Objective A method for dynamically collecting and processing ECG signals was designed to obtain classification
information of abnormal ECG signals. Methods Firstly, the ECG eigenvectors were acquired by real-time acquisition of
ECG signals combined with discrete wavelet transform, and then the ECG fuzzy information entropy was calculated.
Finally, the Euclidean distance was used to obtain the semantic distance of ECG signals, and the classification
information of abnormal signals was obtained. Results The device could effectively identify abnormal ECG signals
on an embedded platform based on the Internet of Things, and improved the diagnosis accuracy of heart diseases.
Conclusion The fuzzy diagnosis device of ECG signal could accurately classify the abnormal signal and output an
online signal classification matrix with a high confidence interval.
WANG Kai1, XU Jicheng2, ZHANG Yu1.Design of Portable Fuzzy Diagnosis Instrument for ECG Signal Based on Internet of Things[J] Chinese Journal of Medical Instrumentation, 2019,V43(5): 341-344
PANIGRAHY D, SAHU P K. Extraction of fetal ECG signal by an improved method using extended Kalman smoother framework from single channel abdominal ECG signal[J].Aust Phy Eng Sci Med.2017, 40(1):191-207
TANG M, YAO J, CHEN Z, et al. Implementation of wearable terminal used for dynamic ECG real-time monitoring based on internet of things[J]. Chin J Med Instrum, 2018, 42(3): 161.
JENKAL W, LATIF R, TOUMANARI A, et al. An efficient algorithm of ECG signal denoising using the adaptive dual threshold filter and the discrete wavelet transform[J].Biocybernetics Biomed Eng.2016, 36(3):499-508