文献类型：期刊文献

英文题名：Multi-Channel ECG Compression Using SAFD-Based Joint Sparse Coding

作者：Tan, Chunyu[1]; Zhang, Liming[2]; Chen, Luyao[1]; Dai, Lei[3]

机构：[1] School of Artificial Intelligence, Anhui University, Hefei, China; [2] University of Macau, Faculty of Science and Technology, China; [3] East China University of Science and Technology, Department of Computer Science and Engineering, Shanghai, China

年份：2025

起止页码：167

外文期刊名：2025 10th International Conference on Signal and Image Processing, ICSIP 2025

收录：EI(收录号：20254319392236)

语种：英文

外文关键词：Biomedical signal processing - Codes (symbols) - Compaction - Compression ratio (machinery) - Electrocardiograms - Quality control - Signal reconstruction

摘要：This paper proposes a novel joint sparse coding for multi-channel electrocardiogram (MECG) signals compression based on statistical adaptive Fourier decomposition (SAFD). SAFD is a newly developed signal processing tool that is suitable for processing multiple signals. It utilizes stochastic maximal selection principle (SMSP) to find common adaptive basis across all signals, thus achieves rapid convergence with high fidelity. Through SAFD, we propose a joint sparse coding implying the common time-frequency distribution of multi-channel signals for MECG signals compression. In order to better evaluate the efficiency of the proposed compression method, in addition to the generic assessment criteria, we also consider the clinical feature assessment to evaluate the diagnostic quality retention of reconstructed signals. Experiments are conducted on the MECG signal database of the PTB database. The results show that the proposed compression algorithm can achieve excellent signal fidelity while maintaining a high compression ratio, and also preserve clinical information. ? 2025 IEEE.