{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T11:27:36Z","timestamp":1786966056489,"version":"build-2736575974"},"reference-count":26,"publisher":"MDPI AG","issue":"16","license":[{"start":{"date-parts":[[2026,8,16]],"date-time":"2026-08-16T00:00:00Z","timestamp":1786838400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100002560","name":"Soonchunhyang University","doi-asserted-by":"publisher","award":["No.20250691"],"award-info":[{"award-number":["No.20250691"]}],"id":[{"id":"10.13039\/501100002560","id-type":"DOI","asserted-by":"publisher"}]},{"award":["No.20250691"],"award-info":[{"award-number":["No.20250691"]}],"id":[{"id":"https:\/\/ror.org\/03qjsrb10","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Electrocardiogram (ECG) signals are essential for arrhythmia detection; however, they are frequently degraded by noise during acquisition, and their evaluation is vulnerable to data-leakage and patient-overlap issues that can compromise model assessment. Therefore, in this study, we propose an image-encoding-based arrhythmia classifier combined with a morphology-aware denoising autoencoder. We evaluate signals under a corrected, patient-independent protocol in which every model-selection decision was made on a separate validation partition. On a leakage-free test partition, the autoencoder achieved an SNR improvement of 5.63 dB and a correlation coefficient of 0.801, improving R-peak amplitude preservation and redetection accuracy under moderate-to-severe noise while introducing measurable morphology degradation when the input was already lightly contaminated. The proposed model encodes the denoised, beat-centered signal into images through an interleaved-grouping outer product with sorting and flipping, and classifies them with a multi-scale three-dimensional convolutional network. Under this protocol, the proposed model obtained the highest macro-F1 among five image-encoding architectures, but did not outperform four models operating directly on the denoised signal (macro-F1 32.5% versus 37.3\u201340.0%), indicating that the proposed encoding does not improve overall five-class classification under rigorous inter-patient evaluation. A controlled test showed that the encoding is exactly invariant to global signal-polarity inversion, unlike the sequential models. This targeted invariance, rather than a general accuracy advantage, is the contribution reported here.<\/jats:p>","DOI":"10.3390\/s26165183","type":"journal-article","created":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T11:00:05Z","timestamp":1786964405000},"page":"5183","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A Deep Learning Framework Based on Denoising and 2D Image Encoding for Arrhythmia Classification"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2790-0370","authenticated-orcid":false,"given":"Ji-Yun","family":"Seo","sequence":"first","affiliation":[{"name":"AI Education Center, Soonchunhyang University, Asan 31518, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1625-952X","authenticated-orcid":false,"given":"Byeong Ho","family":"Park","sequence":"additional","affiliation":[{"name":"AI Education Center, Soonchunhyang University, Asan 31518, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9735-6921","authenticated-orcid":false,"given":"Chang Min","family":"Kim","sequence":"additional","affiliation":[{"name":"AI Education Center, Soonchunhyang University, Asan 31518, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,16]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"11","DOI":"10.5334\/gh.1288","article-title":"The heart of the world","volume":"19","author":"Perel","year":"2024","journal-title":"Glob. Heart"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1634","DOI":"10.1016\/j.procs.2024.04.155","article-title":"ECG based heart disease classification: Advancement and review of techniques","volume":"235","author":"Gour","year":"2024","journal-title":"Procedia Comput. Sci."},{"key":"ref_3","first-page":"718","article-title":"ECG signal approximation using Lagrange\u2013Chebyshev polynomials","volume":"10","author":"Yadav","year":"2018","journal-title":"J. Adv. Res. Dyn. Control Syst."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1196","DOI":"10.1109\/TBME.2004.827359","article-title":"Automatic Classification of Heartbeats Using ECG Morphology and Heartbeat Interval Features","volume":"51","author":"Reilly","year":"2004","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_5","first-page":"86","article-title":"ECG noise sources and various noise removal techniques: A survey","volume":"5","author":"Limaye","year":"2016","journal-title":"Int. J. Appl. Innov. Eng. Manag."},{"key":"ref_6","unstructured":"Elgendi, M., Jonkman, M., and De Boer, F. (2010, January 20\u201323). Frequency bands effects on QRS detection. Proceedings of the BIOSIGNALS 2010\u2014Third International Conference on Bio-inspired Systems and Signal Processing, Valencia, Spain."