{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T04:56:29Z","timestamp":1790744189473,"version":"4.1.0"},"reference-count":34,"publisher":"MDPI AG","issue":"19","license":[{"start":{"date-parts":[[2026,9,26]],"date-time":"2026-09-26T00:00:00Z","timestamp":1790380800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Objective: Multi-label electrocardiogram (ECG) classification requires modeling morphological patterns across different temporal scales and their sequential relationships while maintaining limited model complexity. This study proposes Multi-Scale Channel-Attentive Mamba (MSCA-Mamba), a compact architecture integrating multi-scale feature extraction, branch-wise recalibration, and state-space sequence modeling. Methods: MSCA-Mamba combines three parallel dilated Conv1D branches, branch-wise Squeeze-and-Excitation modules, and a unidirectional Mamba block. The model was evaluated on the PTB-XL diagnostic superclass task using the official patient-wise split and five random seeds. Component ablations, parameter-matched Conv1D, GRU, LSTM, and Transformer controls, unidirectional\u2013bidirectional comparisons, and GPU benchmarks were performed. Results: MSCA-Mamba achieved a Macro ROC-AUC of 0.8998 \u00b1 0.0022, Macro PR-AUC of 0.7651 \u00b1 0.0036, and Macro F1-score of 0.7038 \u00b1 0.0045 with 121,149 parameters. Multi-scale feature extraction provided the clearest benefit, particularly for macro-averaged metrics. Parameter-matched alternatives showed broadly comparable performance. Bidirectional Mamba achieved a modestly higher Macro ROC-AUC but increased the parameter count to 214,077 and inference latency. Conclusions: MSCA-Mamba provides a compact framework for multi-label ECG classification on PTB-XL. Its primary strength lies in compact architectural integration rather than universal superiority of individual components.<\/jats:p>","DOI":"10.3390\/diagnostics16193135","type":"journal-article","created":{"date-parts":[[2026,9,28]],"date-time":"2026-09-28T10:00:34Z","timestamp":1790589634000},"page":"3135","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["MSCA-Mamba: A Compact Multi-Scale Mamba Network for Multi-Label ECG Classification"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0710-2280","authenticated-orcid":false,"given":"Dilber","family":"Cetintas","sequence":"first","affiliation":[{"name":"Department of Computer Engineering, Malatya Turgut Ozal University, 44210 Malatya, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,9,26]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"2982","DOI":"10.1016\/j.jacc.2020.11.010","article-title":"Global burden of cardiovascular diseases and risk factors, 1990\u20132019: Update from the GBD 2019 study","volume":"76","author":"Roth","year":"2020","journal-title":"J. Am. Coll. Cardiol."},{"key":"ref_2","unstructured":"World Health Organization (WHO) (WHO Fact Sheet, 2025). Cardiovascular diseases (CVDs), WHO Fact Sheet."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"465","DOI":"10.1038\/s41569-020-00503-2","article-title":"Artificial intelligence-enhanced electrocardiography in cardiovascular disease management","volume":"18","author":"Siontis","year":"2021","journal-title":"Nat. Rev. Cardiol."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"7","DOI":"10.1016\/j.jelectrocard.2024.05.085","article-title":"ST segment and T wave abnormalities: A narrative review","volume":"85","author":"Scheffer","year":"2024","journal-title":"J. Electrocardiol."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Kim, J., Shon, B., Kim, S., Cho, J., Seo, J.J., Jang, S.Y., and Jeong, S. (2024). ECG data analysis to determine ST-segment elevation myocardial infarction and infarction territory type: An integrative approach of artificial intelligence and clinical guidelines. Front. Physiol., 15.","DOI":"10.3389\/fphys.2024.1462847"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"108097","DOI":"10.1016\/j.cmpb.2024.108097","article-title":"One-shot screening: Utilization of a two-dimensional convolutional neural network for automatic detection of left ventricular hypertrophy using electrocardiograms","volume":"247","author":"Cai","year":"2024","journal-title":"Comput. Methods Programs Biomed."