FaceKit: a Toolkit for Interpretable Facial Phenotyping, Synthetic Image Generation and Privacy Analysis in Rare Diseases
Abstract
Many rare genetic diseases are associated with recognizable craniofacial features. However, traditional approaches for describing facial morphology rely largely on qualitative clinical observation and free-text descriptions, which are often subjective, non-standardized, and difficult to reproduce across observers and institutions. Although the Human Phenotype Ontology (HPO) provides controlled terms for describing facial features, these terms are typically categorical rather than quantitative and may vary depending on examiner experience and interpretation. Here, we present FaceKit, a computational framework for quantitative facial phenotyping from frontal facial photographs. FaceKit extracts standardized measurements of facial landmarks and derived 120 morphological features, then reports feature-level z-scores representing deviation from population reference distributions. The reference distributions are built from the FairFace dataset spanning diverse ancestral groups. We evaluated FaceKit on a curated subset of the GestaltMatcher Database covering 50 rare-disease cohorts. In addition to quantitative facial analysis, FaceKit includes synthetic facial image generation to support rare disease model development and data augmentation. We also performed privacy evaluation to assess whether synthetic images reveal identifiable information from real patient photographs and could compromise patient privacy. Across disease case studies, FaceKit-derived quantitative measurements captured known facial features associated with rare genetic disorders and provided objective support for clinical phenotyping. Together, these results establish FaceKit as a useful tool for quantitative phenotyping, and has the potential to improve rare disease diagnosis, support genotype-phenotype studies, and enable more reproducible clinical characterization across diverse patient populations.
Related
- Interpretable computational SHG microscopy enables quantitative tissue phenotyping of cardiac remodeling in nonischemic heart failure
- PhenoMe: An interactive AI framework for interpretable, microscopy-driven phenotyping
- Evaluating in-context learning with prompting regimes for transformer-based synthetic health tabular data generation
- Exploring the feasibility, acceptability and impact of genomic newborn screening for rare diseases in England: A study protocol for the Generation Study - Process and Impact Evaluation (Version 2)
- Benchmarking Recent Single-Cell Synthetic Data Generation Methods on the OneK1K Dataset
BibTeX
@misc{chen2026facekit,
title = {{FaceKit: a Toolkit for Interpretable Facial Phenotyping, Synthetic Image Generation and Privacy Analysis in Rare Diseases}},
author = {Hongzhuo Chen and Zhanliang Wang and Florent Pollet and Mian Umair Ahsan and Joshua Bie and Tzung-Chien Hsieh and Peter Krawitz and Cong Liu and Wendy K Chung and Chunhua Weng and Gamze Gürsoy and Kai Wang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.09288},
eprint = {2610.09288},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.09288},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - FaceKit: a Toolkit for Interpretable Facial Phenotyping, Synthetic Image Generation and Privacy Analysis in Rare Diseases
AU - Hongzhuo Chen
AU - Zhanliang Wang
AU - Florent Pollet
AU - Mian Umair Ahsan
AU - Joshua Bie
AU - Tzung-Chien Hsieh
AU - Peter Krawitz
AU - Cong Liu
AU - Wendy K Chung
AU - Chunhua Weng
AU - Gamze Gürsoy
AU - Kai Wang
PY - 2026
DA - 2026/10/07
PB - arXiv
DO - 10.48550/arxiv.2610.09288
UR - https://arxiv.org/abs/2610.09288
AB - Many rare genetic diseases are associated with recognizable craniofacial features. However, traditional approaches for describing facial morphology rely largely on qualitative clinical observation and free-text descriptions, which are often subjective, non-standardized, and difficult to reproduce across observers and institutions. Although the Human Phenotype Ontology (HPO) provides controlled terms for describing facial features, these terms are typically categorical rather than quantitative and may vary depending on examiner experience and interpretation. Here, we present FaceKit, a computational framework for quantitative facial phenotyping from frontal facial photographs. FaceKit extracts standardized measurements of facial landmarks and derived 120 morphological features, then reports feature-level z-scores representing deviation from population reference distributions. The reference distributions are built from the FairFace dataset spanning diverse ancestral groups. We evaluated FaceKit on a curated subset of the GestaltMatcher Database covering 50 rare-disease cohorts. In addition to quantitative facial analysis, FaceKit includes synthetic facial image generation to support rare disease model development and data augmentation. We also performed privacy evaluation to assess whether synthetic images reveal identifiable information from real patient photographs and could compromise patient privacy. Across disease case studies, FaceKit-derived quantitative measurements captured known facial features associated with rare genetic disorders and provided objective support for clinical phenotyping. Together, these results establish FaceKit as a useful tool for quantitative phenotyping, and has the potential to improve rare disease diagnosis, support genotype-phenotype studies, and enable more reproducible clinical characterization across diverse patient populations.
ER -