| Description: | Analyze ultrasound images of carotid arteries in vivo to extract the texture of the vessel wall in an attempt to detect the cases of primary hyperaldosteronism from other cases of hypertension and normal controls. It was found previously that there are statistical differences of the intima-media thickness and the Haralick texture parameters between these three classes for expert-selected regions of interest. The aim of this work is to (i) automatically evaluate the intima-media thickness, (ii) to automatically find the regions of interest, and (iii) to create and evaluate a classifier using a bigger dataset and more modern texture descriptors. |
|---|
