Nova Baeza, P.2026-06-302026-06-302026Current problems in surgery, N° 79 (2026) pp. 1-361535-6337https://hdl.handle.net/20.500.12254/7652The study develops computer vision (CV) models to rapidly predict acrylamide (AA) and 5-hydroxymethylfurfural (HMF) levels in crackers, comparing CV estimates with conventional analytical methods. Five baking temperatures (160–200°C) and times (15–35 min) were tested, showing CV accuracy with average errors of 3.10% for AA and 3.28% for HMF. Consumer preference tests (CATA) revealed that crackers baked at 180°C for 25 min, with the lowest AA and HMF levels, were most accepted.engAcceso abierto5-HydroxymethylfurfuralArticleAtribución-NoComercial-CompartirIgual 3.0 Chile (CC BY-NC-SA 3.0 CL)https://orcid.org/0000-0002-1781-062Xhttps://doi.org/10.1016/j.cpsurg.2026.1020271.2.1 - Ciencias de la Computación||2.11.1 - Tecnología de los Alimentos06 - Producción y tecnología industrialIA - Inteligencia Artificial||AA - Analítica Avanzada