METHOD FOR SEGMENTATION OF AN MRI IMAGE OF A TISSUE IN PRESENCE OF PARTIAL VOLUME EFFECTS AND COMPUTER PROGRAM IMPLEMENTING THE METHOD

A method and computer program for segmentation of an MRI image of tissue in presence of partial volume effects, include storing the MRI image in K-space representation as raw dataset, reconstructing N images, each represented by N sets of voxels and N sets of light intensity values, dividing each voxel into a fixed number of subvoxels equal in size, assigning the light intensity value of the voxel to its subvoxels, classifying each subvoxel according to light intensity value, identifying a totality of subvoxels classified with equal probability for a totality of tissue types, labeling the subvoxels as partial volume subvoxels, shifting N−1 images starting with the second image by one subvoxel relatively to the preceding image and gene-rating a new set of overlay subvoxels, determining a new set of probability values for the new set of overlay subvoxels and creating an overlay image of the new set of overlaid subvoxels.