{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T18:01:33Z","timestamp":1785348093759,"version":"3.55.0"},"reference-count":43,"publisher":"Public Library of Science (PLoS)","issue":"7","license":[{"start":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T00:00:00Z","timestamp":1785283200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>The electrocardiogram (ECG) is an essential non-invasive tool for detecting cardiac abnormalities; however, accurate interpretation often requires specialized expertise that may be unavailable in resource-limited clinical settings. While deep learning models have demonstrated high classification performance, many existing architectures remain computationally intensive and lack assessments of predictive reliability, hindering their deployment in clinical decision support systems. In this study, we propose a lightweight hybrid framework integrating Convolutional Neural Networks (CNN) and Fast Fourier Transform (FFT) components. This architecture combines time-domain morphological representations learned from raw ECG signals with physiologically relevant spectral features to enable accurate and efficient classification. Unlike previous approaches, this work emphasizes reliable model evaluation by incorporating probability calibration and a rigorous patient-wise validation protocol. The proposed method was evaluated on the publicly available PTB-XL dataset using the official 10-fold cross-validation protocol for both binary and five-class multi-label classification. In addition to conventional discrimination metrics, model reliability was assessed using Expected Calibration Error (ECE). The model achieved an accuracy of 92.42% and an AUC of 97.8% for binary classification, alongside a macro-AUC of 92.46% for five- class multi-label classification. Calibration analysis demonstrated well-calibrated probability estimates with low ECE values. Despite its competitive performance, the architecture is highly efficient, containing only 87K parameters and 0.26 GFLOPs. These findings highlight the potential of lightweight hybrid architectures combined with calibration-aware evaluation to support reliable AI-assisted ECG diagnostics in resource-constrained healthcare settings.<\/jats:p>","DOI":"10.1371\/journal.pone.0354834","type":"journal-article","created":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T17:38:31Z","timestamp":1785346711000},"page":"e0354834","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["A lightweight hybrid framework integrating convolutional neural networks and fast Fourier transform for reliable and calibrated ECG-based cardiac abnormality detection"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-6525-8346","authenticated-orcid":true,"given":"Abdoul","family":"Malik","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1231-5782","authenticated-orcid":true,"given":"Selim","family":"Aras","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,7,29]]},"reference":[{"issue":"8","key":"pone.0354834.ref001","article-title":"Heart disease and stroke statistics-2022 update: A report from the American Heart Association","volume":"145","author":"CW Tsao","year":"2022","journal-title":"Circulation"},{"issue":"14","key":"pone.0354834.ref002","doi-asserted-by":"crossref","first-page":"1333","DOI":"10.1056\/NEJMoa1406656","article-title":"Demographic and epidemiologic drivers of global cardiovascular mortality","volume":"372","author":"GA Roth","year":"2015","journal-title":"N Engl J Med"},{"issue":"10","key":"pone.0354834.ref003","doi-asserted-by":"crossref","first-page":"1109","DOI":"10.1016\/j.jacc.2007.01.024","article-title":"Recommendations for the standardization and interpretation of the electrocardiogram: part I","volume":"49","author":"P Kligfield","year":"2007","journal-title":"J Am Coll Cardiol"},{"issue":"3","key":"pone.0354834.ref004","doi-asserted-by":"crossref","first-page":"466","DOI":"10.1016\/j.jacc.2023.12.001","article-title":"The cardiovascular workforce crisis: Navigating the present, planning for the future","volume":"83","author":"TM