Accuracy of developing technology to predict the risk of panic attack with AI is 90%
Dec 02, 2024
A joint research team from Yonsei University and Korea University has developed a technology that can predict the risk of panic attacks a day in advance using artificial intelligence (AI).
Panic disorder is a disease in which seizures accompanied by extreme anxiety and fear occur repeatedly in unexpected situations, and is characterized by fear of death along with physical symptoms such as difficulty breathing, heart palpitations, dizziness, and sweating. In particular, anticipation anxiety about not knowing when seizures will come can seriously affect daily life, leading to a decline in the quality of life of patients. With the development of predictive technology by the joint research team, it is expected to give new hope to patients with panic attacks. The joint research team (first author Jang Soo-young, PhD, medical life system information department at Yonsei University, corresponding author Park Yoo-rang, medical life system information department professor at Yonsei University, and Cho Chul-hyun, professor of psychiatry at Korea University Anam Hospital) followed and analyzed the daily life data of 43 patients with mood disorders and anxiety disorders for up to two years.
The research team provided patients with specially designed smartphone apps and wearable devices, and collected various data through them. Bio-information such as heart rate, sleep pattern, and number of steps, as well as lifestyle habits such as daily mood status, energy level, anxiety level, coffee intake, and exercise status, were comprehensively analyzed.
The research team analyzed the collected data through AI algorithms and established a predictive model, and as a result, it succeeded in predicting panic attacks that would occur the next day with an accuracy of 90.5%.
Professor Park Yu-rang explained that "one of the most difficult things for patients with panic disorder is anxiety about 'panic attacks that do not know when they will come'", and "This predictive anxiety leads to a vicious cycle that rather aggravates panic symptoms, and our predictive model will help break this loop."
Professor Cho Chul-hyun said, `Until now, the treatment of panic disorder has been focused on dealing with post-seizure attacks"It is of great significance that this study has made a preemptive response possible" he stressed. It also added that "the development of digital therapeutic devices based on this technology is currently underway."
Professor Cho Chul-hyun then said, "We can utilize high-quality data that monitors digital phenotypes, which are data related to individuals' health collected in real time through digital devices such as smartphones and wearable devices, for 24 hours straight."It can be applied practically based on a comprehensive understanding of the patient's daily life beyond the evaluation that was previously conducted only fragmentarily during hospital visits"
In fact, based on the results of this study, digital treatment devices are being developed through the NIPA project. AimMed Co., Ltd., Department of Psychiatry at Korea University Anam Hospital, Department of Psychiatry at Gangbuk Samsung Hospital, and Department of Medical System Information at Yonsei University have formed a consortium to carry out the 「Development and demonstration of digital treatment devices customized for panic disorder based on digital expression AI」 project.
On the other hand, this study was carried out with the support of the Korea Research Foundation's excellent progress project, the National Institute of Information and Communication Industry Promotion (NIPA) project, and the Information and Communication Planning and Evaluation Institute (IITP) project, and the research paper 'A digital phenotyping database for impressing panic symptoms: a prospective longitude study' was published in the international journal 『Scientific Data" .
Panic disorder is a disease in which seizures accompanied by extreme anxiety and fear occur repeatedly in unexpected situations, and is characterized by fear of death along with physical symptoms such as difficulty breathing, heart palpitations, dizziness, and sweating. In particular, anticipation anxiety about not knowing when seizures will come can seriously affect daily life, leading to a decline in the quality of life of patients. With the development of predictive technology by the joint research team, it is expected to give new hope to patients with panic attacks. The joint research team (first author Jang Soo-young, PhD, medical life system information department at Yonsei University, corresponding author Park Yoo-rang, medical life system information department professor at Yonsei University, and Cho Chul-hyun, professor of psychiatry at Korea University Anam Hospital) followed and analyzed the daily life data of 43 patients with mood disorders and anxiety disorders for up to two years.
The research team provided patients with specially designed smartphone apps and wearable devices, and collected various data through them. Bio-information such as heart rate, sleep pattern, and number of steps, as well as lifestyle habits such as daily mood status, energy level, anxiety level, coffee intake, and exercise status, were comprehensively analyzed.
The research team analyzed the collected data through AI algorithms and established a predictive model, and as a result, it succeeded in predicting panic attacks that would occur the next day with an accuracy of 90.5%.
Professor Park Yu-rang explained that "one of the most difficult things for patients with panic disorder is anxiety about 'panic attacks that do not know when they will come'", and "This predictive anxiety leads to a vicious cycle that rather aggravates panic symptoms, and our predictive model will help break this loop."
Professor Cho Chul-hyun said, `Until now, the treatment of panic disorder has been focused on dealing with post-seizure attacks"It is of great significance that this study has made a preemptive response possible" he stressed. It also added that "the development of digital therapeutic devices based on this technology is currently underway."
Professor Cho Chul-hyun then said, "We can utilize high-quality data that monitors digital phenotypes, which are data related to individuals' health collected in real time through digital devices such as smartphones and wearable devices, for 24 hours straight."It can be applied practically based on a comprehensive understanding of the patient's daily life beyond the evaluation that was previously conducted only fragmentarily during hospital visits"
In fact, based on the results of this study, digital treatment devices are being developed through the NIPA project. AimMed Co., Ltd., Department of Psychiatry at Korea University Anam Hospital, Department of Psychiatry at Gangbuk Samsung Hospital, and Department of Medical System Information at Yonsei University have formed a consortium to carry out the 「Development and demonstration of digital treatment devices customized for panic disorder based on digital expression AI」 project.
On the other hand, this study was carried out with the support of the Korea Research Foundation's excellent progress project, the National Institute of Information and Communication Industry Promotion (NIPA) project, and the Information and Communication Planning and Evaluation Institute (IITP) project, and the research paper 'A digital phenotyping database for impressing panic symptoms: a prospective longitude study' was published in the international journal 『Scientific Data" .
|
bellho@sportschosun.com
