Expectation Maximization for Joint Deconvolution and Statistics Estimation

Biomedical ultrasound image quality is limited due to the blurring of tissue reflectivity introduced by the transducer Point Spread Function (PSF). Deconvolution techniques can be used to obtain the pure tissue response, otherwise called reflectivity function. Typically deconvolution methods are developed in the only purpose of image visual quality improvement. In this work we present an Expectation Maximization (EM) framework for US images deconvolution in which local statistical description of the tissue reflectivity is restored as well, so that features extracted from the deconvolved frame can theoretically be used for classification purposes.