A Patient-Centric Dataset of Images and Metadata for Identifying Melanomas Using Clinical Context

TL;DR

This paper introduces a comprehensive patient-centric skin image dataset that includes multiple lesions per patient and associated metadata, aiming to improve melanoma diagnosis by reflecting real clinical practices.

Contribution

It provides a novel dataset linking multiple skin lesions to individual patients, enabling AI models to incorporate patient-level context for more accurate melanoma detection.

Findings

Dataset includes 2,056 patients and over 33,000 images.

Contains patient identifiers linking multiple lesions.

Includes histopathologically confirmed melanomas.

Abstract

Prior skin image datasets have not addressed patient-level information obtained from multiple skin lesions from the same patient. Though artificial intelligence classification algorithms have achieved expert-level performance in controlled studies examining single images, in practice dermatologists base their judgment holistically from multiple lesions on the same patient. The 2020 SIIM-ISIC Melanoma Classification challenge dataset described herein was constructed to address this discrepancy between prior challenges and clinical practice, providing for each image in the dataset an identifier allowing lesions from the same patient to be mapped to one another. This patient-level contextual information is frequently used by clinicians to diagnose melanoma and is especially useful in ruling out false positives in patients with many atypical nevi. The dataset represents 2,056 patients from three continents with an average of 16 lesions per patient, consisting of 33,126 dermoscopic images and 584 histopathologically confirmed melanomas compared with benign melanoma mimickers.