Low Power Intelligent Wearable Cardiac Sensor Using Discrete Wavelet Compression

Claiming 17.1 million lives a year, cardiovascular diseases are one of the major causes of death in the world especially in rural India. Conventional ECG monitoring instruments are quite bulky and if we are able to miniaturize them, then they can be used to collect data in scenarios that were not possible with the traditional systems. The system proposed here describes a low cost, low power wearable wireless sensor which can be used for issuing early warnings of cardiac disease to doctors, who are mostly mobile or who are away from the patient's location. The knowledge of early symptoms of a cardiac disease will provide an opportunity for the doctor to deliver real-time instructions to a relative/care taker for giving urgent medical care to the patient. Power optimisation has been achieved by using wavelet based compression technique (DWT) and also different modes of operation based on the risk factor of the patient. An algorithm based on discrete wavelet compression for power optimisation is developed and it is simulated in MATLAB. The results obtained shows that this technique achieves a power reduction of 22%. The proposed system can be used to deliver better healthcare to rural India, where availability of experienced doctors is almost scarce compared to urban India.