{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T14:22:32Z","timestamp":1774621352151,"version":"3.50.1"},"reference-count":26,"publisher":"MDPI AG","issue":"12","license":[{"start":{"date-parts":[[2025,11,22]],"date-time":"2025-11-22T00:00:00Z","timestamp":1763769600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Computation"],"abstract":"<jats:p>This study proposes an enhanced deep learning framework for accurate detection of P-peaks in noisy photoplethysmographic (PPG) signals, utilizing a hybrid architecture that integrates wavelet-based analysis with neural network components. The P-peak detection task is formulated as a binary classification problem, where the model learns to identify the presence of a peak at each time step within fixed-length input windows. A temporal attention mechanism is incorporated to dynamically focus on the most informative regions of the signal, improving both localization and robustness. The proposed architecture combines Discrete Wavelet Transform (DWT) for multiscale signal decomposition, Convolutional Neural Networks (CNNs) for morphological feature extraction, and Long Short-Term Memory (LSTM) networks for capturing temporal dependencies. A temporal attention layer is introduced after the recurrent layers to enhance focus on time steps with the highest predictive value. An evaluation was conducted on 30 model variants, exploring different combinations of input types, decomposition levels, and activation functions. The best-performing model\u2014Type30, which includes DWT (3 levels), CNN, LSTM, and attention\u2014achieves an accuracy of 0.918, precision of 0.932, recall of 0.957, and F1-score of 0.923. The findings demonstrate that attention-enhanced hybrid architectures are particularly effective in handling signal variability and noise, making them highly suitable for real-world applications in wearable PPG monitoring, digital twins for Heart Rate Variability (HRV), and intelligent health systems.<\/jats:p>","DOI":"10.3390\/computation13120273","type":"journal-article","created":{"date-parts":[[2025,11,24]],"date-time":"2025-11-24T10:24:17Z","timestamp":1763979857000},"page":"273","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Improved PPG Peak Detection Using a Hybrid DWT-CNN-LSTM Architecture with a Temporal Attention Mechanism"],"prefix":"10.3390","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8017-5537","authenticated-orcid":false,"given":"Galya","family":"Georgieva-Tsaneva","sequence":"first","affiliation":[{"name":"Institute of Robotics, Bulgarian Academy of Science, 1113 Sofia, Bulgaria"}]}],"member":"1968","published-online":{"date-parts":[[2025,11,22]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"14","DOI":"10.2174\/157340312801215782","article-title":"On the analysis of fingertip photoplethysmogram signals","volume":"8","author":"Elgendi","year":"2012","journal-title":"Curr. Cardiol. Rev."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1902","DOI":"10.1109\/TBME.2015.2406332","article-title":"Photoplethysmography-Based Heart Rate Monitoring in Physical Activities via Joint Sparse Spectrum Reconstruction","volume":"62","author":"Zhang","year":"2015","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Park, J., Seok, H.S., Kim, S.-S., and Shin, H. (2022). Photoplethysmogram Analysis and Applications: An Integrative Review. Front. Physiol., 12.","DOI":"10.3389\/fphys.2021.808451"},{"key":"ref_4","unstructured":"Goda, M.\u00c1., Charlton, P.H., and Behar, J.A. (2023). Robust peak detection for photoplethysmography signal analysis. arXiv."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Kazemi, K., Laitala, J., Azimi, I., Liljeberg, P., and Rahmani, A.M. (2022). Robust PPG Peak Detection Using Dilated Convolutional Neural Networks. Sensors, 22.","DOI":"10.3390\/s22166054"},{"key":"ref_6","unstructured":"Whiting, S., Moreland, S., Costello, J., Colopy, G., and McCann, C. (2018). Recognising Cardiac Abnormalities in Wearable Device Photoplethysmography (PPG) with Deep Learning. arXiv."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"382","DOI":"10.1016\/j.bspc.2019.02.028","article-title":"Cuffless Blood Pressure Estimation from Electrocardiogram and Photoplethysmogram Using Waveform-Based ANN\u2013LSTM Network","volume":"51","author":"Tanveer","year":"2019","journal-title":"Biomed. Signal Process. Control"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Jeong, D.U., and Lim, K.M. (2021). Combined deep CNN-LSTM network-based multitasking learning architecture for noninvasive continuous blood pressure estimation using difference in ECG-PPG features. Sci. Rep., 11.","DOI":"10.1038\/s41598-021-92997-0"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Mohammadi, H., Tarvirdizadeh, B., Alipour, K., and Ghamari, M. (2025). Cuff-less blood pressure monitoring via PPG signals using a hybrid CNN-BiLSTM deep learning model with attention mechanism. Sci. Rep., 15.","DOI":"10.1038\/s41598-025-07087-2"},{"key":"ref_10","unstructured":"Zuo, C., Zhao, Y., and Ye, J. (2025). TAU: Modeling Temporal Consistency Through Temporal Attentive U-Net for PPG Peak Detection. arXiv."