A patient-centric dataset of images and metadata for identifying melanomas using clinical context

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 (20.8% with at least one melanoma, 79.2% with zero melanomas) from three continents with an average of 16 lesions per patient, consisting of 33,126 dermoscopic images and 584 (1.8%) histopathologically confirmed melanomas compared with benign melanoma mimickers.

Machine Learning and Deep Learning Methods for Skin Lesion Classification and Diagnosis: A Systematic Review

“…These images were acquired from over 2000 patients. Images in the dataset were decomposed into nine classes in addition to an unknown image class [72]. Table 1 summarizes the total number of images and the total number of images in each class for all of these datasets.…”

Proof-of-principle neural network models for classification, attribution, creation, style-mixing, and morphing of image data for genetic conditions

“…This estimation was not done in some other large-scale studies involving skin images. 19 The fact that our classifier worked relatively well may demonstrate possible use cases. For example, this type of approach could help primary care doctors determine which patients should be prioritized for evaluation by subspecialists like geneticists.…”

“…17,18 For the auxiliary dataset, we downloaded the publicly available SIIM-ISIC Melanoma Classification Challenge Dataset from 2018 to 2020. 11,19 This dataset contains 58,459 images of 9 skin cancer diseases: actinic keratosis, basal cell carcinoma, benign keratosis, dermatofibroma, melanoma, melanocytic nevus, squamous cell carcinoma, vascular lesion, and other unknown skin cancer cases.…”

Co-Design of a Trustworthy AI System in Healthcare: Deep Learning Based Skin Lesion Classifier

“…The ML model currently used for skin lesion classification has been trained on a limited number of publicly available skin images (Mendonça et al, 2013;Codella et al, 2018;Kawahara et al, 2019;Rotemberg et al, 2021). As most of the data was acquired from a small number of distinct research facilities, the dataset suffers from variations in image quality and the occasional presence of artifacts and it cannot be guaranteed that the data is representative of patients' backgrounds (e.g., age, sex, skin tone distribution).…”