Abstract
Congenital syndromes with subtle changes in maxillofacial morphology can pose significant diagnostic challenges, wherein artificial intelligence holds great promise in aiding diagnosis through shape analysis. The authors applied the recently proposed Swap Disentangled Variational Autoencoder (SD-VAE) in diagnosis of Beckwith-Wiedemann syndrome (BWS). The SD-VAE model was trained on a data set primarily comprised of surface 3D head scans [stereophotogrammetry (3D SPG)], gathered using a 3dMD Head System (3dMD LLC). It was also trained on CT scans when available. A total of 72 syndromic scans were used belonging to 56 different BWS patients. Scans of head shapes were pre-processed and annotated with 68 anatomic landmarks. This aided in achieving uniformity between the scans and a template mesh, making a better comparison possible. The SD-VAE model outputs were then visualized in a 2-dimensional space and classified as 'BWS-patient' or 'control'. For each anatomic facial region, the performance of the classification model was evaluated. This allowed us to understand the classification accuracy for each anatomic region as well as calculate the sensitivity and specificity for each region. The model demonstrated perfect diagnostic accuracy for BWS on the test set, with the most characteristic regions being the chin, cheeks, zygoma, eyes, jaw, and supraorbital region. This paper demonstrates how SD-VAE can be applied to 3D head meshes, to quantify the characteristic features of BWS. The authors distinguished BWS-specific features from those of the general population with high diagnostic accuracy. This makes SD-VAE a promising tool for aiding the referral and diagnosis of BWS in the future.