Desmoplastic Melanoma: Clinical Features, Prognostic Implications, and Survival Analysis from a 24-Year Cohort Study

Conclusions: XAI integration in dermatology enhances diagnostic accuracy and clinician trust.Certified devices with XAI functionalities improve workflows by making AI predictions more transparent.However, real-world validation and developer transparency remain essential.Clinician feedback is crucial to ensure AI tools are reliable and interpretable. 2 [2] Future advancements in multimodal XAI as in the ongoing Intelligent Total Body Scanner for Early Detection of Melanoma project (iToBoS) pave the way for better XAI integration in dermatology.Figure 2. Multimodal data integration for holistic risk scoring in skin cancer detection Anonymized metadata (demographics, history, genetics) is combined with TBP and dermoscopic images in the AI dataset.The algorithm generates holistic risk scores for both patients and lesions, supporting personalized diagnostics.