doi:10.1364/BOE.559663engCasado-Moreno, JuanMasia, BelénLu, NanjiCui, LeleConsejo, AlejandraDeep learning-based keratoconus detection from Scheimpflug imagesART-2025-144923This study evaluates the effectiveness of deep learning techniques applied to raw Scheimpflug corneal images for keratoconus detection, with a particular focus on forme fruste (FF) keratoconus, which refers to preclinical cases. Using an original dataset of 22,750 images from 910 eyes, a deep learning model based on transfer learning with a pre-trained VGG16 architecture was trained, incorporating specific preprocessing steps and data augmentation strategies. The proposed approach achieved an overall accuracy of 90.70%, with a sensitivity of 80.57%, and a specificity of 80.56% for FF keratoconus classification, and an AUC of 0.89. For clinical keratoconus, the model demonstrated a sensitivity of 93.28%, a specificity of 99.40%, and an AUC of 1.00. These findings highlight the potential of leveraging raw Scheimpflug images in deep learning-based keratoconus detection, particularly for identifying early-stage structural changes that may not be apparent in conventional topographic assessments.2025http://zaguan.unizar.es/record/16231010.1364/BOE.559663http://zaguan.unizar.es/record/162310oai:zaguan.unizar.es:162310info:eu-repo/grantAgreement/EC/HORIZON EUROPE/101162733/EU/Visual Impairment Screening using Images from Ophthalmology and Novel pathways for Structural Analysis and Fast Evaluation/VISIONSAFEBiomedical Optics Express 16, 8 (2025), 3047-3060byhttps://creativecommons.org/licenses/by/4.0/deed.esinfo:eu-repo/semantics/openAccess