Images Processing and Visualization of Brain Tumors

The chapter offers an experimental computational technique within the framework of which 2D segmentation is performed using MRI data and the construction of a 3D model of a brain tumor. The basis of the proposed technique is: pre-processing of MRI images for noise reduction; brightness and contrast enhancement; contours formation of objects of interest and visualizing the areas under study. The basic computational element in the technique for contours formation and color coding is Shearlet transform for magnetic resonance imaging (MRI) images. In terms of a 3D model generation of a neoplasm a geometric approach (the “central point” method) is proposed. In the experimental part a tumor is studied using MRI images for a specific patient within the framework of personalized medicine technology. At the same time, in the process of processing and visualizing MRI images with a brain tumor the main requests of medical specialists are taken into account. The work shows that based on the developed methodology 3D modeling of brain tumor MRI data from a set of 2D images (DICOM) provide realistic visualization.