Improving Deep Facial Phenotyping for Ultra-rare Disorder Verification Using Model Ensembles

TL;DR

This paper enhances deep facial phenotyping for ultra-rare disorder verification by replacing outdated models with state-of-the-art face recognition techniques, introducing test-time augmentation and model ensembles to improve accuracy on unseen disorders.

Contribution

It introduces a novel ensemble approach combining general and disorder-specific face verification models, improving ultra-rare disorder verification accuracy.

Findings

Achieved state-of-the-art performance on unseen disorders

Demonstrated the effectiveness of model ensembles and test-time augmentation

Showed that replacing DCNN with iResNet and ArcFace improves representations

Abstract

Rare genetic disorders affect more than 6% of the global population. Reaching a diagnosis is challenging because rare disorders are very diverse. Many disorders have recognizable facial features that are hints for clinicians to diagnose patients. Previous work, such as GestaltMatcher, utilized representation vectors produced by a DCNN similar to AlexNet to match patients in high-dimensional feature space to support "unseen" ultra-rare disorders. However, the architecture and dataset used for transfer learning in GestaltMatcher have become outdated. Moreover, a way to train the model for generating better representation vectors for unseen ultra-rare disorders has not yet been studied. Because of the overall scarcity of patients with ultra-rare disorders, it is infeasible to directly train a model on them. Therefore, we first analyzed the influence of replacing GestaltMatcher DCNN with a state-of-the-art face recognition approach, iResNet with ArcFace. Additionally, we experimented with different face recognition datasets for transfer learning. Furthermore, we proposed test-time augmentation, and model ensembles that mix general face verification models and models specific for verifying disorders to improve the disorder verification accuracy of unseen ultra-rare disorders. Our proposed ensemble model achieves state-of-the-art performance on both seen and unseen disorders.