Abstract

AbstractThe area of telemedicine, including wireless communication, has an increasingly significant impact on health care. In recent years, an increase in the importance of compression methods has been noticed in many areas of medicine. Especially in the field of storing, processing and remotely transmitting large amounts of data. Different compression algorithms, based on different methods have been described in the literature so far. Few of them are currently used in monitoring systems and telemedicine.This paper discusses the use of the Matching Pursuit (MP) compression method for electrocardiographic signal transmission using LoRa (Long Range) technology. LoRa technology is one of the communication standards of the Internet of Things. An electrocardiogram (ECG) is a diagnostic tool that measures and records the heart’s electrical activity in detail. The ECG is an important tool used to diagnose heart abnormalities.The paper describes research of the effectiveness of the Matching Pursuit algorithm in compressing the ECG signal, using the Dictionary Learning. The signals from the PTB Diagnostic ECG Database were used for the tests. The analyzed signal fragments were divided into 1000 ms parts, which were then resampled and normalized. The learning data was used to create a dictionary of atoms using the Dictionary Learning method. Using the Orthogonal Matching Pursuit algorithm for the fragments of test data, indices of non-zero coefficients were obtained. In the next step, using the developed extreme compression algorithm, a byte table that could be transmitted over the network was obtained. In the research, the byte array was decompressed and the signal was reconstructed. Consequently, the transmitted to the original signal similarity was measured for different parameters of the algorithm.Thanks to the development of technology, remote patient monitoring, consulting and medical care can be more flexible and convenient.KeywordsSignal compressionMatching pursuitECG

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call