On the Influence of Confounding Factors in Multisite Brain Morphometry Studies of Developmental Pathologies: Application to Autism Spectrum Disorder

摘要: Pooling data acquired on different MR scanners is a commonly used practice to increase the statistical power of studies based MRI-derived measurements. Such are very appealing since they should make it possible detect more subtle effects related pathologies. However, influence confounds introduced by scanner-related variations remains unclear. When studying brain morphometry descriptors, crucial investigate whether scanner-induced errors can exceed effect disease itself. More specifically, in context developmental pathologies such as autism spectrum disorders (ASD), essential evaluate scanner age-related effects. In this paper, we studied dataset composed 159 anatomical images pooled from three scanners, including 75 ASD patients and 84 healthy controls. We quantitatively assessed age, pathology, factors cortical thickness Our results indicate that scan pooling sites would be less fruitful some regions than others. Although age consistent across interaction between important significant specific areas.