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"108235","DOI":"10.1016\/j.compbiomed.2024.108235","article-title":"ECG-based data-driven solutions for diagnosis and prognosis of cardiovascular diseases: A systematic review","volume":"172","author":"Corino","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"1574","DOI":"10.1109\/JBHI.2018.2871510","article-title":"Towards end-to-end ECG classification with raw signal extraction and deep neural networks","volume":"23","author":"Xu","year":"2018","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1519","DOI":"10.1109\/JBHI.2020.3022989","article-title":"Deep learning for ECG analysis: Benchmarks and insights from PTB-XL","volume":"25","author":"Strodthoff","year":"2020","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Popa, C.A., and Dogaru, R. (2025). Lightweight and Interpretable CNNs for Multi-Label ECG Diagnosis on the Edge. Proceedings of the 2025 IEEE 21st International Conference on Intelligent Computer Communication and Processing (ICCP), Cluj-Napoca, Romania, 16\u201318 October 2025, IEEE.","DOI":"10.1109\/ICCP68926.2025.11427107"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Pai, R.R., and Bhat, R. (2025). Enhanced ecg beat classification using a hybrid transformer-cnn model on the ptb-xl dataset. Proceedings of the 2025 4th International Conference on Innovative Mechanisms for Industry Applications (ICIMIA), Tirupur, India, 3\u20135 September 2025, IEEE.","DOI":"10.1109\/ICIMIA67127.2025.11200891"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Xu, G., Xu, P., and Liu, W. (2024). Multi-label ECG classification algorithm based on multiscale features and transformer encoder. Proceedings of the 2024 5th International Symposium on Artificial Intelligence for Medicine Science, Amsterdam, The Netherlands, 13\u201317 August 2024, Association for Computing Machinery.","DOI":"10.1145\/3706890.3707012"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Husain, G., Siddiqua, A., and Toma, M. (2025). Evaluating the Performance of DenseNet in ECG Report Automation. Electronics, 14.","DOI":"10.3390\/electronics14091837"},{"key":"ref_14","first-page":"589","article-title":"Benchmarking approaches: Time series versus feature-based machine learning in ECG analysis on the PTB-XL dataset","volume":"316","author":"Bickmann","year":"2024","journal-title":"Stud. Health Technol. Inform."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"\u015amigiel, S., Pa\u0142czy\u0144ski, K., and Ledzi\u0144ski, D. (2021). Deep learning techniques in the classification of ECG signals using R-peak detection based on the PTB-XL dataset. Sensors, 21.","DOI":"10.3390\/s21248174"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Sheth, K.A., Upreti, C., Prusty, M.R., Satapathy, S.K., Mishra, S., and Cho, S.B. (2024). Time-frequency transformation integrated with a lightweight convolutional neural network for detection of myocardial infarction. BMC Med. Imaging, 24.","DOI":"10.1186\/s12880-024-01502-2"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"110605","DOI":"10.1016\/j.bspc.2026.110605","article-title":"Transfer Learning and Multi-Task Optimization for Multi-Modal Heart Disease Prediction: Integrating Clinical Data and ECG Signals with Uncertainty Quantification","volume":"123","author":"Amiri","year":"2026","journal-title":"Biomed. Signal Process. Control"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Atwa, A.E.M., Atlam, E.S., Ahmed, A., Atwa, M.A., Abdelrahim, E.M., and Siam, A.I. (2025). Interpretable deep learning models for arrhythmia classification based on ECG signals using PTB-X dataset. Diagnostics, 15.","DOI":"10.3390\/diagnostics15151950"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Setu, J.H., Pasha, S.T., Halder, N., Ahmed, E., Islam, A., and Amin, M.A. (2025). Multi-Modal Deep Learning for Myocardial Infarction Detection Using ECG Signals and Images. Proceedings of the 2025 ACM International Joint Conference on Pervasive and Ubiquitous Computing, Espoo, Finland, 12\u201316 October 2025, Association for Computing Machinery.","DOI":"10.1145\/3714394.3756348"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"110674","DOI":"10.1016\/j.compbiomed.2025.110674","article-title":"Enhancing automatic multilabel diagnosis of electrocardiogram signals: A masked transformer approach","volume":"196","author":"Zhou","year":"2025","journal-title":"Comput. Biol. Med."