Please login or create an account. If you do not have access to this content, you will be shown a 30 second preview and licensing options.
- Presentation
Technology in Dermatologic Surgery
Description
The discussion focuses on advancements and practical applications of technology in dermatologic surgery, particularly the use of artificial intelligence (AI) in pathology and specific surgical techniques for scar revision. The speaker highlights the importance of engaging with emerging technologies, including AI algorithms for reading pathology slides, especially in diagnosing basal cell carcinoma through segmentation rather than simple classification. Notable advancements in segmentation models show promise but require further development before they become mainstream in clinical settings. The speaker addresses practical applications for dermatologists, such as mobile apps that assist with surgery criteria and patient risk assessments based on skin cancer data. Furthermore, the presentation covers outdated (paleo) techniques, like effective communication methods in the operating room. The second part is dedicated to strategies for dealing with various types of scars, ranging from atrophic to hypertrophic and contracture, by employing a collagen-focused approach. Specific cases are referenced to illustrate methods like subcision, laser treatments, and Z-plastic techniques to resolve scar issues. The speaker encourages practical use of these advancements in daily practice, emphasizing the importance of ongoing education and adaptation to improve outcomes in dermatologic surgery.
View moreConclusions
- The integration of AI into dermatologic surgery has shown promise but is not yet ready for clinical implementation.
- Current AI algorithms for detecting basal cell carcinoma in frozen section pathology demonstrate variable performance, necessitating further research and improvement.
- Segmentation techniques in AI are challenging due to issues like false positives and lack of robust metrics for evaluation.
- Collaboration and multi-institutional data sharing are essential to improve the generalizability and accuracy of AI algorithms in dermatopathology.
- Clinically relevant outcomes should be prioritized in upcoming studies to ensure AI tools are beneficial for patient care.
- Various practical applications, like risk assessment tools for skin cancer, have been developed and validated for real-world use.
- Holistic patient management strategies, including risk categorization, can guide treatment decisions for skin cancer.
- Traditional communication tools, like headsets, can enhance team efficiency in surgical settings despite having some drawbacks.
- Varra et al: Sn/Sp .71/.75
- Van Zon et al: DICE score 0.66
- Bonnefille et al: 96% of BCC foci (843/877)
- Tan et al: AUC .943, DICE 0.90
- Jambusaria-Pahlajani, et al ., Transplant International. 32: 1259-1267; 2019.
- Gomez-Tomas, et al ., JAMA Dermatol. 159(1): 29-36; 2023.