Faculty, Staff and Student Publications

Language

English

Publication Date

1-1-2025

Journal

Genetics in Medicine Open

DOI

10.1016/j.gimo.2025.103434

PMID

40575353

PMCID

PMC12197971

PubMedCentral® Posted Date

5-9-2025

PubMedCentral® Full Text Version

Post-print

Abstract

Purpose: Vascular Ehlers-Danlos syndrome (vEDS), which is caused by COL3A1 pathogenic variants, is a rare heritable aortic and arterial disorder associated with early mortality, mainly due to spontaneous vascular dissections and ruptures. Improved methods for diagnosing vEDS are needed for guideline-based management to be initiated for preventing deadly complications and differentiating vEDS from overlapping conditions, such as hypermobile EDS (hEDS).

Methods: We implemented an artificial intelligence (AI) facial analysis model based on the PhenoScore framework using a support vector machine trained on facial images of 30 individuals, aged 6 to 65 years, with vEDS from the Montalcino Aortic Consortium, control images from the Chicago Face Database, and publicly available images of individuals with hEDS. Cross-validation was used to train the support vector machine, and statistical measures to evaluate the model performance were calculated. Local Interpretable Model-agnostic Explanations was used to generate facial heatmaps highlighting the features driving the model's predictions.

Results: The AI classifier showed excellent performance with as few as 13 vEDS training images and distinguished vEDS from both controls and individuals with hEDS with high accuracy, achieving an area under the receiver operating characteristic curve ≥ 0.97. Local Interpretable Model-agnostic Explanations highlighted facial regions already established to characterize the facial features of vEDS patients (eg, prominent eyes).

Conclusion: Our results demonstrate the potential of AI-based facial analysis for diagnosing vEDS. This method democratizes the early diagnosis of vEDS by reducing dependence on genetic testing, enabling optimal management and improved outcomes, particularly in resource-limited areas.

Keywords

AI-based screening, Facial phenotype analysis, LIME, Machine learning, vEDS

Published Open-Access

yes

Download

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.