{"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/a-comparison-of-1-d-and-2-d-deep","title":"A Comparison of 1-D and 2-D Deep Convolutional Neural Networks in ECG Classification","arxiv_id":"1810.07088","date":"2018-10-16","proceeding":null,"authors":["Yunan Wu","Feng Yang","Ying Liu","Xuefan Zha","Shaofeng Yuan"],"abstract":"Effective detection of arrhythmia is an important task in the remote\nmonitoring of electrocardiogram (ECG). The traditional ECG recognition depends\non the judgment of the clinicians' experience, but the results suffer from the\nprobability of human error due to the fatigue. To solve this problem, an ECG\nsignal classification method based on the images is presented to classify ECG\nsignals into normal and abnormal beats by using two-dimensional convolutional\nneural networks (2D-CNNs). First, we compare the accuracy and robustness\nbetween one-dimensional ECG signal input method and two-dimensional image input\nmethod in AlexNet network. Then, in order to alleviate the overfitting problem\nin two-dimensional network, we initialize AlexNet-like network with weights\ntrained on ImageNet, to fit the training ECG images and fine-tune the model,\nand to further improve the accuracy and robustness of ECG classification. The\nperformance evaluated on the MIT-BIH arrhythmia database demonstrates that the\nproposed method can achieve the accuracy of 98% and maintain high accuracy\nwithin SNR range from 20 dB to 35 dB. The experiment shows that the 2D-CNNs\ninitialized with AlexNet weights performs better than one-dimensional signal\nmethod without a large-scale dataset.","url_abs":"http://arxiv.org/abs/1810.07088v1","url_pdf":"http://arxiv.org/pdf/1810.07088v1.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":[{"paper_slug":"a-comparison-of-1-d-and-2-d-deep","repo_url":"https://github.com/axelmukwena/biometricECG","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"ecg-classification","task_name":"ECG Classification"},{"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":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"grouped-convolution","method_name":"Grouped Convolution"},{"method_slug":"local-response-normalization","method_name":"Local Response Normalization"},{"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":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}