MDT-HMTL: A Multidimensional Temporal Hierarchical Multi-Task Learning Framework for Myocardial Infarction Detection and State Classification Using PPG Signals
Sensors · 2026
DOI: 10.3390/s26196259 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, 2026. Verfasst von Hui Hui und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Bioengineering, Chongqing University, Chongqing 400030, China. Hauptthemen laut MEDLINE: Myocardial Infarction, Photoplethysmography, Machine Learning. Die Arbeit nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Graduate Research and Innovation Foundation of Chongqing.
Abstract (Originalsprache)
Continuous assessment of myocardial infarction (MI) using photoplethysmography (PPG) signals has potential for out-of-hospital cardiovascular monitoring, yet conventional single-channel and short-window approaches have limitations in capturing complex hemodynamic variations and dynamic disease-state transitions. To address these limitations, we propose MDT-HMTL, a Multidimensional Temporal Hierarchical Multi-Task Learning framework tailored for simultaneous MI detection and disease-state classification. By integrating multi-channel physiological representations with hierarchical temporal modeling, MDT-HMTL captures local morphological characteristics and longer-range temporal dependencies within a unified multi-task architecture. Evaluation on the MIMIC-III clinical database achieved an average area under the ROC curve (AUC) of 0.908 and an overall accuracy of 86.4%. Furthermore, ablation studies demonstrated the contributions of multi-dimensional representation learning and hierarchical temporal modeling to the overall performance. The proposed framework demonstrates the feasibility of PPG-based MI detection and longitudinal disease-state assessment, highlighting its potential for future non-invasive and wearable-compatible cardiovascular monitoring.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- College of Bioengineering, Chongqing University, Chongqing 400030, China.
- Key Laboratory of Biorheological Science and Technology, Chongqing University, Ministry of Education, Chongqing 400044, China.
Förderung
- National Natural Science Foundation of China
- Graduate Research and Innovation Foundation of Chongqing
Zitieren
APA
Hui, H., Wang, J., Cheng, J., Zhao, H., & Ji, Z. (2026). MDT-HMTL: A Multidimensional Temporal Hierarchical Multi-Task Learning Framework for Myocardial Infarction Detection and State Classification Using PPG Signals. Sensors. https://doi.org/10.3390/s26196259
Vancouver
Hui H, Wang J, Cheng J, Zhao H, Ji Z. MDT-HMTL: A Multidimensional Temporal Hierarchical Multi-Task Learning Framework for Myocardial Infarction Detection and State Classification Using PPG Signals. Sensors. 2026. doi:10.3390/s26196259
BibTeX
@article{Hui2026,
author = {Hui, Hui and Wang, Jun and Cheng, Jinxiu and Zhao, Huimeng and Ji, Zhong},
title = {MDT-HMTL: A Multidimensional Temporal Hierarchical Multi-Task Learning Framework for Myocardial Infarction Detection and State Classification Using PPG Signals},
journal = {Sensors},
year = {2026},
doi = {10.3390/s26196259}
}
EUVIMED