Comparative Technique and Performance Results on Novel Learned Snakes in Two Dissimilar Medical Domains

by Samuel D. Fenster, and John R. Kender

IEEE Computer Vision and Pattern Recognition, v. II pp. 706-713. June 2000, Hilton Head.

We review our work on how to teach deformable models to maximize image segmentation correctness based on user-specified criteria. We then present new variants and applications of learned snakes, modeled by four different probability density functions (PDFs), at three scales, and in the two very different medical domains of abdominal CT slices and echocardiograms.

We review and extend our method for evaluating which criteria work best. Success depends on the relation of objective function (the PDF) output to shape correctness. This relationship, for all the above learned snake variants and domains, is evaluated on perturbed ground truth shapes in three ways: by the incidence of ``false positives'' (scoring better than ground truth) of randomized shapes; by the monotonicity of the objective function versus shape closeness to ground truth, as given by a correlation coefficient; and by the distance of this relationship to the nearest monotonically increasing function, a new performance measure which we introduce here.

We exhaustively demonstrate such evaluations on traditional snakes, and on snakes for which image intensity and perpendicular gradient are learned separately, and with their covariances, and with separate learning over equal-length "sectors". Optimal blur appears to depend on domain. Both sectoring and the use of covariance markedly improve results in abdominal CT images, where nearby image landmarks (i.e. organs) stabilize learning. Results on echocardiograms, however, are less striking, although the use of covariance does show improvements; on investigation this appears due to the non-Gaussian distribution of image features in this domain.

Entire article: PDF (289K), compressed PostScript (252K)