Resumen
This research revolutionizes dermatological diagnostics by integrating artificial intelligence (AI) and hyperspectral imaging (HSI) to identify skin cancer lesions, particularly Mycosis fungoides (MF). It differentiates MF from conditions such as psoriasis and atopic dermatitis using a dataset of 1659 skin images. This study employs a novel AI algorithm alongside advanced techniques for precise lesion segmentation and classification, moving diagnosis from color to spectral analysis. This non-invasive and efficient method marks a significant advancement in the early and accurate detection of skin malignancies. The model?s high performance is validated by its sensitivity, specificity, and accuracy, making it a vital tool in dermatology for identifying skin cancers and inflammatory conditions.