Skin Lesion Phenotyping via Nested Multi-modal Contrastive Learning

TL;DR

This paper presents SLIMP, a novel nested contrastive learning method that integrates image and metadata to improve skin lesion classification, addressing challenges of variability and lack of context in medical imaging.

Contribution

SLIMP introduces a new multi-modal pre-training approach that combines images and metadata for enhanced skin lesion representation learning.

Findings

Improved classification performance over existing methods.

Effective integration of image and metadata modalities.

Enhanced representation quality for skin lesion analysis.

Abstract

We introduce SLIMP (Skin Lesion Image-Metadata Pre-training) for learning rich representations of skin lesions through a novel nested contrastive learning approach that captures complex relationships between images and metadata. Melanoma detection and skin lesion classification based solely on images, pose significant challenges due to large variations in imaging conditions (lighting, color, resolution, distance, etc.) and lack of clinical and phenotypical context. Clinicians typically follow a holistic approach for assessing the risk level of the patient and for deciding which lesions may be malignant and need to be excised, by considering the patient's medical history as well as the appearance of other lesions of the patient. Inspired by this, SLIMP combines the appearance and the metadata of individual skin lesions with patient-level metadata relating to their medical record and other clinically relevant information. By fully exploiting all available data modalities throughout the learning process, the proposed pre-training strategy improves performance compared to other pre-training strategies on downstream skin lesions classification tasks highlighting the learned representations quality.