Eye on the future
Artificial intelligence, new cancer treatments converge on dermatology.
F001 – Future of Dermatology Symposium
What does the future of dermatology hold? Fortunately, you don’t need a crystal ball to find out.
It seems inevitable that artificial intelligence (AI) is a big part of the discussion about the future. But the technology is moving more rapidly than you might think, according to James Weinstein, DO, MS, vice president of Microsoft Research Health Futures, and a presenter during Thursday’s session, F001 – Future of Dermatology Symposium.
“What we’re seeing with this era of AI is that it’s happening so fast that human beings aren’t able to keep up,” Dr. Weinstein said. “People can only adapt at a certain rate, and AI has already passed that rate. We have to realize that people need time to adapt to these technologies, so we have to work as a society to not leave people behind.”
Part of that work, he said, involves health care adapting to AI. One of the major concerns that exists with AI is that if it’s not properly handled, it could exacerbate existing disparities within the health care system.
“One of my biggest fears is the disparities could actually get worse, because we’ll have the haves and have nots in AI, and those who have it may be much more advantaged in some areas than those who don’t,” he believes. “So, as a society we have to be thoughtful about this ecosystem transformation and not just putting this new technology in without looking at how it affects the public good, what it is going to cost, who’s going to pay for it, and how we can be sure we don’t leave people behind. I’m pretty excited about it, but I’m also cautious at this point.”
Contrary to popular belief
Shannon Wongvibulsin, MD, PhD, a dermatology resident at the University of California Los Angeles (UCLA) and member of the AAD Augmented Intelligence Committee’s Standards Workgroup, said that despite some claims that AI will replace the job of the dermatologist, there are pitfalls when it comes to adopting AI in dermatology. She emphasized caution regarding data, potential bias, patient safety, and privacy as well as cost considerations.
The key, she said, is for dermatologists to get out in front and guide the use of AI within the specialty, rather than the other way around.
“AI tools need to be cautiously evaluated prior to use,” she said. “And dermatologist involvement is essential.”
And still other technologies
In addition to AI, the session showcased other innovative, emerging technologies from around the world. Edward Brennan, MD, FACS, chief medical officer of Medicus Pharma in Toronto, presented phase I study results from SkinJect, a Medicus subsidiary that evaluated the safety and tolerability of D-MNA in basal cell carcinoma patients.
The study involved 13 participants and used a 3+3 dose escalation design with four dose groups and a placebo control group. The study found there were no serious adverse events or dose-limiting toxicities, and the D-MNA was well-tolerated across all dose levels, with minimal local reactions.
In addition, the treatment showed positive responses in basal cell carcinoma lesions with six patients showing complete lesion response at each dose level. Dr. Brennan said these results hold tremendous promise for a better treatment option for basal cell carcinoma.
“Basal cell carcinoma is the most prevalent cancer worldwide," Dr. Brennan said. "Over five million new cases are reported in the United States annually. Current standard of care involves surgical procedures which, while effective, are often painful, expensive, and not aesthetic. There is a significant need for a non-invasive, cost-effective, and aesthetically acceptable treatment option like SkinJect.”
Still another abstract focused on site identification. Site identification is one of the leading causes of medical errors in dermatology. Julian Henke, a third-year medical student at Emory University School of Medicine in Atlanta, highlighted a future where artificial intelligence in dermatologic image analysis, diagnosis and triangulation could help eliminate this problem with the use of specially designed smart glasses.
“Seamlessly integrating AI, augmented reality and LIDAR technology into an affordable pair of lightweight smart glasses gives dermatology providers across specialties an unmet solution for accurate, efficient lesion follow-up,” Henke said. “Based on the infrastructure of current technology advancements, superposition of previous lesion-borders pinpointed on a LIDAR-scanned patient will refine dermatologic examination through standardization and objective confirmation.”
Directors for this popular session included Hensin Tsao, MD, PhD, FAAD, April W. Armstrong, MD, MPH, FAAD, Paul Nghiem, MD, PhD, FAAD, Bobby Y. Reddy, MD, FAAD, and Sandy Sharon Tsao, MD, FAAD. In addition to Drs. Weinstein and Wongvbulsin, a bevy of speakers included Aaron Fanous, Eran Gwillim, MD, Julian Henke, Jeffrey A. Klein, MD, FAAD, Justin W. Marson, MD, Tasneem Mohammad, MD, FAAD, Matthew Molenda, MD, FAAD, Yevgeniy Semenov, MD, MS, FAAD, and Madison Weisz.