A novel segmentation approach for implementation of MRAC in head PET/MRI employing Short-TE MRI and 2-point Dixon method in a fuzzy C-means framework

Results for both 3 T and 1.5 T systems show that bone segmentation applied in several slices yields average accuracy, sensitivity and specificity higher than 90%. Results indicate that FCM is an appropriate technique for tissue classification in the sinusoidal area where there is air-bone interface. Furthermore, using Dixon method, fat and brain tissues were successfully separated.