The primary objective of the current prospective study was to examine developmental patterns of voxel-by-voxel gray and white matter volumes (GMV, WMV, respectively) that would predict psychosis in adolescents with 22q11.2 deletion syndrome (22q11.2DS), the most common known genetic risk factor for schizophrenia. We performed a longitudinal voxel-based morphometry analysis using structural T1 MRI scans from 19 individuals with 22q11.2DS and 18 typically developing individuals. In 22q11.2DS, univariate analysis showed that greater reduction in left dorsal prefrontal cortical (dPFC) GMV over time predicted greater psychotic symptoms at Time2. This dPFC region also showed significantly reduced volumes in 22q11.2DS compared to typically developing individuals at Time1 and 2, greater reduction over time in 22q11.2DS COMT(Met) compared to COMT(Val), and greater reduction in those with greater decline in verbal IQ over time. Leave-one-out Multivariate pattern analysis results (MVPA) on the other hand, showed that patterns of GM and WM morphometric changes over time in regions including but not limited to the dPFC predicted risk for psychotic symptoms (94.7-100% accuracy) significantly better than using univariate analysis (63.1%). Additional predictive brain regions included medial PFC and dorsal cingulum. This longitudinal prospective study shows novel evidence of morphometric spatial patterns predicting the development of psychotic symptoms in 22q11.2DS, and further elucidates the abnormal maturational processes in 22q11.2DS. The use of neuroimaging using MVPA may hold promise to predict outcome in a variety of neuropsychiatric disorders.

This paper presents a learning based method for automatic extraction of the major cortical sulci from MRI volumes or extracted surfaces. Instead of using a few pre-defined rules such as the mean curvature properties, to detect the major sulci, the algorithm learns a discriminative model by selecting and combining features from a large pool of candidates. We used the Probabilistic Boosting Tree algorithm to learn the model, which implicitly discovers and combines rules based on manually annotated sulci traced by neuroanatomists. The algorithm almost has no parameters to tune and is fast because of the adoption of integral volume and 3D Haar filters. For a given approximately registered MRI volume, the algorithm computes the probability of how likely it is that each voxel lies on a major sulcus curve. Dynamic programming is then applied to extract the curve based on the probability map and a shape prior. Because the algorithm can be applied to MRI volumes directly, there is no need to perform preprocessing such as tissue segmentation or mapping to a canonical space. The learning aspect makes the approach flexible and it also works on extracted cortical surfaces.