A Sharper Eye
2020, Vol. 38, No. 1
A Sharper Eye: How Advanced Imaging Technologies and Artificial Intelligence Are Transforming Melanoma Diagnosis
The Feature
In This Issue
Pushing Technology Forward
Allan C. Halpern, MD, has long advocated for dermatologists to use the latest technology to assist in the diagnosis of melanoma. In fact, he discussed the topic in his very first issue as editor of this publication. It was early 1996, in the Volume 14, No. 1 issue. It featured an article by Robert H. Johr, MD, then director of the Pigmented Lesion Clinic at the University of Miami School of Medicine, on the use of dermoscopy, “which he believes has been overlooked by most American physicians,” Dr. Halpern wrote.
The state of the art then to distinguish melanomas from other lesions had been created by the first editor of this publication: Alfred W. Kopf, MD, and colleagues at NYU: the “ABCD” rule (for asymmetry, border irregularity, color variegation and diameter greater than 6 mm). And when doctors did use a dermatoscope, diagnostic accuracy improved dramatically.
When Ashfaq A. Marghoob, MD, joined Dr. Halpern as an editor in 2001, his first issue (Volume 19, No. 1) was titled “Misdiagnosis of Melanoma,” and stated that while significant advances had been made in early detection of melanoma, clinical diagnostic accuracy was still less than optimal. The doctors said, “Techniques such as dermoscopy and the ‘ABCD’ rule of melanoma are improving diagnostic accuracy.” Then they looked forward, predicting, “We may eventually have ‘computer vision’ — computers with artificial intelligence that provide automated diagnosis of early melanoma.”
They were right. Now, 19 years later (they’re too humble to say it, so I am saying it for them), that “vision” is becoming a reality, thanks to the pioneering efforts of Dr. Halpern, Dr. Marghoob and the colleagues who have joined them. There is still room for improvement, of course, both in awareness and in adapting existing technology for widespread clinical use.
That is why we wanted to feature their work — and their knowledge — in this special issue. We recruited writer Kenneth Miller, an LA-based journalist and contributing editor at Discover magazine, to interview our editors on the emerging technologies in imaging and artificial intelligence and translate what these advances mean for dermatology in a way that will be useful for a broad range of medical professionals.
The pandemic has made the practice of medicine difficult across all specialties. It has also helped to push technology forward quickly, with enactment of special regulations to allow expansion of telemedicine, including how doctors may be reimbursed and what is covered by insurance. Many dermatologists who were once reluctant realized that since skin cancer is the cancer you can see, being able to share images and interact with patients via video technology has never been more crucial.
In a June 2020 article on the Harvard Health Blog, Kristina Liu, MD, wrote, “Even before the COVID-19 pandemic, studies found that teledermatology diagnoses can be accurate, result in high patient satisfaction and allow for better access to dermatologists in areas where their numbers are limited or wait times are high.”
During this ongoing health crisis, dermatologists are learning by doing — and discovering challenges such as the need for imaging standards. But many are also finding that teledermatology is an effective, if not perfect, way to help determine what can wait a little longer for biopsy or treatment — and what cannot.
The July 2020 JAAD featured a letter from Sara Perkins, MD, and colleagues in the Department of Dermatology at Yale School of Medicine about their successful transition from a completely office-based practice to a primarily teledermatolology model over just three weeks. While many dermatologists have not been as successful, a time of crisis does show that while moving technology forward rapidly can feel disruptive, it can lead to lifesaving and lasting change.
Julie Bain • Senior Director of Science & Education