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
Dermatopathology and Augmented Intelligence (AUI)
Description
In her presentation, Shruthi Agarwal, a dermatologist and dermatopathologist at Mayo Clinic, discusses the evolution of dermatopathology towards digital pathology and the integration of Augmented Intelligence (AI) in this field. She highlights the transition from traditional glass slides to digital imaging, enabling the development of AI models that assist in detecting, characterizing, quantifying, and predicting outcomes in dermatopathology. While many dermatopathologists express optimism about AI enhancing diagnostics, some harbor concerns about job security and the technology’s reliability. Agarwal emphasizes AI's potential in detecting melanoma metastases and classifying tumors accurately, while also pointing to areas where AI is still developing, such as evaluating inflammatory skin conditions. The discussion notes the importance of combining AI with comprehensive patient data for improved outcomes and personalization of care. Agarwal warns against over-reliance on AI, stressing the necessity for pathologists to maintain critical diagnostic skills, especially in recognizing subtle mimickers of diseases. She concludes by underscoring the need for continued education and careful assessment of AI’s role in dermatopathology, while acknowledging the exciting possibilities it offers for the future.
View moreConclusions
- Digital pathology is leading to enhanced diagnostic workflows in dermatopathology through the use of augmented intelligence models.
- Dermatopathologists have optimistic views on the potential of augmented intelligence to revolutionize their field, though some harbor concerns about being replaced by AI.
- Augmented intelligence is viewed as more beneficial for diagnosing cutaneous tumors compared to inflammatory skin diseases.
- Models trained through augmented intelligence demonstrate high sensitivity and accuracy for detecting conditions like melanoma and other skin tumors.
- There is a growing emphasis on the potential to use AI for stratifying prognosis and improving treatment outcome predictions based on histopathological features.
- Education and training in augmented intelligence can enhance dermatopathology but also requires the integration of informatics knowledge and responsibility for both input and output of AI systems.
- Challenges such as automation bias and the need for continuous updates in diagnostic algorithms are critical considerations in the implementation of AI in dermatopathology.
- Xiyue Wang # 1 2, Junhan Zhao # 1 3, Eliana Marostica 1 4, Wei Yuan 5, Jietian Jin 6, Jiayu Zhang 5, Ruijiang Li 2, Hongping Tang 7, Kanran Wang 8, Yu Li9,Fang Wang10, Yulong Peng 11, Junyou Zhu 12, Jing Zhang 5, Christopher R Jackson 1 13 14, Jun Zhang 15, Deborah Dillon 16, Nancy U Lin 17, Lynette Shol| 16 18, Thomas Denize 16 18, David Meredith 16, Keith L Ligon 16 18, Sabina Signoretti 16 18, Shuji Ogino 16 19 20 , Jeffrey A Golden 16 21, MacLean P Nasrallah 22, Xiao Han 15, Sen Yang 23 24,
- Brian Potter, M.D ., and Salve G. Ronan, M.D ., Chicago, IL, 2025, J Am Acad Dermatol, Computerized dermatopathologic diagnosis.
- Noora Neittaanmäki2,5 and John Paoli 1,2, 2024, Am J Dermatopathol, Attitudes Toward Artificial Intelligence Within Dermatopathology: An International Online Survey.
- Shoko Vos 1, Konnie Hebeda 2, Megan Milota 3, Martin Sand 4, Jojanneke Drogt 3, Katrien Grünberg 2, Karin Jongsma 3, 2025, Mod Pathol, Making Pathologists Ready for the New Artificial Intelligence Era: Changes in Required Competencies.
- Alan N. Snyder, MD, Dan Zhang MS1, Steffen L. Dreesen BS1, Christopher A. Baltimore BS *, Dan R. Lopez-Garcia, 2025, Histologic Screening of Malignant Melanoma, Spitz, Dermal and Junctional Melanocytic Nevi Using a Deep Learning Model.
- Kristina Bang Christensen b, Jeanette Bæhr Georgsen a b, Patricia Switten Nielsen a b, 2025, Objective assessment of tumor infiltrating lymphocytes as a prognostic marker in melanoma using machine learning algorithms.