Search

Energy functional minimization is a popular technique for medical image segmentation. The segmentation must be initialized, weights for competing terms of an energy functional must be tuned, and the functional minimized. There is a substantial amount of guesswork involved. We reduce this guesswork by analytically determining the optimal weights and minimizing a convex energy functional independent of the initialization. We demonstrate improved results over state of the art on a set of 470 clinical examples.

Energy functional minimization is an increasingly popular technique for image segmentation. However, it is far too commonly applied with hand-tuned parameters and initializations that have only been validated for a few images. Fixing these parameters over a set of images assumes the same parameters are ideal for each image. We highlight the effects of varying the parameters and initialization on segmentation accuracy and propose a framework for attaining improved results using image adaptive parameters and initializations. We provide an analytical definition of optimal weights for functional terms through an examination of segmentation in the context of image manifolds, where nearby images on the manifold require similar parameters and similar initializations. Our results validate that fixed parameters are insufficient in addressing the variability in real clinical data, that similar images require similar parameters, and demonstrate how these parameters correlate with the image manifold. We present significantly improved segmentations for synthetic images and a set of 470 clinical examples.

Spinal cord analysis is an important problem relating to the study of various neurological diseases. We present a novel approach to spinal cord segmentation in magnetic resonance images. Our method uses 3D “deformable organisms” (DefOrg) an artificial life framework for medical image analysis that complements classical deformable models (snakes and deformable meshes) with high-level, anatomically-driven control mechanisms. The DefOrg framework allows us to model the organism’s body as a growing generalized tubular spring-mass system with an adaptive and predominantly elliptical cross section, and to equip them with spinal cord specific sensory modules, behavioral routines and decision making strategies. The result is a new breed of robust DefOrgs, “spinal crawlers”, that crawl along spinal cords in 3D images, accurately segmenting boundaries, and providing sophisticated, clinically-relevant structural analysis. We validate our method through the segmentation of spinal cords in clinical data and provide comparisons to other segmentation techniques.