Repositori institucional URV
| URV's Author/s: | Abdellatif Fatahallah Ibrahim Mahmoud, Hatem / Abdelnasser Mohamed Mahmoud, Mohamed / Puig Valls, Domènec Savi |
| Author, as appears in the article.: | Rashwan, Hatem A; Marques-Pamies, Montserrat; Ruiz, Sabina; Gil, Joan; Asensio-Wandosell, Diego; Martinez-Momblan, Maria-Antonia; Vazquez, Federico; Salinas, Isabel; Ciriza, Raquel; Jorda, Mireia; Chanson, Philippe; Valassi, Elena; Abdelnasser, Mohamed; Puig, Domenec; Puig-Domingo, Manel |
| Author's mail: | domenec.puig@urv.cat hatem.abdellatif@urv.cat mohamed.abdelnasser@urv.cat |
| Author identifier: | 0000-0002-0562-4205 0000-0001-5421-1637 0000-0002-1074-2441 |
| Journal publication year: | 2025 |
| Publication Type: | Journal Publications |
| Paper original source: | Pituitary. 28 (3): 50- |
| Abstract: | PurposeTo describe the development of the AcroFace system, an AI-based system for early detection of acromegaly, based on facial photographs analysis.MethodsTwo types of features were explored: (1) the visual/texture of a set of 2D facial images, and (2) geometric information obtained from a reconstructed 3D model from a single image. We optimized acromegaly detection by integrating SVM for geometric features and CNNs for visual features, each chosen for their strength in processing distinct data types effectively. This combination enhances overall accuracy by leveraging SVM's capability to manage structured, quantitative data and CNNs' proficiency in interpreting complex image textures, thus providing a comprehensive analysis of both geometric alignment and textural anomalies. ResNet-50, VGG-16, MobileNet, Inception V3, DensNet121 and Xception models were trained with an expert endocrinologist-based score as a ground truth.ResultsResNet-50 model as a feature extractor and Support Vector Regression (SVR) with a linear kernel showed the best performance (accuracy delta 1 of 75% and delta 3 of 89%), followed by the VGG-16 as a feature extractor and SVR with a linear kernel. Geometric features yield less accurate results than visual ones. The validation cohort showed the following performance: precision 0.90, accuracy 0.93, F1-Score 0.92, sensitivity 0.93 and specificity 0.93.ConclusionAcroFace system shows a good performance to discriminate acromegaly and non-acromegaly facial traits that may serve for the detection of acromegaly at an early stage as a screening procedure at a population level. |
| Article's DOI: | 10.1007/s11102-025-01515-2 |
| Link to the original source: | https://link.springer.com/article/10.1007/s11102-025-01515-2 |
| Paper version: | info:eu-repo/semantics/publishedVersion |
| licence for use: | https://creativecommons.org/licenses/by/3.0/es/ |
| Department: | Enginyeria Informàtica i Matemàtiques |
| Licence document URL: | https://repositori.urv.cat/ca/proteccio-de-dades/ |
| Thematic Areas: | Ciências biológicas ii Educação física Endocrinology Endocrinology & metabolism Endocrinology, diabetes and metabolism Enfermagem Farmacia Interdisciplinar Medicina i Medicina ii Medicina iii Saúde coletiva |
| Keywords: | Acromegaly Acromegaly detectio Acromegaly detection Artificial intelligence Facial analysis Facial changes |
| Entity: | Universitat Rovira i Virgili |
| Record's date: | 2025-05-12 |
| Search you record at: |  |
| File | Description | Format | |
|---|---|---|---|
| DocumentPrincipal | DocumentPrincipal | application/pdf |
© 2011 Universitat Rovira i Virgili