Voxel-based morphometry: current perspectives

Cristina Scarpazza, Maria De Simone
2016 Neuroscience and Neuroeconomics  
Voxel-based morphometry (VBM) is a widely used automated technique for the analysis of neuroanatomical images. This work, which reviews important VBM clinical findings of the last few years, is divided into two main sections. After briefly introducing VBM methodology, in the first part, VBM findings on neurological and psychiatric diseases have been discussed separately. The reported studies were divided into studies that examine the diagnostic value of VBM results and their usefulness for the
more » ... usefulness for the differential diagnosis between two disorders; studies investigating the potential of VBM for the diagnosis in an early stage of the illness, and studies that examine the utility of VBM for predicting the transition from prodromal phase to full-blown disease. In the second part, this review focuses on the most recent findings on the single-case approach. This analysis is useful for promoting the translational impact of VBM results in clinical practice, where clinicians need to make inferences at the level of the individual patient. Finally, within the single-case approach, a paragraph is dedicated to the potential forensic applications of VBM. Indeed, if used to support and integrate results obtained with classical forensic evaluations, VBM may provide objective data that could be used to reduce controversy in forensic psychiatric evaluations of mental insanity. In this second part, particular emphasis is given to the problem of results interpretation, which should be based mainly on the presence of anatomoclinical correlation. The review finishes with a provocative note reporting an interesting result of a single-case VBM analysis that highlights the risk to fall into the "reverse inference" reasoning. This example has been chosen because it effectively highlights the risks that are encountered when this technique is inappropriately used.
doi:10.2147/nan.s66439 fatcat:gkmuy3zdjjc7nh7bi4ffdjeukm