3 February 2023 in IEEE Transactions on Instrumentation and Measurement
To detect and identify QRS complexes and R-peak is one of the crucial steps in the field of electrocardiogram (ECG) signals research, and their detection accuracy directly affects the performance of the subsequent ECG signal processing and analysis. However, the task involving the detection and identification of the ECG features becomes so complex for the traditional threshold-based detection methods. At present, researchers turn to deep learning approaches but little work has been done on the detection of spatiotemporal features of the QRS complexes. In this paper, for better QRS complexes and R-peak detection, we propose a novel method based on the improved U-net model, called ST-Res U-net. This uniquely designed component of the improved U-net model is made up of four levels of ST Block (the block extracting spatial-temporal features) and Res Path (the residual path). The entire framework contains three steps, data preprocessing dealing with denoising of the raw ECG signals, the key component ST-Res U-net dealing with the spatiotemporal feature extraction, and a threshold screening algorithm for locating the R-peaks. Our experiments are purposedly designed to test various combination structures of the ST blocks. The training and testing are from the MIT-BIH arrhythmia database (MITDB) and the CPSC2019 database. We adopt a commonly used set of evaluation criteria with the following experimental results: 99.76% and 90.01% for sensitivity, 99.87% and 93.5% for positive predictivity rate, 99.81% and 91.75% for F1 value, and 0.37% and 15.24% for detection error rate. The former numbers are for MITDB and the latter CPSC2019. Further, we test the proposed method on the PTB-XL database (without labeled R-peak positions) and the method can accurately locate QRS complexes and R-peaks The experimental results demonstrate that the proposed model and method is quite effective for the automatic classification and annotation of ECG signals and thus greatly improves the accuracy of diagnosis of arrhythmia diseases.
29 July 2022 in IEEE Transactions on Instrumentation and Measurement
In the field of arrhythmia classification, classification accuracy has always been a research hotspot. However, the noises of electrocardiogram (ECG) signals, the class imbalance of ECG data, and the complexity of spatiotemporal features of ECG data are all important factors affecting the accuracy of ECG arrhythmias classification. In this article, a novel DSCSSA ECG arrhythmias classification framework is proposed. First, discrete wavelet transform (DWT) is used to denoise and reconstruct ECG signals to improve the feature extraction ability of ECG signals. Then, the synthetic minority oversampling technique (SMOTE) oversampling method is used to synthesize a new minority sample ECG signal to reduce the impact of ECG data imbalance on classification. Finally, a convolutional neural network (CNN) and sequence-to-sequence (Seq2Seq) classification model with attention mechanism based on bidirectional long short-term memory (Bi-LSTM) as the codec is used for arrhythmias classification, and the model can give corresponding weight according to the importance of heartbeat features and can improve the ability to extract and filter the spatiotemporal features of heartbeats. In the classification of five heartbeat types, including normal beat (N), supraventricular ectopic beat (S), ventricular ectopic beat (V), fusion beat (F), and unknown beat (Q), the proposed method achieved the overall accuracy (OA) value and Macro- $F1$ score of 99.28% and 95.70%, respectively, in public the Massachusetts Institute of Technology-Boston’s Beth Israel Hospital (MIT-BIH) arrhythmia database. These methods are helpful to improve the effectiveness and clinical reference value of computer-aided ECG automatic classification diagnosis.