Skin Lesion Analysis Toward Melanoma Detection: A Challenge at the 2017 International Symposium on Biomedical Imaging (ISBI), Hosted by the International Skin Imaging Collaboration (ISIC)

TL;DR

This paper presents a comprehensive benchmark challenge for automated melanoma detection using dermoscopic images, involving multiple tasks and a large international participant base to advance skin cancer diagnosis algorithms.

Contribution

It introduces a large-scale, standardized challenge dataset and evaluation framework for lesion segmentation, feature detection, and disease classification in melanoma detection.

Findings

Largest dermoscopic image analysis challenge to date

Engaged 593 participants with 46 final submissions

Provided a benchmark dataset for future research

Abstract

This article describes the design, implementation, and results of the latest installment of the dermoscopic image analysis benchmark challenge. The goal is to support research and development of algorithms for automated diagnosis of melanoma, the most lethal skin cancer. The challenge was divided into 3 tasks: lesion segmentation, feature detection, and disease classification. Participation involved 593 registrations, 81 pre-submissions, 46 finalized submissions (including a 4-page manuscript), and approximately 50 attendees, making this the largest standardized and comparative study in this field to date. While the official challenge duration and ranking of participants has concluded, the dataset snapshots remain available for further research and development.