ECG Signal Compression and Reconstruction Based on CNN-LSTM-Attention Model.
W, L., D, C., Z, Z., Y, C., Y, L., Z, G., & L, L. (2026). ECG Signal Compression and Reconstruction Based on CNN-LSTM-Attention Model.. Sensors (Basel, Switzerland). https://doi.org/10.3390/s26133983
W L, D C, Z Z, Y C, Y L, Z G, et al. ECG Signal Compression and Reconstruction Based on CNN-LSTM-Attention Model.. Sensors (Basel, Switzerland). 2026; doi: 10.3390/s26133983
W L, D C, Z Z, et al. ECG Signal Compression and Reconstruction Based on CNN-LSTM-Attention Model.[J]. Sensors (Basel, Switzerland). 2026. DOI: 10.3390/s26133983.
@article{w2026,
author = {Liu W and Chen D and Zhang Z and Cao Y and Liu Y and Gui Z and Liu L},
title = {ECG Signal Compression and Reconstruction Based on CNN-LSTM-Attention Model.},
journal = {Sensors (Basel, Switzerland)},
year = {2026},
doi = {10.3390/s26133983},
note = {PMID: 42451227},
}
TY - JOUR AU - Liu W AU - Chen D AU - Zhang Z AU - Cao Y AU - Liu Y AU - Gui Z AU - Liu L TI - ECG Signal Compression and Reconstruction Based on CNN-LSTM-Attention Model. T2 - Sensors (Basel, Switzerland) PY - 2026 DO - 10.3390/s26133983 AN - PMID:42451227 ER -
The high prevalence of cardiovascular diseases and the extensive application wearable electrocardiogram (ECG) devices for long-term monitoring have posed significant challenges for the transmission, storage, and real-time processing of massive amounts of ECG data. Consequently, efficient ECG compression and reconstruction have become a research priority in remote ECG monitoring. Traditional compressed sensing is complex and has high computational overhead, while single deep learning models cannot simultaneously extract local waveforms and model temporal dependencies. To address these shortcomings in the reconstruction process, this paper presents a CNN-LSTM-Attention hybrid model. This model utilizes a convolutional neural network (CNN) to capture local ECG waveform features, employs a long short-term memory (LSTM) network to learn long-term temporal dependencies, and introduces an attention mechanism to weight and fuse key diagnostic features, enabling accurate focus on key components including the QRS complex and ST segment. Experimental results on the MIT-BIH Arrhythmia dataset demonstrate that across the full compression range of 0.1-0.9, the proposed model achieves favorable comprehensive performance. Its PRD is stabilized at 10-12%, the SNR stays above 20 dB, and the RMSE is mostly lower than 0.25 mV. In terms of reconstruction accuracy and stability, our model outperforms the single CNN and CNN-LSTM models by a large margin.