Convolutional Neural Networks and Cluster Analysis for Heart Rate Data Analysis and Monitoring

DOI: https://doi.org/10.62517/jbdc.202301307

Author(s)

Jiaqiang Peng, Zhiyun Li, Rongrong Huang, Yan Liang*

Affiliation(s)

School of Science, Guangdong University of Petrochemical Technology, Maoming, Guangdong, China *Corresponding Author.

Abstract

Heart rate variation is a dynamic process that requires real-time monitoring and timely diagnostic categorization in clinical settings. Addressing the challenge of reducing manual workload while ensuring diagnostic accuracy through intelligent diagnostic technologies is a crucial concern in clinical practice. Frequency domain analysis of heart rate time series can reflect the operation of the autonomic nervous system, thereby enhancing the sensitivity and accuracy of heart rate variation monitoring. In this study, through the fusion of temporal and time-frequency domain features in electrocardiogram (ECG) data using Convolutional Neural Networks (CNN), and based on inter-sample clustering analysis and Short-Time Fourier Transform (STFT) evaluation of heart rate variability, a model for heart rate changes in a normal population is established. Following the analysis and processing of selected heart rate data, this model provides a foundational approach for predicting abnormal heart rate conditions and health management. The use and introduction of data analysis techniques in this paper serve as an important reference for medical and health management research. Continuous experimentation with related algorithms to minimize errors will enhance the accuracy and comprehensiveness of mathematical-based heart rate monitoring and analysis methods, thereby promoting their widespread application in relevant fields.

Keywords

Convolutional Neural Networks; Cluster Analysis; Short-time Fourier Transform; Heart Rate Monitoring

References

[1] Wang Jilong. Intelligent Approaches for Data Analysis in ECG Monitoring Systems in Chinese. Hunan University, 2021. DOI: 10.27135/d.cnki.ghudu.2021.004611. [2] Wang Wenxue. Advanced Research on Wearable Technology in Training and Performance Monitoring of Golfers in Chinese. Sporting Goods and Technology, 2023 (23): 157-159. [3] Jiang Zongkang, Lv Xiaogang, Zhang Jianxin, et al. Overview of Deep Learning Methods for MR Brain Tumor Image Segmentation in Chinese. Journal of Image and Graphics, 2020, 25 (2): 215-228. [4] Wang Yingying, Xue Chao, Yin Zhaofang, et al. Development and Application Progress of Intelligent Diagnosis in Electrocardiogram Diagnosis in Chinese. Journal of Cardiovascular Rehabilitation Medicine, 2019, 28 (04): 502-505. [5] Zeng Yuchen, He Zhaosheng, Hu Shulin, et al. ECG Arrhythmia Classification with Multi-Domain Feature Fusion using Convolutional Neural Networks and Attention Mechanism in Chinese. Information and Computer (Theoretical Edition), 2023, 35 (01): 75-79. [6] Chawla N V, Bowyer K W, Hall L O, et al. SMOTE: Synthetic Minority Over-sampling Technique. Journal of Artificial Intelligence Research, 2002, 16 (1): 321-357. [7] Yang Li. Design of Health Monitoring System Based on ARM in Chinese. Electronic Fabrication, 2023, 31 (22): 97-99+96. DOI: 10.16589/j.cnki.cn11-3571/tn.2023.22.014. [8] Guo Chen, Jian Tao, Xu Cong'an, et al. Radar Ship Target Recognition Based on Deep Pseudo Two-Dimensional Convolutional Neural Network in Chinese. Journal of Electronics & Information Technology, 2019, 41 (6): 1302-1309. [9] Moody G B, Mark R G. The impact of the MIT-BIH arrhythmia database. IEEE Engineering in Medicine and Biology Magazine, 2001, 20 (3): 45-50. [10] Li Jia, Zhang Yang. Research on High-Frequency Noise Removal Algorithm in ECG Surrounding Signals Based on Wavelet Analysis in Chinese. Science and Technology Innovation, 2019, (19): 33-34.