Optimal Filter Length to Identify Uninterpretable Electrocardiograms

The ubiquity of wearable health monitors has drastically improved the quality of life for end-users whose wellbeing relies on continuous monitoring. The main challenge with telemedicine services lies in the quality of the data that the system receives from the user. Therefore, it is paramount that the telehealth processing system has a way to check and identify corrupted data so that it can be corrected or excluded from further processing. The Simple Differential Filter (SDF) aims to solve the challenge of artifact detection in ECG signals. In this paper, we research the optimal value for filter length. This parameter is significant not only because of its implications on performance and accuracy, but memory management in big data systems as well.