• General Dermatology
  • Eczema
  • Alopecia
  • Aesthetics
  • Vitiligo
  • COVID-19
  • Actinic Keratosis
  • Precision Medicine and Biologics
  • Rare Disease
  • Wound Care
  • Rosacea
  • Psoriasis
  • Psoriatic Arthritis
  • Atopic Dermatitis
  • Melasma
  • NP and PA
  • Anti-Aging
  • Skin Cancer
  • Hidradenitis Suppurativa
  • Drug Watch
  • Pigmentary Disorders
  • Acne
  • Pediatric Dermatology
  • Practice Management

AI Transforming Medicine: Insights from GI to Dermatology

News
Article

AI's transformative impact on medicine, explored by Ryan Stidham, MD, reveals potential applications in dermatology, emphasizing caution, ethical considerations, and the need for collaboration across specialties.

Artificial intelligence (AI) is revolutionizing the field of medicine, and Ryan Stidham, MD, gastroenterologist and associate professor of medicine at the University of Michigan in Ann Arbor, Michigan shed light on its applications, particularly in inflammatory bowel disease (IBD). In his Masterclasses in Dermatology session “AI in IBD: What is the Gastroenterology Perspective,” he emphasized the challenges faced by healthcare practitioners in measuring diseases, with current metrics having limitations in capturing all features and presenting subjectivity. “As patient populations grow and resources become strained, the need for innovative solutions becomes crucial,” Stidham said.1

Joseph Merola, MD, MMSc, dermatologist and rheumatologist at UT Southwestern Medical Center in Dallas, Texas explained in an interview with Dermatology Times what clinicians should consider based on Stidham’s AI experience. “I think that the potential for AI and dermatology is huge. I think one of the things we learned from the IBD group and the IBD lecture with AI is a little bit the double-edged sword of AI as it comes to dermatology,” Merola said.

Stidham introduced core AI technologies driving advancements in medicine, focusing on automated information extraction through computer vision and natural language processing. Computer vision, utilizing technologies like convolutional neural networks, allows machines to "see," while natural language processing enables machines to read and comprehend vast amounts of text. Additionally, generative pre-training transformers (GPT), large language models like GPT-3, are mentioned as machines capable of generating human-like text.

The integration of AI in dermatology was discussed, with a focus on image analysis. Stidham highlighted the widespread use of AI in categorizing lesions, replicating expert judgment, and providing routine care. From skin cancer detection to disease quantitation in conditions like psoriasis, AI is proving instrumental in enhancing diagnostic accuracy and reducing subjectivity.

“So on the positive side, we're certainly all dealing with burnout, issues of wellness in medicine globally. And certainly in dermatology, it's nice to see the potential of AI in supporting many of our functions, including note writing documentation, potentially helping us to appropriately code based on that documentation and on visits, potentially supporting documentation for prior authorization and other functions of the office, which would be very much welcomed, Merola explained. “I think [dermatologists] could see the potential for supporting our decision making around, you know, disease severity, disease activity, a typical lesion monitoring among a broad range of positives.”However, Stidham stressed the need for caution and ethical considerations. He emphasizes the importance of addressing biases in AI models, citing a case where a healthcare model exhibited racial bias, impacting patient outcomes. This highlights the necessity of understanding and accounting for population variations in disease manifestation.

Moving beyond image analysis, Stidham delved into large language models like GPT, showcasing their potential in automating documentation tasks. Ambient conversation documentation, where conversations between practitioners and patients are recorded and transcribed, is presented as a practical application. This transcript can then be used to generate notes, letters, and procedure reports with minimal human intervention.

However, the talk also acknowledges the challenges and threats posed by AI, such as complacency among practitioners relying too heavily on AI-generated insights. Issues of equity, legal implications, and burnout due to increased complexity in patient cases are highlighted as potential concerns.2

Merola responded to AI concerns and said, “I think my eyes were opened a bit to some of the potential pitfalls and concerns, including payer use of AI to stratify the patients that we might see or how we might get reimbursed for those patients. You can imagine also, what was interesting study from the IBD literature, [is] that sometimes we can rely on AI a bit too much and make our own brains a little lazy, if you will, to what we're seeing in front of us. In fact, we have to be vigilant to that and continue to do what we do what we do best.”

In conclusion, Stidham envisions a future where AI plays a crucial role in managing larger patient populations, allowing healthcare practitioners to focus on more critical aspects of patient care. However, he emphasized the need for active involvement, education, and ethical considerations to ensure the responsible and effective integration of AI in healthcare.

“[I’ve accepted] using AI as a tool, but not as the only tool. In our practice. I think there's no question that AI is here is here to stay. I think it'llbe interesting to see how we as a community decide to utilizeand embrace portions of AI to augment our practice, but not certainly to replace what we do and only we can do best,” Merola concluded from a dermatologist’s perspective.

As the AI future unfolds, collaboration between different medical specialties, such as dermatology and gastroenterology, may provide valuable insights and shared learnings.

References

  1. Stidham RW. AI in IBD: What is the Gastroenterology Perspective. Presented at: Masterclasses in Dermatology February 16-19, 2024; Puerto Rico.
  2. Stidham RW, Takenaka K. Artificial Intelligence for Disease Assessment in Inflammatory Bowel Disease: How Will it Change Our Practice?. Gastroenterology. 2022;162(5):1493-1506. doi:10.1053/j.gastro.2021.12.238
Related Videos
© 2024 MJH Life Sciences

All rights reserved.