Best Practices for Clinical Skin Image Acquisition in Translational Artificial Intelligence Research

Article Dermatology

Evaluation of Artificial Intelligence-Assisted Diagnosis of Skin Neoplasms: A Single-Center, Paralleled, Unmasked, Randomized Controlled Trial

Seung Seog Han et al.

Summary: This study validated the role of artificial intelligence (AI) in assisting nonexpert physicians in dermatology diagnosis in real-world settings, indicating the potential of AI algorithms in improving diagnostic accuracy.

JOURNAL OF INVESTIGATIVE DERMATOLOGY (2022)

Add to Collection

Review Dermatology

Checklist for Evaluation of Image-Based Artificial Intelligence Reports in Dermatology CLEAR Derm Consensus Guidelines From the International Skin Imaging Collaboration Artificial Intelligence Working Group

Roxana Daneshjou et al.

Summary: The use of AI in dermatology is accelerating, and it is important to establish comprehensive criteria for development and evaluation to ensure fairness, reliability, and safety. In this consensus statement, experts provide guidelines for developers and reviewers to achieve clinically effective AI.

JAMA DERMATOLOGY (2022)

Add to Collection

Letter Dermatology

Bias in, bias out: Underreporting and underrepresentation of diverse skin types in machine learning research for skin cancer detection- A scoping review

Lisa N. Guo et al.

JOURNAL OF THE AMERICAN ACADEMY OF DERMATOLOGY (2022)

Add to Collection

Editorial Material Dermatology

Clinical photography in skin of colour: tips and best practices

J. C. Lester et al.

BRITISH JOURNAL OF DERMATOLOGY (2021)

Add to Collection

Editorial Material Dermatology

Equity in skin typing: why it is time to replace the Fitzpatrick scale

U. K. Okoji et al.

BRITISH JOURNAL OF DERMATOLOGY (2021)

Add to Collection

Article Oncology

Association between different scale bars in dermoscopic images and diagnostic performance of a market-approved deep learning convolutional neural network for melanoma recognition

Julia K. Winkler et al.

Summary: This study investigated the impact of scale bars on the diagnostic accuracy of a CNN analyzing dermoscopic images, finding that superimposed scale bars may lead to increased false-positive diagnoses. The research recommends avoiding the use of scale bars in images intended for CNN analysis to prevent impairment of diagnostic accuracy.

EUROPEAN JOURNAL OF CANCER (2021)

Add to Collection

Review Dermatology

Lack of Transparency and Potential Bias in Artificial Intelligence Data Sets and Algorithms A Scoping Review

Roxana Daneshjou et al.

Summary: This scoping review identified issues in data sets used for developing and testing clinical AI algorithms for skin disease, including lack of characterization and transparency, nonstandard disease labels, and inability to fully assess patient diversity.

JAMA DERMATOLOGY (2021)

Add to Collection

Editorial Material Dermatology