This study investigates whether structural brain asymmetries can be leveraged to classify hemispheric orientation and predict handedness using voxel-based morphometry (VBM) data from the Amsterdam Open MRI Collection (AOMIC). While hemispheric specialization is well established, the structural basis of handedness remains debated due to small effect sizes and methodological variability. Recent work in neuroimaging has emphasized predictive modelling, such as decoding approaches, as a complement to traditional explanatory analyses. In this context, we use support vector classifiers (SVCs) on pre-processed T1-weighted images to test four research questions: (1) Can an SVC reliably distinguish original from flipped brains, reflecting structural lateralization? (2) Are left-handed individuals classified as structurally flipped brains, relative to a right-handed template? (3) Can an SVC reliably distinguish between left and right handers? The classifier identifies flipped brains with over 98% accuracy, confirming consistent hemispheric asymmetries. However, left-handers are not classified as flipped brains, and classification performance on handedness is poor, with no consistent accuracy across full-brain or regional models. These results suggest that while structural lateralization is robust and detectable with VBM, it does not directly account for individual differences in handedness.