Statistical shape analysis (SSA) is an advanced computer vision and graphics technique for investigating anatomical structures and their geometric variations. We optimized an SSA-based pipeline to investigate brainstem deformations in a population of Joubert Syndrome (JS) patients. Using 3T T1-weighted MRI brain scans at 1 mm³  resolution of a group of 120 subjects ( 32 JS patients and  88 controls), we segmented brainstem masks to create triangular mesh representations. The pipeline defines a group-based template shape, on which we build a model assessing the subject-to-subject shape differences. Principal component analysis (PCA) identified key anatomical variations, constituting the features for binary classification between Joubert patients and controls. Classification achieved high balanced accuracy (90.3%), and we validated the results through a cross-validation procedure. Additionally, we developed a visualization framework to qualitatively assess the subject specific deformation to support the diagnostic task. Visual inspections in 3D, 2D MRI planes and their combinations, confirmed the ability of our method to detect main local brainstem deformations associated with Joubert syndrome, such as pons reduction, altered cerebellar peduncles interface, and mesencephalon dilatation. We also identified a susceptibility of the proposed tool to segmentation-related noise, probably related to a segmentation pipeline still not optimized for SSA-based analyses. Limitations, such as noisy data, exploration of methodological choices, and template selection, suggest directions for future works. In conclusion, our work offers insights into shape-based exploration and aims to provide a valuable analysis tool for the medical community both for clinical applications and research.