Maddox","year":"2024","journal-title":"J Am Coll Cardiol"},{"key":"pone.0354834.ref005","doi-asserted-by":"crossref","first-page":"105420","DOI":"10.1016\/j.bspc.2023.105420","article-title":"Classification of cardiac disorders using weighted visibility graph features from ECG signals","volume":"87","author":"G Kutluana","year":"2024","journal-title":"Biomedical Signal Processing and Control"},{"key":"pone.0354834.ref006","doi-asserted-by":"crossref","first-page":"107187","DOI":"10.1016\/j.knosys.2021.107187","article-title":"Deep learning in ECG diagnosis: A review","volume":"227","author":"X Liu","year":"2021","journal-title":"Knowledge-Based Systems"},{"key":"pone.0354834.ref007","author":"P Rajpurkar","year":"2017"},{"issue":"1","key":"pone.0354834.ref008","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1038\/s41591-018-0268-3","article-title":"Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network","volume":"25","author":"AY Hannun","year":"2019","journal-title":"Nat Med"},{"key":"pone.0354834.ref009","volume-title":"Advanced methods and tools for ECG data analysis","author":"GD Clifford","year":"2006"},{"issue":"20","key":"pone.0354834.ref010","doi-asserted-by":"crossref","first-page":"1920","DOI":"10.1161\/CIRCULATIONAHA.115.001593","article-title":"Machine learning in medicine","volume":"132","author":"RC Deo","year":"2015","journal-title":"Circulation"},{"issue":"5","key":"pone.0354834.ref011","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":"N Strodthoff","year":"2021","journal-title":"IEEE J Biomed Health Inform"},{"key":"pone.0354834.ref012","doi-asserted-by":"crossref","first-page":"106006","DOI":"10.1016\/j.cmpb.2021.106006","article-title":"Automated ECG classification using a non-local convolutional block attention module","volume":"203","author":"J Wang","year":"2021","journal-title":"Comput Methods Programs Biomed"},{"key":"pone.0354834.ref013","doi-asserted-by":"crossref","first-page":"105114","DOI":"10.1016\/j.bspc.2023.105114","article-title":"A novel data augmentation approach for enhancement of ECG signal classification","volume":"86","author":"MF Safdar","year":"2023","journal-title":"Biomedical Signal Processing and Control"},{"key":"pone.0354834.ref014","doi-asserted-by":"crossref","unstructured":"Haque A, Akhtar N. Ensemble model for ECG Classification. In: 2022 5th International Conference on Multimedia, Signal Processing and Communication Technologies (IMPACT), 2022. https:\/\/doi.org\/10.1109\/impact55510.2022.10029230","DOI":"10.1109\/IMPACT55510.2022.10029230"},{"key":"pone.0354834.ref015","doi-asserted-by":"crossref","first-page":"102179","DOI":"10.1016\/j.artmed.2021.102179","article-title":"Real-time frequency-independent single-Lead and single-beat myocardial infarction detection","volume":"121","author":"H Martin","year":"2021","journal-title":"Artif Intell Med"},{"key":"pone.0354834.ref016","doi-asserted-by":"crossref","first-page":"201","DOI":"10.1016\/j.jelectrocard.2023.09.010","article-title":"Multi-center atrial fibrillation electrocardiogram (ECG) classification using Fourier space convolutional neural networks (FD-CNN) and transfer learning","volume":"81","author":"L Vasconcelos","year":"2023","journal-title":"J Electrocardiol"},{"issue":"3","key":"pone.0354834.ref017","doi-asserted-by":"crossref","first-page":"648","DOI":"10.1109\/TETCI.2023.3235374","article-title":"A multi-view multi-scale neural network for multi-label ECG classification","volume":"7","author":"S Yang","year":"2023","journal-title":"IEEE Trans Emerg Top Comput Intell"},{"key":"pone.0354834.ref018","first-page":"1120","article-title":"Improving automatic detection of ECG abnormality with less manual annotations using siamese network","volume":"2021","author":"F Yang","year":"2021","journal-title":"Annu Int Conf IEEE Eng Med Biol Soc"},{"issue":"4","key":"pone.0354834.ref019","doi-asserted-by":"crossref","first-page":"471","DOI":"10.3390\/e24040471","article-title":"CNN-FWS: A model for the diagnosis of normal and abnormal ECG with feature adaptive","volume":"24","author":"J