Characteristics measurement for blood vessels diseases detection based on cone-beam CT images

This paper describes characteristics measurement for blood vessels diseases detection based on high-resolution 3D blood vessel images obtained by cone-beam CT. The important objective here is to show how the representation of blood vessel morphology can lead to feature measurement and identification of abnormal regions. In our method, the blood vessel structures are extracted by a graph description of blood vessel centerlines and surfaces representation using curvatures. The measurements of the anatomical information, such as the blood vessel's orientation, cross-sectional area, surface shapes, and abnormal regions volumes, are based on the structure description. In order to alert the doctor's attention to the location of the blood vessel's abnormality, we attempt to enhance the difference between malformed and normal shapes using the measured blood vessel characteristics. In this paper, we focus on a method based on surfaces representation and we present examples of cone-beam CT images of a patient's blood vessels.