Disorders of the aortic valve represent a common cardiovascular disease and an important public-health problem. Pathological valves are currently determined from 2D images through elaborate qualitative evaluations and complex measurements, potentially inaccurate and tedious to acquire. This paper presents a novel diagnosis method, which identifies diseased valves based on 3D geometrical models constructed from volumetric data. Robust segmentation is applied to estimate the patient specific model from CT volumes. A learned discriminative distance function determines the corresponding pathology class based on the shape information. Experiments on 63 patients affected by various diseases demonstrated an accuracy of 94% correctly classified valves.     «

Disorders of the aortic valve represent a common cardiovascular disease and an important public-health problem. Pathological valves are currently determined from 2D images through elaborate qualitative evaluations and complex measurements, potentially inaccurate and tedious to acquire. This paper presents a novel diagnosis method, which identifies diseased valves based on 3D geometrical models constructed from volumetric data. Robust segmentation is applied to estimate the patient specific model...     »