Frequency prioritised queuing in real-time electrocardiograph systems

Real-time Electrocardiogram (ECG) telemonitoring is currently being envisaged to provide improved healthcare services for elderly and chronic care patients. A major challenge that arises in this situation is the need to reliably transmit ECG data over bandwidth constrained communications channels where congestion can lead to loss of ECG data. To overcome this problem, we propose a frequency prioritised transmission scheme which uses the sampling rate itself to separate out the ECG data into four frequency components, ranging from low to high frequency. We design a priority queuing system which gives highest priority to ECG packets containing low frequency components during transmission over a bandwidth constrained communications channel. The real-time performance of the proposed prioritisation algorithm is tested on a purposely built ECG machine. Our results show that the proposed scheme is robust and can withstand up to 65% of lost packets from a standard sampling rate of 320 Hz without any significant degradation in ECGs of healthy and arrhythmia sustained subjects.