TIGP (SNHCC) -- AI in ECG for Cardiac Arrhythmia Classification: Super-Resolution Techniques and Applications
- LecturerDr. Tsai-Min Chen (Research Center for Information Technology Innovation, Academia Sinica)
Host: TIGP (SNHCC) - Time2025-03-31 (Mon.) 14:00 ~ 16:00
- LocationAuditorium 106 at IIS new Building
Abstract
Electrocardiography (ECG) signals are crucial for diagnosing cardiac arrhythmias, yet continuous high-resolution monitoring remains challenging due to battery constraints and data transmission limitations in wearable devices. To address this issue, Dr. Tsai-Min Chen presents an innovative AI-driven framework utilizing super-resolution techniques (SRECG) to enhance low-resolution ECG signals for accurate arrhythmia classification. His approach integrates convolutional neural networks (CNNs) and gated recurrent units (GRUs) with attention mechanisms, optimizing data resolution and power efficiency. This pioneering work significantly improves ECG classification accuracies at lower sampling frequencies, crucial for practical deployment in consumer electronics. Furthermore, Dr. Chen points out the broader potential of the SRECG methodology for other biomedical signals, such as EEG and EMG, underscoring its impact across interdisciplinary domains.
BIO
Dr. Tsai-Min Chen is a Postdoctoral Scholar at Academia Sinica's AI Cooperative, known for his interdisciplinary expertise bridging data science, biomedical signal processing, and AI applications. He earned his PhD in Data Science from National Taiwan University in 2024, focusing on AI-enhanced ECG classification for cardiac arrhythmias. His accolades include the 2023 Google PhD Fellowship and the IEEE Chester Sall Award (2023). Dr. Chen previously served as an AI Consultant at Taiwan AI Academy and Research Assistant at Academia Sinica's Institute of Biomedical Sciences. His research spans diverse fields, reflecting a profound commitment to integrating AI and advanced computational methods into medical and consumer electronics.