Quantification of brain tissue volumes using MR/MR fusion

We propose in this article to compute White Matter (WM), Gray Matter (GM) and Cerebrospinal Fluid (CF) volumes from MR images of a human brain using possibilistic logic and a fusion scheme. The fusion process is divided into three steps: fuzzy tissue maps are first computed on all images using a possibilistic clustering algorithm. Fusion is then achieved for all tissues with a context based fusion operator. The final segmentation allows for the computation of brain tissue volumes. Applications on a brain model show a great agreement between the results and the true segmentation, thus affording a relevant tissue classification on clinical data