The World Health Organization estimates that every 30 s, a lower-limb amputation occurs due to diabetes mellitus, with studies indicating that in Mexico, nearly 50% of these amputations could be prevented through early detection and timely intervention. Classifying tissue within and around diabetic foot ulcers, as well as monitoring their size, is crucial for effective diagnosis and treatment. To address this, we propose a mobile virtual assistant that integrates artificial intelligence algorithms for the automatic segmentation of ulcers and affected tissue from images. The models were trained using the Foot Ulcer Segmentation (FUSeg) Challenge 2021 dataset from MICCAI, which includes 1,010 binary segmentation images, and the DFUTissue dataset, consisting of 110 images annotated with callus, fibrinous, and granulation tissues. The segmentation models, based on the Unet architecture, were validated using 68 images from Mexican adults aged 65-70 years with diabetic foot ulcers, collected during specialist-assisted care, achieving 96% accuracy in ulcer segmentation with a Dice coefficient close to 0.8. The developed mobile application integrates these models, providing additional metrics of interest and enabling users to track their progress over time.