Skin Cancer and Artificial Intelligence
Artificial intelligence (AI) is a specialty within computer science that is focused on creating systems that can replicate human intelligence and problem-solving abilities. It utilizes and processes data, learning from the past to help streamline and improve future tasks.
This may sound like something out of a science fiction book or movie, AI was first created over 65 years ago. The term “AI” was first used in 1955 by John McCarthy of MIT (Massachusetts Institute of Technology) while holding a workshop at Dartmouth College.
AI in medicine (AIM) creates opportunities for personalized medicine and can be used for predictive models for disease diagnosis, prediction of therapeutic responses and potentially preventive medicine. While the discussion and full application of AIM goes beyond the scope of this blog post, a review of its potential utility in diagnosing dermatologic malignancies seems worthy for skin cancer awareness month.
AI software is now being used to identify and distinguish between benign nevi and melanoma. Skin lesion images are broken down to the pixel level for individual analysis using the techniques of broad categorization of algorithms that encompass subcategories, including machine learning (ML), natural language processing (NLP), and deep learning (DL) to predict and classify malignancies. One landmark paper (Jones et al, 2022) showed some systems have high sensitivities and specificities when distinguishing malignant from benign lesions. There are studies exploring metastatic disease from skin cancer. One such study (Jansen et al, 2023) looked at the histological tissue sections of sentinel lymph nodes in the convolutional neural network models to identify the presence of metastases with high sensitivity and specificity.
Machine learning techniques are also being applied to Mohs micrographic surgery slides, incorporating to classify basal cell carcinomas. While other researchers have developed models to interpret indirect immunofluorescent microscopies to help classify bullous dermatoses.
Before clinicians worry that we are entering “The Matrix,” know that this research is still in its infancy and is fraught with a myriad of issues. Biases, regulatory hurdles, interpretability, data sets, and how this would fit into current clinical workflows are just a few of the complex issues that need to be addressed before these tools are widely used in daily clinical practice. Dermatology NPs and PAs will hopefully be a part of an interdisciplinary effort to understand the workings of AI in dermatology in a fair, safe, and responsible manner.
References:
Jones, OT , Matin, RN , van der Schaar, M , Prathivadi Bhayankaram, K , Ranmuthu, CKI , Islam, MS, et al. Artificial intelligence and machine learning algorithms for early detection of skin cancer in community and primary care settings: a systematic review. Lancet Digit Health. (2022) 4:e466–76. doi: 10.1016/S2589-7500(22)00023-1
Jansen, P , Baguer, DO , Duschner, N , Arrastia, JL’C , Schmidt, M , Landsberg, J, et al. Deep learning detection of melanoma metastases in lymph nodes. Eur J Cancer. (2023) 188:161–70. doi: 10.1016/j.ejca.2023.04.023
Esteva, A , Kuprel, B , Novoa, RA , Ko, J , Swetter, SM , Blau, HM, et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature. (2017) 542:115–8. doi: 10.1038/nature21056
Kaul, V., Enslin,S., & Gross, S. (2020). History of artificial intelligence in medicine. Gastrointestinal Endoscopy, (92)4: 807-12. Accessed from https://www.giejournal.org/article/S0016-5107%2820%2934466-7/pdf on April 15, 2024.
What is the history of artificial intelligence? Tableau.com. Accessed from https://www.tableau.com/data-insights/ai/history#definition on April 15, 2024.
Omiye JA, Gui H, Daneshjou R, Cai ZR and Muralidharan V (2023) Principles, applications, and future of artificial intelligence in dermatology. Front. Med. 10:1278232. doi: 10.3389/fmed.2023.1278232. Accessed from: https://www.frontiersin.org/articles/10.3389/fmed.2023.1278232/full#:~:text=3.-,Applications%20of%20AI%20in%20dermatology,%2C%20to%20text%2Dbased%20analyses on April 15, 2024.
Justin Love MPAS, PA-C resides in the blue zone of Loma Linda CA. He works for Loma Linda University Department of Dermatology. In his spare time, he enjoys any ocean related activities and spending time with his family.
Victoria Garcia-Albea, BSN, MSN, RN, PNP, DCNP, is a medical dermatology nurse practitioner at Lahey Clinic in Burlington, MA. She is the director of the Lahey Clinic Dermatology NP Training Program. She spends most of her free time with her husband and two school-aged boys, and volunteers at her public library.
IMPORTANT NOTICES: Any medical information on this website is not intended as a substitute for informed medical advice and you should not take any action before consulting with a licensed healthcare professional.
Any licensed healthcare professional in a healthcare encounter must independently make decisions on medical testing, diagnosis, medication prescribing, and treatment administration based on their professional judgment and training, regardless of the information on this website.