Compressed Sensing and Classification of Cardiac Beats using Patient Specific Dictionaries

In this paper, we investigated the benefits of compressed acquisition for monitoring applications of patients with various heart diseases. The possibility of heartbeat acquisition followed by classification into one of two classes, namely, normal beats or pathological has been approached using patient-specific dictionaries. Moreover, several types of projection matrices (matrices with random i.i.d. elements sampled from the Gaussian or Bernoulli distributions, and matrices optimized for the particular dictionary used in reconstruction by means of appropriate algorithms) have been compared. The dictionaries used in the reconstruction phase were built with and without centred R waves.