Independent component analysis in automated segmentation of brain tumors

Independent component analysis (ICA) is a powerful method for removing artifacts and separating independent sources from the multispectral magnetic resonance images (MRI) of the brain. Segmentation of tumor from MR images following feature extraction with ICA has been shown to be superior to conventional segmentation algorithms. However, the performance of most of the algorithms still falls far below expectations and thus cannot be utilized in clinical applications. In this paper, we review the main approaches to automated segmentation of brain tumors and the concepts involved in the application of ICA technique to MR images. The main features of the segmentation algorithms coupled with ICA are analyzed pointing out their strengths and weaknesses. A qualitative and quantitative comparison of the results of the approaches is also presented. Finally, possible future approaches to tumor segmentation are discussed.