Zhu","year":"2022","journal-title":"Entropy (Basel)"},{"issue":"9","key":"pone.0354834.ref020","doi-asserted-by":"crossref","first-page":"1121","DOI":"10.3390\/e23091121","article-title":"ECG signal classification using deep learning techniques based on the PTB-XL dataset","volume":"23","author":"S \u015amigiel","year":"2021","journal-title":"Entropy (Basel)"},{"issue":"24","key":"pone.0354834.ref021","doi-asserted-by":"crossref","first-page":"8174","DOI":"10.3390\/s21248174","article-title":"Deep learning techniques in the classification of ECG signals using r-peak detection based on the PTB-XL dataset","volume":"21","author":"S \u015amigiel","year":"2021","journal-title":"Sensors (Basel)"},{"issue":"3","key":"pone.0354834.ref022","doi-asserted-by":"crossref","first-page":"904","DOI":"10.3390\/s22030904","article-title":"Study of the few-shot learning for ECG classification based on the PTB-XL dataset","volume":"22","author":"K Pa\u0142czy\u0144ski","year":"2022","journal-title":"Sensors (Basel)"},{"issue":"1","key":"pone.0354834.ref023","doi-asserted-by":"crossref","first-page":"22040","DOI":"10.1038\/s41598-024-72382-3","article-title":"Deep residual 2D convolutional neural network for cardiovascular disease classification","volume":"14","author":"HA Elyamani","year":"2024","journal-title":"Sci Rep"},{"key":"pone.0354834.ref024","author":"K Sharma","year":"2022"},{"key":"pone.0354834.ref025","doi-asserted-by":"crossref","first-page":"102492","DOI":"10.1016\/j.bspc.2021.102492","article-title":"FFT-based deep feature learning method for EEG classification","volume":"66","author":"M Li","year":"2021","journal-title":"Biomedical Signal Processing and Control"},{"key":"pone.0354834.ref026","doi-asserted-by":"crossref","first-page":"106479","DOI":"10.1016\/j.engappai.2023.106479","article-title":"Electroencephalogram signal classification based on Fourier transform and Pattern Recognition Network for epilepsy diagnosis","volume":"123","author":"Q Gao","year":"2023","journal-title":"Engineering Applications of Artificial Intelligence"},{"key":"pone.0354834.ref027","doi-asserted-by":"crossref","first-page":"104102","DOI":"10.1016\/j.medengphy.2024.104102","article-title":"ECG arrhythmia detection in an inter-patient setting using Fourier decomposition and machine learning","volume":"124","author":"B Fatimah","year":"2024","journal-title":"Med Eng Phys"},{"issue":"1","key":"pone.0354834.ref028","doi-asserted-by":"crossref","first-page":"7","DOI":"10.3390\/biomedinformatics5010007","article-title":"Time\u2013frequency transformations for enhanced biomedical signal classification with convolutional neural networks","volume":"5","author":"G Lekkas","year":"2025","journal-title":"BioMedInformatics"},{"key":"pone.0354834.ref029","doi-asserted-by":"crossref","first-page":"1584299","DOI":"10.3389\/fphys.2025.1584299","article-title":"A multi-modal deep learning approach for stress detection using physiological signals: Integrating time and frequency domain features","volume":"16","author":"J-Z Xiang","year":"2025","journal-title":"Front Physiol"},{"issue":"1","key":"pone.0354834.ref030","doi-asserted-by":"crossref","first-page":"154","DOI":"10.1038\/s41597-020-0495-6","article-title":"PTB-XL, a large publicly available electrocardiography dataset","volume":"7","author":"P Wagner","year":"2020","journal-title":"Sci Data"},{"key":"pone.0354834.ref031","doi-asserted-by":"crossref","first-page":"104196","DOI":"10.1016\/j.medengphy.2024.104196","article-title":"Leadwise clustering multi-branch network for multi-label ECG classification","volume":"130","author":"F Zhou","year":"2024","journal-title":"Med Eng Phys"},{"issue":"1","key":"pone.0354834.ref032","doi-asserted-by":"crossref","first-page":"230","DOI":"10.1186\/s12916-019-1466-7","article-title":"Calibration: the Achilles heel of predictive analytics","volume":"17","author":"Topic Group \u2018Evaluating diagnostic t","year":"2019","journal-title":"BMC Med"},{"key":"pone.0354834.ref033","volume-title":"Bioelectrical signal processing in cardiac and neurological applications","author":"L