Verma Kritin K, Grabow Kurt M, Koch Ryan S, Friedmann Daniel P, Tarbox Michelle B
Texas Tech University Health Sciences Center School of Medicine, Lubbock, Texas, USA.
College of Medicine, Texas A&M University College of Medicine, Dallas, Texas, USA.
Proc (Bayl Univ Med Cent). 2025 May 5;38(4):577-578. doi: 10.1080/08998280.2025.2489873. eCollection 2025.
The use of artificial intelligence (AI) in dermatology, particularly for the diagnosis of melanoma, has demonstrated potential in improving early detection of cancer. Current AI-based systems, such as DermaSensor and Nevisense, have shown high sensitivity. In addition, open-source models like All Data Are Ext (ADAE) continue to show promise. Ethical, practical, and privacy concerns remain despite these advancements. Key challenges with these models include maintaining transparency with patients, ensuring privacy of patient data, and addressing discrepancies between AI and clinical determinations. Additional research, regulatory guidance, and open conversations are necessary to realize AI's full potential in the field of dermatology while preserving patient trust.
人工智能（AI）在皮肤病学中的应用，尤其是用于黑色素瘤的诊断，已显示出在改善癌症早期检测方面的潜力。当前基于AI的系统，如DermaSensor和Nevisense，已显示出高灵敏度。此外，像All Data Are Ext（ADAE）这样的开源模型也持续展现出前景。尽管有这些进展，但伦理、实际操作和隐私问题仍然存在。这些模型的关键挑战包括对患者保持透明度、确保患者数据的隐私以及解决AI与临床诊断之间的差异。需要进行更多研究、监管指导和公开讨论，以在保持患者信任的同时，充分发挥AI在皮肤病学领域的潜力。
