{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/atrial-fibrillation-detection-using-deep","title":"Atrial Fibrillation Detection Using Deep Features and Convolutional Networks","arxiv_id":"1903.11775","date":"2019-03-28","proceeding":null,"authors":["Sara Ross-Howe","H. R. Tizhoosh"],"abstract":"Atrial fibrillation is a cardiac arrhythmia that affects an estimated 33.5\nmillion people globally and is the potential cause of 1 in 3 strokes in people\nover the age of 60. Detection and diagnosis of atrial fibrillation (AFIB) is\ndone noninvasively in the clinical environment through the evaluation of\nelectrocardiograms (ECGs). Early research into automated methods for the\ndetection of AFIB in ECG signals focused on traditional bio-medical signal\nanalysis to extract important features for use in statistical classification\nmodels. Artificial intelligence models have more recently been used that employ\nconvolutional and/or recurrent network architectures. In this work, significant\ntime and frequency domain characteristics of the ECG signal are extracted by\napplying the short-time Fourier trans-form and then visually representing the\ninformation in a spectrogram. Two different classification approaches were\ninvestigated that utilized deep features in the spectrograms construct-ed from\nECG segments. The first approach used a pretrained DenseNet model to extract\nfeatures that were then classified using Support Vector Machines, and the\nsecond approach used the spectrograms as direct input into a convolutional\nnetwork. Both approaches were evaluated against the MIT-BIH AFIB dataset, where\nthe convolutional network approach achieved a classification accuracy of\n93.16%. While these results do not surpass established automated atrial\nfibrillation detection methods, they are promising and warrant further\ninvestigation given they did not require any noise prefiltering, hand-crafted\nfeatures, nor a reliance on beat detection.","url_abs":"http://arxiv.org/abs/1903.11775v1","url_pdf":"http://arxiv.org/pdf/1903.11775v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"arrhythmia-detection","task_name":"Arrhythmia Detection"},{"task_slug":"atrial-fibrillation-detection","task_name":"Atrial Fibrillation Detection"},{"task_slug":"electrocardiography-ecg","task_name":"Electrocardiography (ECG)"},{"task_slug":"classification","task_name":"General Classification"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-block","method_name":"Dense Block"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/atrial-fibrillation-detection-on-mit-bih-af","task":"Atrial Fibrillation Detection","dataset":"MIT-BIH AF","model":"Spectrogram + ConvNet","rank_in_archive_order":6,"of":6,"metrics":{"Accuracy":"93.16%"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}