S\u00f6rnmo","year":"2005"},{"key":"pone.0354834.ref034","doi-asserted-by":"crossref","unstructured":"Clifford GD, Liu C, Moody B, Lehman LH, Silva I, Li Q, et al. AF classification from a short single lead ECG recording: The PhysioNet\/computing in cardiology challenge 2017. In: 2017 computing in cardiology (CinC). IEEE. 2017.","DOI":"10.22489\/CinC.2017.065-469"},{"key":"pone.0354834.ref035","doi-asserted-by":"crossref","unstructured":"Mun\u00f5z A, Torres-Santamaria J, Iregui M, Romero E, Cruz-Roa A. Comparative Analysis of Deep Neural Network Architectures for Heart Disease Classification in Electrocardiography Signals. In: 2024 20th International Symposium on Medical Information Processing and Analysis (SIPAIM), 2024. 1\u20134. https:\/\/doi.org\/10.1109\/sipaim62974.2024.10783577","DOI":"10.1109\/SIPAIM62974.2024.10783577"},{"issue":"11","key":"pone.0354834.ref036","doi-asserted-by":"crossref","first-page":"992","DOI":"10.1016\/j.jacc.2008.12.015","article-title":"AHA\/ACCF\/HRS recommendations for the standardization and interpretation of the electrocardiogram: Part V: electrocardiogram changes associated with cardiac chamber hypertrophy","volume":"53","author":"EW Hancock","year":"2009","journal-title":"J Am Coll Cardiol"},{"issue":"11","key":"pone.0354834.ref037","doi-asserted-by":"crossref","first-page":"976","DOI":"10.1016\/j.jacc.2008.12.013","article-title":"AHA\/ACCF\/HRS recommendations for the standardization and interpretation of the electrocardiogram: part III: Intraventricular conduction disturbances","volume":"53","author":"B Surawicz","year":"2009","journal-title":"J Am Coll Cardiol"},{"key":"pone.0354834.ref038","doi-asserted-by":"crossref","first-page":"75","DOI":"10.1016\/j.hjc.2024.08.011","article-title":"Detection and classification of electrocardiography using hybrid deep learning models","volume":"81","author":"IJ Selvam","year":"2025","journal-title":"Hellenic J Cardiol"},{"issue":"16","key":"pone.0354834.ref039","doi-asserted-by":"crossref","first-page":"7758","DOI":"10.3390\/app11167758","article-title":"K-labelsets method for multi-label ECG signal classification based on SE-ResNet","volume":"11","author":"J Yoo","year":"2021","journal-title":"Appl Sci (Basel)"},{"issue":"10","key":"pone.0354834.ref040","doi-asserted-by":"crossref","first-page":"5877","DOI":"10.1109\/JBHI.2024.3416897","article-title":"Scaling representation learning from ubiquitous ECG with state-space models","volume":"28","author":"K Avramidis","year":"2024","journal-title":"IEEE J Biomed Health Inform"},{"key":"pone.0354834.ref041","doi-asserted-by":"crossref","first-page":"108447","DOI":"10.1016\/j.bspc.2025.108447","article-title":"Interpretable deep learning framework for multi-label ECG classification: Enhancing cardiac diagnostics through feature attention and explainability","volume":"112","author":"W Zeng","year":"2026","journal-title":"Biomedical Signal Processing and Control"},{"key":"pone.0354834.ref042","doi-asserted-by":"crossref","first-page":"108210","DOI":"10.1016\/j.compbiomed.2024.108210","article-title":"Enhanced multi-label cardiology diagnosis with channel-wise recurrent fusion","volume":"171","author":"W Wen","year":"2024","journal-title":"Comput Biol Med"},{"issue":"11","key":"pone.0354834.ref043","doi-asserted-by":"crossref","first-page":"18710","DOI":"10.1109\/JSEN.2024.3392017","article-title":"Spatiotemporal 3-D variations modeling with self-attention for multilabel ECG classification","volume":"24","author":"P Xia","year":"2024","journal-title":"IEEE Sens J"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0354834","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T17:38:40Z","timestamp":1785346720000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0354834"}},"subtitle":[],"editor":[{"given":"Choon-Hian","family":"Goh","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,7,29]]},"references-count":43,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7,29]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0354834","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,29]]}}}