Abstract In this paper, we present a textural features analysis applied to a medical image segmentation problem where other methods fail, i.e. the localization of thrombotic tissue in the aorta. This problem is extremely relevant because many clinical applications are being developed for the computer assisted, image driven planning of vascular intervention, but standard segmentation techniques based on edges or gray level thresholding are not able to differentiate thrombus from surrounding tissues like vena, pancreas having similar HU average and noisy patterns. Our work consisted in a deep analysis of the texture segmentation approaches used for CT scans, and on experimental tests performed to find out textural features that better discriminate between thrombus and other tissues. Finding that Run Length codes are those performing better both in literature and experiments, we tried to understand the reason of their success suggesting a revision of this approach with feature selection and the use of specifically thresholded Run Lengths that improve the discriminative power of measures reducing the computational cost.
[1]
Robert M. Haralick,et al.
Textural Features for Image Classification
,
1973,
IEEE Trans. Syst. Man Cybern..
[2]
D. L. Kirch,et al.
Abdominal organ segmentation using texture transforms and a Hopfield neural network
,
1999,
IEEE Transactions on Medical Imaging.
[3]
Mary M. Galloway,et al.
Texture analysis using gray level run lengths
,
1974
.
[4]
Jacob D. Furst,et al.
A Texture Dictionary for Human Organs Tissues’ Classification
,
2004
.
[5]
Fritz Albregtsen,et al.
Low dimensional adaptive texture feature vectors from class distance and class difference matrices
,
2004,
IEEE Transactions on Medical Imaging.
[6]
A. H. Mir,et al.
Texture analysis of CT images
,
1995
.
[7]
M. Ito,et al.
Trabecular texture analysis of CT images in the relationship with spinal fracture.
,
1995,
Radiology.
[8]
Xiaoou Tang,et al.
Texture information in run-length matrices
,
1998,
IEEE Trans. Image Process..