An Unsupervised Linear Discriminant Analysis Approach to Multispectral MRI Images Classification

Abstract

Magnetic Resonance Imaging (MRI) is a useful medical instrument in medical science because it provides unparallel capability of revealing soft tissue characterization as well as 3-D visualization and proposes the diagnosis without needing to intrude into the human body. MRI produces a sequence of multiple spectral images of tissues with a variety of contrasts, but the multi-spectral images cannot be conveniently used to be a pathology diagnosis correctly. In general, we need to transform the multispectral images to an enhanced image which is easier to be used for doctor's clinical diagnosis. One of the potential applications of MRI in clinical practice is the brain parenchyma classification. In this paper, we present a new approach called Unsupervised Linear Discriminant Analysis (ULDA) for the classification of multi-spectral MRI images. The ULDA consists of two processes, Target Generation Process (TGP) and Linear Discriminant Analysis (LDA) classification. As a result, ULDA can be used to search for a specific target in unknown scenes. Finally, the effectiveness of ULDA in target classification is evaluated by several MRI images experiments. In order to further evaluate its performance, ULDA is compared with Fuzzy C-mean for the medical image segmentation. Several experiment results show that the ULDA has the much better effective segmentation for multispectral MRI images and is robust to the noise disturbance in the image.