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"451","DOI":"10.1007\/s40031-020-00488-z","article-title":"Arrhythmia detection in ECG signal using fractional wavelet transform with principal component analysis","volume":"101","author":"Gupta","year":"2020","journal-title":"J. Inst. Eng. India Ser. B"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Bhalerao, S., Ansari, I.A., and Kumar, A. (2020). Performance comparison of SVM and ANN for reversible ECG data hiding. Soft Computing: Theories and Applications, Springer.","DOI":"10.1007\/978-981-15-4032-5_20"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1109\/JBHI.2020.2982935","article-title":"An adaptive Kalman filter bank for ECG denoising","volume":"25","author":"Hesar","year":"2020","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"106621","DOI":"10.1016\/j.compeleceng.2020.106621","article-title":"Time-frequency approach to ECG classification of myocardial infarction","volume":"84","author":"Kayikcioglu","year":"2020","journal-title":"Comput. Electr. Eng."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"326","DOI":"10.1109\/TIM.2013.2279001","article-title":"Application of cross wavelet transform for ECG pattern analysis and classification","volume":"63","author":"Banerjee","year":"2013","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"60806","DOI":"10.1109\/ACCESS.2019.2912036","article-title":"Noise reduction in ECG signals using fully convolutional denoising autoencoders","volume":"7","author":"Chiang","year":"2019","journal-title":"IEEE Access"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"100249","DOI":"10.1016\/j.jnlest.2024.100249","article-title":"Efficient ECG classification based on Chi-square distance for arrhythmia detection","volume":"22","author":"Kadhim","year":"2024","journal-title":"J. Electron. Sci. Technol."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"107","DOI":"10.35882\/jeeemi.v6i2.363","article-title":"An approach to ECG-based gender recognition using random forest algorithm","volume":"6","author":"Arif","year":"2024","journal-title":"J. Electron. Electromed. Eng. Med. Inform."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"105884","DOI":"10.1016\/j.bspc.2023.105884","article-title":"A lightweight hybrid CNN\u2013LSTM explainable model for ECG-based arrhythmia detection","volume":"90","author":"Alamatsaz","year":"2024","journal-title":"Biomed. Signal Process. Control"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"120383","DOI":"10.1016\/j.ins.2024.120383","article-title":"Application of spatial uncertainty predictor in CNN\u2013BiLSTM model using coronary artery disease ECG signals","volume":"665","author":"Seoni","year":"2024","journal-title":"Inf. Sci."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1080\/10255842.2024.2378105","article-title":"Arrhythmia detection in inter-patient ECG signals using entropy rate features and RR intervals with CNN architecture","volume":"29","author":"Berrahou","year":"2026","journal-title":"Comput. Methods Biomech. Biomed. Eng."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Ullah, A., Rehman, S.U., Tu, S., Mehmood, R.M., and Ehatisham-ul-Haq, M. (2021). A hybrid deep CNN model for abnormal arrhythmia detection based on cardiac ECG signal. Sensors, 21.","DOI":"10.3390\/s21030951"},{"key":"ref_19","first-page":"8683855","article-title":"Arrhythmia classification algorithm based on a two-dimensional image and modified EfficientNet","volume":"2022","author":"Zhao","year":"2022","journal-title":"Comput. Intell. Neurosci."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"11509","DOI":"10.1002\/int.23041","article-title":"Graph convolutional network-based deep feature learning for cardiovascular disease recognition from heart sound signals","volume":"37","author":"Rezaee","year":"2022","journal-title":"Int. J. Intell. Syst."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"2873","DOI":"10.1007\/s12652-023-04528-6","article-title":"Fusing handcrafted and deep features for multi-class cardiac diagnostic decision support model based on heart sound signals","volume":"14","author":"Jabari","year":"2023","journal-title":"J. Ambient Intell. Humaniz. Comput."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"106441","DOI":"10.1016\/j.bspc.2024.106441","article-title":"A deep learning framework for ECG denoising and classification","volume":"94","author":"Peng","year":"2024","journal-title":"Biomed. Signal Process. Control"},{"key":"ref_23","unstructured":"PhysioNet (2025, October 01). MIT-BIH Arrhythmia Database. Available online: https:\/\/physionet.org\/content\/mitdb\/1.0.0\/."},{"key":"ref_24","unstructured":"PhysioNet (2025, October 01). Noise Stress Test Database. Available online: https:\/\/physionet.org\/content\/nstdb\/1.0.0\/."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Kim, C., and Lee, W. (2023). Human activity recognition by the image type encoding method of 3-axial sensor data. Appl. Sci., 13.","DOI":"10.3390\/app13084961"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"230","DOI":"10.1109\/TBME.1985.325532","article-title":"A real-time QRS detection algorithm","volume":"BME-32","author":"Pan","year":"1985","journal-title":"IEEE Trans. Biomed. Eng."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/16\/5183\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T11:06:09Z","timestamp":1786964769000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/16\/5183"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,16]]},"references-count":26,"journal-issue":{"issue":"16","published-online":{"date-parts":[[2026,8]]}},"alternative-id":["s26165183"],"URL":"https:\/\/doi.org\/10.3390\/s26165183","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,16]]}}}