Panic Attack Prediction for Patients With Panic Disorder via Machine Learning and Wearable Electrocardiography Monitoring: Model Development and Validation Study
Journal of Medical Internet Research · 2025
DOI: 10.2196/69045 ↗At a glance
Validierungsstudie in Journal of Medical Internet Research, 2025. Written by Hayoung Oh and 8 co-authors. Institution: Department of Artificial Intelligence Convergence, Sungkyunkwan University, 25-2, Seonggyun-gwan-ro, Jongno-gu, Seoul, 03063, Republic of Korea, 82 1…. Major MEDLINE topics: Panic Disorder, Machine Learning, Wearable Electronic Devices, Electrocardiography. 33 references. The full text is freely available in PubMed Central.
Abstract (original language)
Background
Panic attack prediction remains a critical challenge in mental health care due to the high interindividual variability of physiological responses and the limitations of subjective psychological assessments.
Objective
This study aims to develop a multimodal deep learning framework that integrates real-time physiological signals from wearable electrocardiogram (ECG) monitors and psychological assessments to improve the accuracy of panic attack prediction.
Methods
We adapted the ConvNetQuake architecture, originally designed for seismic detection, to extract temporal patterns from ECG signals. The model was pretrained on the PTB-XL ECG dataset and fine-tuned using wearable ECG data collected from adult participants. In parallel, psychological profiles based on the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition criteria and Panic Disorder Severity Scale assessments were encoded as auxiliary inputs. The multimodal framework was evaluated using standard performance metrics.
Results
The proposed model achieved an accuracy of 71.43%, precision of 83.72%, recall of 70.59%, and F1 score of 76.60% in detecting heart rate variability anomalies associated with panic episodes. Experimental comparisons demonstrated that the integration of physiological and psychological modalities significantly outperformed unimodal baselines in prediction reliability.
Conclusions
This study provides empirical support for wearable-based early warning systems for panic attacks. The proposed approach demonstrates the feasibility of just-in-time digital interventions and underscores the potential of wearable artificial intelligence in advancing affective computing and digital psychiatry.
Subjects
MeSH terms (MEDLINE)
Author keywords
Authors and institutions
- Department of Artificial Intelligence Convergence, Sungkyunkwan University, 25-2, Seonggyun-gwan-ro, Jongno-gu, Seoul, 03063, Republic of Korea, 82 10-5389-5996.
- Department of Mechanical Engineering, Sungkyunkwan University, Seoul, Republic of Korea.
- Department of Child Psychology and Education, Sungkyunkwan University, Seoul, Republic of Korea.
- Department of Computer Science and Engineering, Sungkyunkwan University, Seoul, Republic of Korea.
- Department of Applied Data Science, Sungkyunkwan University, Seoul, Republic of Korea.
- Department of Applied Artificial Intelligence, Sungkyunkwan University, Seoul, Republic of Korea.
Cite
APA
Oh, H., Do, H., Maeng, C., Park, J., Yoon, T., Kim, J., Hwang, H., Choi, S., & Huilin, P. (2025). Panic Attack Prediction for Patients With Panic Disorder via Machine Learning and Wearable Electrocardiography Monitoring: Model Development and Validation Study. Journal of Medical Internet Research. https://doi.org/10.2196/69045
Vancouver
Oh H, Do H, Maeng C, Park J, Yoon T, Kim J, et al. Panic Attack Prediction for Patients With Panic Disorder via Machine Learning and Wearable Electrocardiography Monitoring: Model Development and Validation Study. Journal of Medical Internet Research. 2025. doi:10.2196/69045
BibTeX
@article{Oh2025,
author = {Oh, Hayoung and Do, Hunmin and Maeng, Chaehyun and Park, Jinsuk and Yoon, Taejun and Kim, Jihwan and Hwang, Hyeran and Choi, Seoin and Huilin, Piao},
title = {Panic Attack Prediction for Patients With Panic Disorder via Machine Learning and Wearable Electrocardiography Monitoring: Model Development and Validation Study},
journal = {Journal of Medical Internet Research},
year = {2025},
doi = {10.2196/69045}
}
EUVIMED