Phantom-based performance evaluation: Application to brain segmentation from magnetic resonance images

This paper presents a new technique for assessing the accuracy of segmentat ion algorithms, applied to the performance evaluation of brain editing and brain tissue segmentation algorithms for magnetic resonance images. We prop ose performance evaluation criteria derived from the use of the realistic d igital brain phantom Brainweb. This 'ground truth' allows us to build dista nce-based discrepancy features between the edited brain or the segmented br ain tissues (such as cerebro-spinal fluid, grey matter and white matter) an d the phantom model, taken as a reference. Furthermore, segmentation errors can be spatially determined, and ranged in terms of their distance to the reference. The brain editing method used is the combination of two segmenta tion techniques. The first is based on binary mathematical morphology and a region growing approach. It represents the initialization step, the result s of which are then refined with the second method, using an active contour model. The brain tissue segmentation used is based on a Markov random fiel d model. Segmentation results are shown on the phantom for each method, and on real magnetic resonance images for the editing step; performance is eva luated by the new distance-based technique and corroborates the effective r efinement of the segmentation using active contours. The criteria described here can supersede biased Visual inspection in order to compare, evaluate and validate any segmentation algorithm. Moreover, provided a 'ground truth ' is given, we are able to determine quantitatively to what extent a segmen tation algorithm is sensitive to internal parameters, noise, artefacts or d istortions. (C) 2000 Elsevier Science B.V. All rights reserved.