Facilitate the early diagnosis of Alagille syndrome by facial image analysis
Patients
Photos
Countries
Why?
Alagille syndrome is a rare genetic disorder that can affect multiple organs, including the liver, heart, and kidneys. It is caused by mutations in the JAG1 or NOTCH2 genes. The syndrome is characterized by a distinctive facial appearance, which can be subtle and difficult to recognize. However, Early diagnosis is important for timely medical intervention and management of the disease. In general, the treatment outcomes are better when the disease is diagnosed early. Therefore, this project aims to improve the diagnosis of Alagille syndrome by using artificial intelligence to analyze facial images.
How?
Many patients with Alagille syndrome have distinctive facial dysmorphic features, and artificial intelligence (AI), such as GestaltMatcher (Hsieh et al. Nature Genetics, 2022), has proven the ability to recognize the rare disorder by facial image analysis. Over the last few years, we published GestaltMatcher to recognize 1,115 rare disorders and GestaltMatcher-Arc (Hustinx et al. WACV, 2023) to further improve the performance by 100%, considered state-of-the-art. Moreover, we built GestaltMatcher Database (GMDB) consisting of more than 10,000 facial images with 792 disorders that can be used for model training.
This project will be built upon our GestaltMatcher approach and database. We first collect facial photos of Alagille Syndrome patients to train the AI tool to help physicians diagnose accurately and efficiently.
How to participate in this project?
If you're interested in this project and want to contribute to it, please feel free to contact us or schedule up a meeting with us.
Schedule a Meeting, now!
Team

Dr. Tzung-Chien Hsieh
IGSB, University Hospital Bonn, Germany
PostDoc and Principal Investigator

Ms. Sophia Kaptain
IGSB, University Hospital Bonn, Germany
Data curator

Dr. Jing-Mei Li
IGSB, University Hospital Bonn, Germany
Data scientist and software developer

Ms. Merle ten Hagen
IGSB, University Hospital Bonn, Germany
Data curator

Ms. Hannah Weiland
IGSB, University Hospital Bonn, Germany
Data curator

Prof. Peter M. Krawitz
IGSB, University Hospital Bonn, Germany
Medical consultor

Prof. Shahida Moosa
Stellenbosch University, Cape Town, South Africa
Medical consultor
References
GestaltMatcher
Hsieh, TC., Bar-Haim, A., Moosa, S. et al. GestaltMatcher facilitates rare disease matching using facial phenotype descriptors. Nat Genet (2022).
https://doi.org/10.1038/s41588-021-01010-x (PDF)
GestaltMatcher Database
Lesmann, H. et al. GestaltMatcher Database - a FAIR database for medical imaging data of rare disorders. medRxiv (2023)
https://https://doi.org/10.1101/2023.06.06.23290887