Quispe-Enriquez, OmarLerma, José Luis2026-01-232026-01-232025-12-239788413963112https://riunet.upv.es/handle/10251/231914[EN] Cranial 3D models are essential for the analysis and early detection of deformities in infants. While they have traditionally been obtained using manual methods or 3D scanners, mobile photogrammetry offers an accessible and precise alternative. This study examines the use of PhotoMeDAS (Photogrammetric Medical Deformation Assessment Solutions), a mobile-based solution for generating 3D models with the accuracy required to assess and plan corrective interventions. Additionally, the potential of Artificial Intelligence (AI) for classifying cranial deformities is explored, although its implementation faces the challenge of limited data availability. To address this limitation, the generation of semi-synthetic data from 3D point clouds obtained through mobile photogrammetry is proposed, using a total of 60 cases (actual data). These data are classified according to different cranial pathologies, introducing controlled variations along the X, Y, and Z axes to simulate morphological changes. As a result, 2700 cases were generated, enriching the available datasets and facilitating the evaluation and application of classification algorithms such as decision trees, Random Forest, and multilayer perceptron neural networks. Preliminary results highlight the potential of PhotoMeDAS as a practical and cost-effective tool, provided that AI is efficiently integrated into future versions. However, further validation is required to confirm its applicability in actual clinical environments. 6Reconocimiento - No comercial - Compartir igual (by-nc-sa)Cranial 3D modelsPhotoMeDASPhotogrammetrySemi-Synthetic DataMachine learningArtificial IntelligenceComunicación en congreso10.4995/CiGeo2025.2025.19358Abierto