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"4004611","DOI":"10.1109\/TIM.2026.3664602","article-title":"Semi-Supervised Contrastive Learning for Label-Efficient Electrocardiogram Noise Detection","volume":"75","author":"Rahman","year":"2026","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Lee, J., and Shin, M. (2025). Cross-database learning framework for electrocardiogram arrhythmia classification using two-dimensional beat-score-map representation. Appl. Sci., 15.","DOI":"10.3390\/app15105535"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"110740","DOI":"10.1016\/j.bspc.2026.110740","article-title":"An integrated architecture for ECG classification combining state-space model Mamba with CNNs","volume":"125","author":"Tian","year":"2026","journal-title":"Biomed. Signal Process. Control"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Qiang, Y., Dong, X., Liu, X., Yang, Y., Hu, F., and Wang, R. (2024). Ecgmamba: Towards ecg classification with state space models. Proceedings of the 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Lisbon, Portugal, 3\u20136 December 2024, IEEE.","DOI":"10.1109\/BIBM62325.2024.10822754"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"7887","DOI":"10.1109\/JBHI.2026.3669893","article-title":"Mamba-Based Prototypical Contrastive Learning with Augmented Feature Separation for Common and Rare Arrhythmia Classification","volume":"30","author":"Guo","year":"2026","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"e70572","DOI":"10.1002\/cpe.70572","article-title":"High-Fidelity Atrial Fibrillation Electrocardiogram Signal Generation Based on Mamba-WGAN-GP","volume":"38","author":"Chen","year":"2026","journal-title":"Concurr. Comput. Pract. Exp."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"176","DOI":"10.1007\/s11760-026-05188-7","article-title":"Dmkformer: A dual-path routing Transformer with hybrid Mamba and KAN for medical time series","volume":"20","author":"Zeng","year":"2026","journal-title":"Signal Image Video Process."},{"key":"ref_28","unstructured":"Wagner, P., Strodthoff, N., Bousseljot, R.D., Samek, W., and Schaeffter, T. (2026, September 24). PTB-XL, A Large Publicly Available Electrocardiography Dataset. Available online: https:\/\/physionet.org\/content\/ptb-xl\/1.0.3\/."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"154","DOI":"10.1038\/s41597-020-0495-6","article-title":"PTB-XL: A Large Publicly Available ECG Dataset","volume":"7","author":"Wagner","year":"2020","journal-title":"Sci. Data"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"792","DOI":"10.1038\/s44360-026-00096-z","article-title":"PhysioNet as a global platform for biomedical research","volume":"1","author":"Pollard","year":"2026","journal-title":"Nat. Health"},{"key":"ref_31","unstructured":"Gu, A., and Dao, T. (2023). Mamba: Linear-time sequence modeling with selective state spaces. arXiv."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Reddy, L., Talwar, V., Alle, S., Bapi, R.S., and Priyakumar, U.D. (2021). Imle-net: An interpretable multi-level multi-channel model for ecg classification. Proceedings of the 2021 IEEE International Conference on Systems, Man, and Cybernetics (SMC), Melbourne, Australia, 17\u201320 October 2021, IEEE.","DOI":"10.1109\/SMC52423.2021.9658706"},{"key":"ref_33","unstructured":"Cheng, R., Zhuang, Z., Zhuang, S., Xie, L., and Guo, J. (2023). Msw-transformer: Multi-scale shifted windows transformer networks for 12-lead ecg classification. arXiv."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Qiang, Y., Dong, X., Liu, X., Yang, Y., Fang, Y., and Dou, J. (2024). Ecgmamba: Towards efficient ecg classification with bissm. arXiv.","DOI":"10.1109\/BIBM62325.2024.10822754"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/19\/3135\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T04:13:26Z","timestamp":1790741606000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/19\/3135"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,26]]},"references-count":34,"journal-issue":{"issue":"19","published-online":{"date-parts":[[2026,10]]}},"alternative-id":["diagnostics16193135"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16193135","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,26]]}}}