Deep learning for dermatologists: Part I. Fundamental concepts - 25/05/21

Doi : 10.1016/j.jaad.2020.05.056

Dennis H. Murphree, PhD ^a,^b,^⁎ , Pranav Puri, BA ^b,^c, Huma Shamim, MBBS ^b,^d, Spencer A. Bezalel, MD ^b,^d, Lisa A. Drage, MD ^b,^d, Michael Wang, MD ^e, Mark R. Pittelkow, MD ^b,^f, Rickey E. Carter, PhD ^g, Mark D.P. Davis, MD ^b,^d, Alina G. Bridges, DO ^b,^d,^h, Aaron R. Mangold, MD ^b,^f, James A. Yiannias, MD ^f, Megha M. Tollefson, MD ^b,^d, Julia S. Lehman, MD ^b,^d,^h, Alexander Meves, MD ^b,^d, Clark C. Otley, MD ^b,^d, Olayemi Sokumbi, MD ^b,^i,^j, Matthew R. Hall, MD ^b,ⁱ, Nneka Comfere, MD ^b,^d,^h
^a Department of Health Sciences Research, Division of Digital Health Sciences, Mayo Clinic, Rochester, Minnesota
^b Mayo Clinic Office of Artificial Intelligence in Dermatology
^c Mayo Clinic Alix School of Medicine, Scottsdale, Arizona
^d Department of Dermatology, Mayo Clinic, Rochester, Minnesota
^e Department of Dermatology, University of California San Francisco, San Francisco, California
^f Department of Dermatology, Mayo Clinic, Scottsdale, Arizona
^g Department of Health Sciences Research, Division of Biomedical Statistics and Informatics, Mayo Clinic, Jacksonville, Florida
^h Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota
ⁱ Department of Dermatology, Mayo Clinic, Jacksonville, Florida
^j Department of Laboratory Medicine and Pathology, Mayo Clinic, Jacksonville, Florida

^∗Correspondence to: Dennis H. Murphree, PhD, 200 First St SW, Rochester, MN 55905.200 First St SWRochesterMN55905

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Tuesday 25 May 2021
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder

Abstract

Artificial intelligence is generating substantial interest in the field of medicine. One form of artificial intelligence, deep learning, has led to rapid advances in automated image analysis. In 2017, an algorithm demonstrated the ability to diagnose certain skin cancers from clinical photographs with the accuracy of an expert dermatologist. Subsequently, deep learning has been applied to a range of dermatology applications. Although experts will never be replaced by artificial intelligence, it will certainly affect the specialty of dermatology. In this first article of a 2-part series, the basic concepts of deep learning will be reviewed with the goal of laying the groundwork for effective communication between clinicians and technical colleagues. In part 2 of the series, the clinical applications of deep learning in dermatology will be reviewed and limitations and opportunities will be considered.

Le texte complet de cet article est disponible en PDF.

Key words : artificial intelligence, deep learning, dermatology, machine learning

Abbreviations used : AI, BCC, DL, DNN, GPU, ML