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Sarkar, P., and Etemad, A. (2020). CardioGAN: Attentive Generative Adversarial Network with Dual Discriminators for Synthesis of ECGfrom PPG. arXiv.","DOI":"10.1609\/aaai.v35i1.16126"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Almarshad, M.A., Islam, M.S., Al-Ahmadi, S., and BaHammam, A.S. (2022). Diagnostic Features and Potential Applications of PPG Signal in Healthcare: A Systematic Review. Healthcare, 10.","DOI":"10.3390\/healthcare10030547"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Liu, Z., Zhang, Y., and Zhou, C. (2024). BiGRU-attention for Continuous blood pressure trends estimation through single channel PPG. Comput. Biol. Med., 168.","DOI":"10.1016\/j.compbiomed.2023.107795"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"149","DOI":"10.1038\/s41597-023-02020-6","article-title":"A benchmark for machine-learning based non-invasive blood pressure estimation using photoplethysmogram","volume":"10","author":"Hsieh","year":"2023","journal-title":"Sci. Data"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"135","DOI":"10.1016\/j.neunet.2023.04.026","article-title":"Long short-term memory with activation on gradient","volume":"164","author":"Qin","year":"2023","journal-title":"Neural Netw."},{"key":"ref_16","unstructured":"Arpit, D., Kanuparthi, B., Kerg, G., Ke, N.R., Mitliagkas, I., and Bengio, Y. (2018). h-detach: Modifying the LSTM Gradient Towards Better Optimization. arXiv."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Mej\u00eda-Mej\u00eda, E., and Kyriacou, P.A. (2022). Photoplethysmography-Based Pulse Rate Variability and Haemodynamic Changes in the Absence of Heart Rate Variability: An In-Vitro Study. Appl. Sci., 12.","DOI":"10.3390\/app12147238"},{"key":"ref_18","unstructured":"Zhao, J., Huang, F., Lv, J., Duan, Y., Qin, Z., Li, G., and Tian, G. (2020). Do RNN and LSTM have Long Memory?. arXiv."},{"key":"ref_19","unstructured":"Pimentel, M., Johnson, A., Charlton, P., and Clifton, D. (2025, October 09). BIDMC PPG and Respiration Dataset. Available online: https:\/\/physionet.org\/content\/bidmc\/1.0.0\/."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"E215","DOI":"10.1161\/01.CIR.101.23.e215","article-title":"PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals","volume":"101","author":"Goldberger","year":"2000","journal-title":"Circulation"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Georgieva-Tsaneva, G., Cheshmedzhiev, K., and Lebamovski, P. (2024). A Wavelet Based Hybrid Method for Time Interval Series Determining. CompSysTech \u201824: Proceedings of the International Conference on Computer Systems and Technologies 2024, ACM International Conference Proceeding Series, Association for Computing Machinery.","DOI":"10.1145\/3674912.3674913"},{"key":"ref_22","unstructured":"Georgieva-Tsaneva, G.N., Tsanev, Y.A., and Cheshmedzhiev, K. (2025, January 9\u201311). Deep-SimPPG: A GAN-Based Hybrid Framework for Realistic Photoplethysmographic Signal Synthesis. Proceedings of the International Conference Automatics and Informatics 2025 (ICAI\u201925), Varna, Bulgaria."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"5438","DOI":"10.1109\/JBHI.2025.3548771","article-title":"Dynamic Beat-to-Beat Measurements of Blood Pressure Using Multimodal Physiological Signals and a Hybrid CNN-LSTM Model","volume":"29","author":"Xiang","year":"2025","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Esgalhado, F., Fernandes, B., Vassilenko, V., Batista, A., and Russo, S. (2021). The Application of Deep Learning Algorithms for PPG Signal Processing and Classification. Computers, 10.","DOI":"10.3390\/computers10120158"},{"key":"ref_25","first-page":"777","article-title":"Algorithm for Real-Time Pulse Wave Detection Dedicated to Non-Invasive Pulse Sensing","volume":"39","author":"Iliev","year":"2012","journal-title":"Comput. Cardiol."},{"key":"ref_26","first-page":"1","article-title":"A Robust Beat-to-Beat Artifact Detection Algorithm for Pulse Wave","volume":"2020","author":"Hu","year":"2020","journal-title":"Math. Probl. Eng."}],"container-title":["Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2079-3197\/13\/12\/273\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,24]],"date-time":"2025-11-24T11:10:03Z","timestamp":1763982603000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2079-3197\/13\/12\/273"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,22]]},"references-count":26,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["computation13120273"],"URL":"https:\/\/doi.org\/10.3390\/computation13120273","relation":{},"ISSN":["2079-3197"],"issn-type":[{"value":"2079-3197","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,22]]}}}