Improving MRI segmentation with probabilistic GHSOM and multiobjective optimization

In the last years, the improvements in Magnetic Resonance Imaging systems (MRI) provide new and additional ways to diagnose some brain disorders such as schizophrenia or the Alzheimer disease. One way to figure out these disorders from a MRI is through image segmentation. Image segmentation consist in partitioning an image into different regions. These regions determine different tissues present on the image. This results in a very interesting tool for neuroanatomical analyses. In this paper we present a segmentation method based on the Growing Hierarchical Self-Organizing Map and multiobjective-based feature selection to optimize the performance of the segmentation process. Since the features extracted from the image result crucial for the final performance of the segmentation process, optimized features are computed to maximize the performance of the segmentation process on each plane. The experiments performed on this paper use real brain scans from the Internet Brain Segmentation Repository (IBSR) and the Alzheimer Disease Neuroimaging Initiative (ADNI). Moreover, a comparison with other methods using the IBSR database shows that our method outperforms other algorithms.