Patch-based sparse representation for bacterial detection

In this paper, we propose an unsupervised approach for bacterial de- tection in optical endomicroscopy images. This approach splits each image into a set of overlapping patches and assumes that observed intensities are linear combinations of the actual intensity values as- sociated with background image structures, corrupted by additive Gaussian noise and potentially by a sparse outlier term modelling anomalies (which are considered to be candidate bacteria). The ac- tual intensity term representing background structures is modelled as a linear combination of a few atoms drawn from a dictionary which is learned from bacteria-free data and then fixed while analyzing new images. The bacteria detection task is formulated as a minimization problem and an alternating direction method of multipliers (ADMM) is then used to estimate the unknown parameters. Simulations con- ducted using two ex vivo lung datasets show good detection and cor- relation performance between bacteria counts identified by a trained clinician and those of the proposed method.