Exploring deep features and ECG attributes to detect cardiac rhythm classes

• Published in: Knowledge-based systems Vol. 232; p. 107473
• Main Authors: Murat, Fatma, Yildirim, Ozal, Talo, Muhammed, Demir, Yakup, Tan, Ru-San, Ciaccio, Edward J., Acharya, U. Rajendra
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 28.11.2021; Elsevier Science Ltd
• Subjects: Accuracy; Arrhythmia; Artificial neural networks; Cardiac arrhythmia; Cardiac rhythm; Classifiers; Deep learning; Diagnostic systems; ECG signals; Electrocardiography; Feature extraction; Feature maps; Inspection; Machine learning; Performance enhancement; Perturbation; Principal components analysis; Rhythm; Training; Deep learning; ECG signals; Feature extraction; Cardiac rhythm
• DOI: 10.1016/j.knosys.2021.107473
• ISSN: 0950-7051

Arrhythmia is a condition characterized by perturbation of the regular rhythm of the heart. The development of computerized self-diagnostic systems for the detection of these arrhythmias is very popular, thanks to the machine learning models included in these systems, which eliminate the need for visual inspection of long electrocardiogram (ECG) recordings. In order to design a reliable, generalizable and highly accurate model, large number of subjects and arrhythmia classes are included in the training and testing phases of the model. In this study, an ECG dataset containing more than 10,000 subject records was used to train and diagnose arrhythmia. A deep neural network (DNN) model was used on the data set during the extraction of the features of the ECG inputs. Feature maps obtained from hierarchically placed layers in DNN were fed to various shallow classifiers. Principal component analysis (PCA) technique was used to reduce the high dimensions of feature maps. In addition to the morphological features obtained with DNN, various ECG features obtained from lead-II for rhythmic information are fused to increase the performance. Using the ECG features, an accuracy of 90.30% has been achieved. Using only deep features, this accuracy was increased to 97.26%. However, the accuracy was increased to 98.00% by fusing both deep and ECG-based features. Another important research subject of the study is the examination of the features obtained from DNN network both on a layer basis and at each training step. The findings show that the more abstract features obtained from the last layers of the DNN network provide high performance in shallow classifiers, and weight updates of DNN network also increases the performance of these classifiers. Hence, the study presents important findings on the fusion of deep features and shallow classifiers to improve the performance of the proposed system. •Both clinical ECG features and features obtained from deep learning model layers are used for detection cardiac rhythms.•A big ECG dataset containing more than 10,000 subject records was used.•Lead-II ECG signals from individual subjects grouped into four rhythm classes were analyzed.•The random forest (RF) classifier yielded a patient-level arrhythmia classification accuracy of 98% ± 0.64 using the fused features.