Files and links (1)
CC BY-SA V4.0, Open Access
Thesis
Real-Time Arrhythmia Detection using Convolutional Neural Network
Washington State University
Master of Science (MS), Washington State University
01/2021
DOI:
https://doi.org/10.7273/000001873
Handle:
https://hdl.handle.net/2376/120714
Abstract
Arrhythmia, defined as conditions related to abnormal rhythm of the heart, has become one of the leading cause of death in the United States. Currently, the most common diagnostic method of arrhythmia is through the analysis of an electrocardiogram (ECG or EKG) by a medical personnel. This method can be time-consuming as an entire ECG recording may be several minutes long. In this thesis, we present the study into real-time arrhythmic detection using neural networks. Most existing studies either look into arrhythmia classification but not in real-time, or propose a real-time method that does not have an in depth real-time analysis of the run time. We develop a simple convolutional neural network, which takes images of ECG segments as input, and classifies the arrhythmia conditions. One of the limitation of an image-based approach is that, for a time-series dataset, is not the most efficient method for classification. We carry out extensive experiments and evaluated the computational cost of each step of the classification workflow, and our result shows real-time arrhythmic detection using neural network is indeed possible. To further demonstrate the flexibility of this approach, we used the trained model with processed data collected by a customized wearable sensor from a lab setting, and the result shown that the model is highly accurate and efficient.
Metrics
4 File views/ downloads
25 Record Views
Details
- Title
- Real-Time Arrhythmia Detection using Convolutional Neural Network
- Creators
- Thong Vu
- Contributors
- Xinghui Zhao (Advisor)Scott Wallace (Committee Member)Ben McCamish (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Engineering and Computer Science (VANC), School of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University
- Number of pages
- 78
- Identifiers
- 99900606650701842
- Language
- English
- Resource Type
- Thesis