Latent Space Analysis for Melanoma Prevention
This work addresses the need for early, interpretable diagnostic tools in melanoma prevention, offering a method that bridges predictive performance with clinical applicability to enhance trust in AI-assisted diagnosis.
The paper tackles the problem of limited clinical insight from binary skin lesion classification models by introducing a Conditional Variational Autoencoder that learns a structured latent space for interpretable risk modeling, enabling continuous assessment of morphological differences and achieving strong performance in differentiating benign nevi from melanomas.
Melanoma represents a critical health risk due to its aggressive progression and high mortality, underscoring the need for early, interpretable diagnostic tools. While deep learning has advanced in skin lesion classification, most existing models provide only binary outputs, offering limited clinical insight. This work introduces a novel approach that extends beyond classification, enabling interpretable risk modelling through a Conditional Variational Autoencoder. The proposed method learns a structured latent space that captures semantic relationships among lesions, allowing for a nuanced, continuous assessment of morphological differences. An SVM is also trained on this representation effectively differentiating between benign nevi and melanomas, demonstrating strong and consistent performance. More importantly, the learned latent space supports visual and geometric interpretation of malignancy, with the spatial proximity of a lesion to known melanomas serving as a meaningful indicator of risk. This approach bridges predictive performance with clinical applicability, fostering early detection, highlighting ambiguous cases, and enhancing trust in AI-assisted diagnosis through transparent and interpretable decision-making.