Authors:Delia Mitrea¹;Sergiu Nedevschi¹;Monica Platon-Lupsor² andRadu Badea³

Affiliations:¹Technical University of Cluj-Napoca, Romania;²Department of Medical Imaging, I. Hatieganu University of Medicine and Pharmacy, Romania;³University of Medicine and Pharmacy Iuliu Hatieganu Cluj Napoca, Romania

Abstract:This chapter describes specific, texture-based methods for the detection,characterization and recognition of some severe affections and of theirevolution phases, using only information from ultrasound images. We performthe recognition of the considered affections in supervised manner, and we alsodiscover the disease evolution phases in unsupervised manner. In both cases,the imagistic textural model is defined, consisting of: the relevant features forthe characterization of the disease, respectively of its evolution phase; the specificvalues of the relevant textural features: arithmetic mean, standard deviation,probability distribution. Advanced texture analysis techniques, consistingof textural microstructure co-occurrence matrices based on Laws’ features, areinvolved in this process. At the end, the imagistic textural model is validatedthrough powerful, supervised classifiers, the resulting accuracy being around90%.