Statistical shape analysis in neuroimaging : methods, challenges, validation : applications to the study of brain asymmetries in schizophrenia

Abstract

The study of brain shape and its patterns of variations can provide insights into the understanding of normal and pathological brain development and brain degenerative processes. This thesis focuses on the in vivo analysis of human brain shape as extracted from three-dimensional magnetic resonance images. Major automatic methods for the analysis of brain shape are discussed particularly focusing on the computation of shape metrics, the subsequent inference procedures, and their applications to the study of brain asymmetries in schizophrenia. Methodological challenges as well as possible biological factors that complicate the analysis of brain shape, and its validation, are also discussed. The contributions of this research work are as it follows.

First, a novel automatic method for the statistical shape analysis of local interhemispheric asymmetries is presented and applied to the study of cerebral structural asymmetries in schizophrenia. The method extracts and analyzes smooth surface representations approximating the gross shape of the outlines of cerebral hemispheres.

Second, a novel and fully automatic image processing framework for the validation of measures of brain asymmetry is proposed. The framework is based on the synthesis of realistic three-dimensional magnetic resonance images with a known asymmetry pattern. It employs a parametric model emulating the normal interhemispheric bending of the human brain while retaining other subject-specific features of brain anatomy. The framework is applied for the quantitative validation of measures of asymmetry in brain tissues' composition as computed by voxel-based morphometry. Particularly, the framework is used to investigate the dependence of voxel-based measures of brain asymmetry on the spatial normalization scheme, template space, and amount of spatial smoothing applied. The developed automatic framework is made available as open-source software.

Third, a novel Simplified Reeb Graph based descriptor of the human striatum is proposed. The effectiveness of such a descriptor is demonstrated for the purposes of automatic registration, decomposition, and comparison of striatal shapes in schizophrenia patients and matched normal controls.

In conclusion, this thesis proposes novel methods for shape representation and analysis within three-dimensional magnetic resonance brain images, an original way for validating these methods, and applies the methods for the study of brain asymmetries in schizophrenia. The impact of this research lies in its potential implications for the development of biomarkers aiming to a better understanding of the brain in normal and pathological conditions, early diagnosis of a number of brain diseases, and development of novel therapeutic strategies for improving the quality of life of affected individuals. In addition, the distribution of simulated data and automatic tools for validation of morphometric measures of brain asymmetry is expected to have a great impact in enabling systematic validation of novel and existing methods for the analysis of brain asymmetries, quantitatively comparing them, and possibly clarifying contradicting findings in the neuroimaging literature of brain lateralizations.