Dataset Information

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.

SUBMITTER: Rotemberg V

PROVIDER: S-EPMC7843971 | biostudies-literature | 2021 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

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

Rotemberg Veronica V Kurtansky Nicholas N Betz-Stablein Brigid B Caffery Liam L Chousakos Emmanouil E Codella Noel N Combalia Marc M Dusza Stephen S Guitera Pascale P Gutman David D Halpern Allan A Helba Brian B Kittler Harald H Kose Kivanc K Langer Steve S Lioprys Konstantinos K Malvehy Josep J Musthaq Shenara S Nanda Jabpani J Reiter Ofer O Shih George G Stratigos Alexander A Tschandl Philipp P Weber Jochen J Soyer H Peter HP

Scientific data 20210128 1

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 be ...[more]

PMID: 33510154

Similar Datasets

Project description:Undergraduate subject pools are prevalent across disciplines in the United States. The Health Education Research Experience (HERE) Program was the first known course-based subject pool entirely managed and conducted online for online students enrolled in an introductory health education/health promotion course. The program was conducted within five semesters from Spring 2012 through Summer 2013. The HERE Program encompassed 13 studies embedded in two sections of an undergraduate online course at the University of Florida. The studies were all related to course topics and current research topics in health education/promotion (as identified through the Healthy People 2020 Framework). The topics ranged from the relatively less sensitive health aspects of college life (i.e., technology use) to studies assessing more sensitive health topics (i.e., intimate partner violence and sexual assault). In alignment with a best practice in survey design, the HERE Program's survey instruments included one metadata item embedded in each survey to identify which devices students used to complete the surveys. Understanding which devices students used for survey completion has ramifications for survey designers and survey researchers. In contrast to the relative uniformity of pen and paper surveys and control of the survey completion environment, online surveys may not look identical across personal devices and may be completed in increasingly varied environments. All studies, study procedures and protocols, and metadata collection procedures were approved by the university's Institutional Review Board. The data presented here were extracted from each survey's data files and aggregated. The aggregated metadata are available through Mendeley Data in a.csv file for widespread access. Descriptive statistics are presented in tables. The data provided in this article will benefit researchers interested in survey methodology, questionnaire design, modes of survey collection, and survey metadata. The data are hosted in the following Mendeley Data repository: https://data.mendeley.com/datasets/ht9jmd3cdt/2.

Dataset Information

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

Publications

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets