Explainable Deep Learning System for Automatic Detection of Thyroid Eye Disease Using Facial Images - 11/08/25
, Kelvin Kam Lung Chong 3, 4, 6, 7, ⁎ Résumé |
Purpose |
To report an explainable deep learning (XDL) system to automatically detect thyroid eye disease (TED) using facial images.
Design |
Prospective study to develop and evaluate a deep-learning diagnostic algorithm.
Methods |
A dataset consisting of 302 and 289 facial images of newly diagnosed, treatment-naïve, TED patients and healthy subjects were compiled, annotated, and applied to train the XDL model. It consisted of a periocular landmarks localization network that identified the periocular landmarks on facial images, and the TED detection network (TDN) that uses a binary classification to detect TED using facial images. The generalizability of the XDL system was evaluated using a threefold cross-validation strategy and further validated using 100 facial images of TED patients from an independent thyroid eye clinic.
Results |
The area under the receiver operating characteristic curve was 99.7%, sensitivity 99.7%, and specificity 94.5% (95% confidence interval: 99.6%-99.9%). Heatmaps demonstrated upper and lower eyelids as key regions of interest. The validation cohort achieved area under the receiver operating characteristic curve of 98.9%, sensitivity 92%, and specificity 93%.
Conclusions |
This XDL system detected TED using facial images with excellent accuracy and explainability. It should be further evaluated in prospective Graves’ disease cohorts at nonspecialist setting for early detection and referral of progressive TED.
Le texte complet de cet article est disponible en PDF.Plan
| Xiaodan Sui and Kenneth Ka Hei Lai contributed equally as co-first authors while Hongsheng Li and Kelvin Kam Lung Chong contributed equally as co-corresponding authors. |
Vol 277
P. 323-334 - septembre 2025 Retour au numéro
Article précédent
- RPGR-Related Retinal Dystrophy in Latin America—A Collaborative Study
- Malena Daich Varela, Rene Moya, José D. Luna, Marcela Ciccioli, M. Eugenia Inga, Julieta Gras, Pedro J. Nuova, Luciana Capalbo, Alejandra Antacle, Laura Echandi, Angelica Moussali, Alejandro Sanders Villa, Marcela Pérez Araya, Tamara Muhlberger, Rocio A. Villafuerte-de la Cruz, Adda Villanueva, Tania Barragán Arévalo, Olivia Araujo Zin, Thiago Carvalho Barros de Oliveira, Fernanda Belga Ottoni Porto, Gustavo Barreto de Melo, Caio Henrique Marques Texeira, Mariana Vallim Salles, Renata Moreto, Jose Ronaldo Lima de Carvalho, Mariana Matioli da Palma, Rebeca Azevedo Souza Amaral, Cecília Francini Cabral de Vasconcellos, Gabriela Doná Rodrigues, Rosane Guazi Resende, Fabiana Louise Teixeira Motta, Luiz Alexandre Rassi Gabriel, Debora Moraes, Nancy Gelvez, Greizy López, Martha L. Tamayo, Juliana Maria Ferraz Sallum
- The Relationship of Hydroxychloroquine Retinopathy Progression to Stage at Cessation of Therapy
- David J. Browning, Donnie R. Koonce, Akshay Mentreddy, Omar Punjabi
Bienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.
Déjà abonné à cette revue ?