extreme anxiety and fear when faced with unexpected situations. Its physical symptoms include shortness of breath, heart palpitations, dizziness and sweating as well as a fear of impending death. (Yonhap News)

By Yoo Yeon Gyeong

A domestic research team has developed a technology that uses artificial intelligence (AI) to detect the risk of panic attack a day in advance.

The team of Yonsei University and Korea University researchers on Dec. 2 released the results of its analysis of daily life data on 43 patients with mood and anxiety disorders for up to two years.

The study's first author was Ph.D. candidate Jang Sooyoung and its corresponding author Dr. Park Yu Rang both from the Department of Biomedical Systems Informatics of Yonsei's College of Medicine, and another corresponding author was Dr. Cho Chul-Hyun from Korea University's psychiatry and biomedical systems departments.

Using a specially designed smartphone app and wearable devices, the team analyzed the patient's biometrics data like heart rate, sleep patterns and number of steps as well as daily habits like everyday mood, energy and anxiety levels, coffee intake and exercise.

The team achieved a 90.5% accuracy rate in predicting a panic attack a day in advance based on analyses of data collected through an AI algorithm and devising a prediction model.

The study's results will be used to develop a digital treatment device to help patients proactively respond to panic attacks.

dusrud21@korea.kr