AUTOMATIC CARDIAC MR IMAGE SEGMENTATION USING EDGE-DETECTION BY TISSUE CLASSIFICATION IN PIXEL NEIGHBORHOODS

A highly sensitive edge detector has been developed that uses tissue c lassification of pixels based on analysis of data in their local neigh borhoods, In conjunction with recursive region growing, it has been us ed successfully to define regions of interest (ROI) when applied speci fically to gradient echo MR images of the heart, The detector adapts t o nonuniformity by carrying out an independent analysis at each locati on. If two tissues are present in a neighborhood and the pixel at that location cannot be classified with the seed pixel, a region edge has been crossed and recursion is stopped, No geometric assumptions relati ng to object shape such as definition of a region center and radial se arch are required, The detector was applied to multi-slice, multi-phas e images of the heart from 26 subjects, A segmentation strategy specif ied slice processing order, graded ROIs, and used successfully detecte d ROls to guide subsequent detection, Segmentation of all images resul ted in a 90.3% median edge pixel detection efficiency.