A New Shape Diffusion Descriptor for Brain Classification

摘要: In this paper, we exploit spectral shape analysis techniques to detect brain morphological abnormalities. We propose a new descriptor able encode morphometric properties of image or region using diffusion geometry based on the local Heat Kernel. Using approach, it is possible design versatile signature, employed in case classify between normal subjects and patients affected by schizophrenia. Several strategies are assessed verify robustness proposed under different deformation variations. A dataset consisting MRI scans from 30 control utilized test which achieves promising classification accuracies, up 83.33%. This constitutes drastic improvement comparison with other description techniques.