Non-invasive classification of coronary perfusion pressure during CPR using smartphone-based skin video and deep learning
Computer Methods and Programs in Biomedicine · 2026-07 · 281 · 109348
DOI: 10.1016/j.cmpb.2026.109348 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computer Methods and Programs in Biomedicine, 2026-07. Verfasst von Soyoon Kwon und 8 weiteren Autor:innen. Beteiligte Einrichtung: Interdisciplinary Program in Bioengineering, Graduate School, Seoul National University, Seoul 08826, Republic of Korea; Integrated Major in Innovati…. Hauptthemen laut MEDLINE: Smartphone, Deep Learning, Cardiopulmonary Resuscitation, Coronary Circulation. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 14 Literaturangaben. Gefördert von: Ministry of Trade, Industry and Energy, Ministry of Food and Drug Safety, Ministry of Health and Family Welfare.
Abstract (Originalsprache)
BACKGROUND AND OBJECTIVE: Coronary perfusion pressure (CPP) is an important determinant of myocardial blood flow and an indicator during cardiopulmonary resuscitation (CPR). However, conventional CPP monitoring methods are invasive and unsuitable for out-of-hospital settings. This study proposes a non-invasive approach to classify CPP levels using skin video recorded with a smartphone camera and deep learning. METHODS: Video and biosignal data were collected simultaneously from 15 pigs during CPR. An integrated deep learning model that combines backbone model (CNN, EfficientNetV2-B0, ConvNeXt-Nano, FastViT-T8) with gated recurrent unit (GRU) was developed to classify whether CPP exceeded the clinically relevant threshold of 20 mmHg. Gradient-weighted Class Activation Mapping (Grad-CAM) was used to identify regions attended to by the model, and statistical analysis conducted. RESULTS: Among four different backbone models used for training and evaluation, the EfficientNetV2-B0-GRU architecture demonstrated the best performance, achieving accuracy of 84.60 % and F1-score of 76.04 %. Statistical analysis revealed significant differences in YCrCb channel values between correctly classified groups. In addition, active regions of Grad-CAM showed greater variances than inactive regions, indicating that the model focused on regions with higher color variabilities related to perfusion. CONCLUSION: A deep learning-based approach using skin video enables the non-invasive classification of CPP during CPR. Grad-CAM and quantitative YCrCb analyses improve interpretability, while the proposed method provides proof-of-concept for cost-effective and accessible perfusion-related assessment using skin video recorded with a smartphone, with potential future utility for real-time CPR quality assessment and decision-making in emergency or out-of-hospital settings.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Interdisciplinary Program in Bioengineering, Graduate School, Seoul National University, Seoul 08826, Republic of Korea; Integrated Major in Innovative Medical Science, Graduate School, Seoul National University, Seoul…
- Institute of Medical and Biological Engineering, Medical Research Center, Seoul National University, Seoul 03080, Republic of Korea.
- Research Center for Disaster Medicine, Medical Research Center, Seoul National University, Seoul 03080, Republic of Korea; Department of Emergency Medicine, Seoul National University Hospital, Seoul 03080, Republic of K…
- National Medical Center, Seoul 04564, Republic of Korea.
- Graduate School of Convergence Science and Technology, Seoul National University, Seoul 08826, Republic of Korea.
- Graduate School of Convergence Science and Technology, Seoul National University, Seoul 08826, Republic of Korea; Blue Robin inc., Seoul, 06524, Republic of Korea.
Förderung
- Ministry of Trade, Industry and Energy
- Ministry of Food and Drug Safety
- Ministry of Health and Family Welfare
- Ministry of Education
- Korea Medical Device Development Fund
- National Research Foundation
Zitieren
APA
Kwon, S., Shin, D. A., Kim, T., Kim, K. S., Suh, G. J., Sim, J., Hur, S., Park, J., & Lee, J. C. (2026). Non-invasive classification of coronary perfusion pressure during CPR using smartphone-based skin video and deep learning. Computer Methods and Programs in Biomedicine, 281, 109348. https://doi.org/10.1016/j.cmpb.2026.109348
Vancouver
Kwon S, Shin DA, Kim T, Kim KS, Suh GJ, Sim J, et al. Non-invasive classification of coronary perfusion pressure during CPR using smartphone-based skin video and deep learning. Computer Methods and Programs in Biomedicine. 2026;281:109348. doi:10.1016/j.cmpb.2026.109348
BibTeX
@article{Kwon2026,
author = {Kwon, Soyoon and Shin, Dong Ah and Kim, Taegyun and Kim, Kyung Su and Suh, Gil Joon and Sim, Jaehoon and Hur, Sungmoon and Park, Jaeheung and Lee, Jung Chan},
title = {Non-invasive classification of coronary perfusion pressure during CPR using smartphone-based skin video and deep learning},
journal = {Computer Methods and Programs in Biomedicine},
year = {2026},
volume = {281},
pages = {109348},
doi = {10.1016/j.cmpb.2026.109348}
}
EUVIMED