Robust techniques for designing remote real-time arrhythmias classification system

This work describes a robust methods used in the development of "remote real-time arrhythmias classification system for multi-patients" (RACSM) to assist the medical diagnosis of cardiac arrhythmias. A coder based on a function of linear predictors and Huffman coding was chosen and found to give a compression ratio of at least 2:1 and as high as 3:1 on real-world electrocardiogram (ECG) signals. Four techniques have been used to achieve a reduction in Hardware energy consumption of 43.30%. For setting up the communication link for multi-patient; a wireless sensor protocol (WSP) was proposed and implemented using time division duplex (TDD) scheme. A robust method to accurately classify cardiac arrhythmias through a combination of wavelets and fuzzy logic has been introduced. The method performed well when evaluated with the MIT BIH Arrhythmia Database. At last, the algorithms and methods mentioned above are applied in the (two-patient) ECG remote diagnosis system to classify and diagnose 13 kinds of arrhythmia.