| Title |
High-efficiency ECG Data Analysis Scheme using Modified Residual Convolutional Neural Network Model |
| Authors |
고경남(Kyeong-Nam Ko) ; 강 문 식(Moon-Sik Kang) |
| DOI |
https://doi.org/10.5573/ieie.2021.58.10.42 |
| Keywords |
MIT-BIH arrhythmia 데이터베이스; ResNet; ResNeXt; Adabound; 주입기법 |
| Abstract |
In this paper, we propose an efficient ECG data analysis scheme to classify health status using a convolutional neural network model. The proposed system trains a model using the MIT-BIH arrhythmia database, and is designed to classify cardiovascular diseases into five categories. The model is constructed using the concepts used in ResNet and ResNeXt, which are widely used for image classification. In order to evaluate the performance of the implemented model, some data in the MIT-BIH arrhythmia database is selected and analyzed. As a result, the classification accuracy is 98.812%, and the F1-score obtained by considering the imbalance of data used for performance evaluation is 0.9326, and these results show the excellent performance. of the proposed system. |