Brain tissue segmentation based on DWI/DTI data

We present a method for tissue classification based on diffusion-weighted imaging (DWI)/diffusion tensor imaging (DTI) data. Our motivation is that independent tissue segmentation based on DWI/DTI images provides complementary information to the tissue segmentation result using structural MRI data alone. The basis idea is to classify the brain into two compartments by utilizing the tissue contrast exiting in a single channel, e.g., apparent diffusion coefficient (ADC) image can be used to separate CSF and non-CSF, and the fractional anisotropy (FA) image can be used to separate WM from non-WM tissues. Other channels such as eigen values of the tensor, relative anisotropy (RA), and volume ratio (VR) can also be used to separate tissues. We employ the STAPLE algorithm to combine these two-class maps to obtain a complete segmentation of CSF, GM, and WM