Wavelet-Based Arrhythmia Detection in Medical Diagnostics Sensor Networks
This paper describes an application of wavelet transform for non-equidistant time series analysis in distributed sensor networks. Based on an idea of modern technologies of the Internet of Things and Big Data implementation in digital medicine there is outlined a problem of uneven time series analysis specific for medical diagnostics, specifically electrocardiogram (ECG) monitoring. As a solution there is proposed an original approach and algorithms of calculating the wavelet transform coefficients, using only those samples of the time series that are contained in the width of the wavelet. The advantage of this approach is that the result of the transformation is an even representation. The velocity of the algorithm is improved by taking into account the effective radius of the mother wavelet and calculating its width. The method and software tool for wavelet-based analysis of ECG signals are proposed for arrhythmia detection task. Experimental results show that proposed wavelet-based method of ECG analysis can detect signs of arrhythmia. Results of wireless channel speed test confirm that the chosen hardware meets the requirements of wireless protocol bandwidth. Proposed solutions are suitable for portable heart monitoring systems.
KeywordsMedical diagnostics The Internet of Things ECG analysis Wavelet transform
- 3.Ivaschenko, A., Minaev, A.: Multi-agent solution for adaptive data analysis in sensor networks at the intelligent hospital ward. In: Ślȩzak, D., et al. (eds.) International Conference on Active Media Technology. LNCS, vol. 8610, pp. 453–463. Springer, Switzerland (2014)Google Scholar
- 8.Peng, Z., Wang, G.: A novel ECG eigenvalue detection algorithm based on wavelet transform. Biomed. Res. Int. 2017, 5168346 (2017)Google Scholar
- 9.Gutiérrez-Gnecchia, J.A., Morfin-Magaña, R., Lorias-Espinoza, D., Tellez-Anguiano, A., Reyes-Archundia, E., Méndez-Patiñoa, A., Castañeda-Mirandac, R.: DSP-based arrhythmia classification using wavelet transform and probabilistic neural network. Biomed. Signal Process. Control 32, 44–56 (2017)CrossRefGoogle Scholar
- 11.Kuzmin, A., Safronov, M., Bodin, O., Petrovsky M., Sergeenkov A.: Device and software for mobile heart monitoring. In: Proceedings of the 19th Conference of Open Innovations Association FRUCT, pp. 121–127. FRUCT Oy, Helsinki (2016)Google Scholar