Multimodal skin lesion classification using deep learning

TLDR

CNNs have been used for skin lesion classification, but prior work typically relies on a single macroscopic image and produces only binary outcomes. The study proposes a multimodal approach that fuses multiple imaging modalities and patient metadata to enhance automated skin lesion diagnosis. The method was tested on 2,917 cases with dermatoscopic, macroscopic images and metadata, evaluating both binary and five‑class classification tasks. The multimodal classifier achieved higher performance than the single‑image baseline, with AUC 0.866 versus 0.784 for binary melanoma detection and mAP 0.729 versus 0.598 for five‑class classification, and dermatoscopic images further improved accuracy over macroscopic images.

Abstract

Abstract While convolutional neural networks (CNNs) have successfully been applied for skin lesion classification, previous studies have generally considered only a single clinical/macroscopic image and output a binary decision. In this work, we have presented a method which combines multiple imaging modalities together with patient metadata to improve the performance of automated skin lesion diagnosis. We evaluated our method on a binary classification task for comparison with previous studies as well as a five class classification task representative of a real‐world clinical scenario. We showed that our multimodal classifier outperforms a baseline classifier that only uses a single macroscopic image in both binary melanoma detection ( AUC 0.866 vs 0.784) and in multiclass classification ( mAP 0.729 vs 0.598). In addition, we have quantitatively showed the automated diagnosis of skin lesions using dermatoscopic images obtains a higher performance when compared to using macroscopic images. We performed experiments on a new data set of 2917 cases where each case contains a dermatoscopic image, macroscopic image and patient metadata.