摘要
The cardiovascular disease is the one of the major cause of death in the today's world. The Holter recorder is used to monitor the Electrocardiogram (ECG) signal to record the heart activity which is a popular method to detect the disease. Unfortunately, finding professionals to examine a big volume of ECG data takes up far too much medical time and money. The machine learning-based methods for recognising ECG characteristics have important role for extraction of features and ECG signal classification for Arrhythmia detection. The traditional processed used for the cardiac treatment have certain disadvantages, such as the need to understand the exact cardiac problem to recognize the disease and long learning curve to study the arrhythmia process manually. In this study twelve different heartbeat micro-classes are used with MIT-BIH Arrhythmia database for the classification. An efficient and robust virtually 12-layer deep traditionally one-dimensional CNN (convolutional neural network) is used with ELM (Extreme learning Machine) is used to propose the classification procedure. In the experiment different heart beat characteristics are identified, and the wavelet self-adaptive threshold denoising approach is applied by a combined algorithm using ELM and CNN. The results reveal that the model described in this paper performing better than the random forest, BP neural network, and other GA networks in terms of accuracy, sensitivity, resilience, and anti-noise capabilities. In this work ELM based classification on Convolutional Neural Network is developed for arrhythmia detection from the ECG signal. Its precise classification effectively conserves medical resources, which benefits clinical practise. In this study the classification result produces 98.82% accuracy which is quite satisfactory with compared with other similar types of algorithms used classification results.
学科与关键词
- ECG signal classification and arrhythmia detection using ELM-RNNKuila, Sumanta*; Dhanda, Namrata; Joardar, SubhankarMULTIMEDIA TOOLS AND APPLICATIONS, 2022, 81(18): 25233-25249.被引 18
- Automated arrhythmia classification based on a combination network of CNN and LSTMChen, Chen; Hua, Zhengchun; Zhang, Ruiqi; Liu, Guangyuan; Wen, Wanhui*Biomedical Signal Processing and Control, 2020, 57: 101819.被引 135
- A Hybrid Deep CNN Model for Abnormal Arrhythmia Detection Based on Cardiac ECG SignalUllah, Amin; Rehman, Sadaqat ur; Tu, Shanshan*; Mehmood, Raja Majid; Fawad; Ehatisham-ul-haq, MuhammadSENSORS, 2021, 21(3): 951.被引 76
