Preparing for the AI Revolution: A Guide for Dermatology Nursing Professionals
There is growing evidence regarding the role of artificial intelligence (AI) in dermatology, including its applications in diagnosing skin conditions, enhancing clinical workflows, and aiding research. While there is a paucity of data on the utilization of AI in dermatology nursing, in my work, I explain the latest evidence and discuss the implications for nursing practice and innovation. Specifically, in this article, I explain basic AI concepts and how AI impacts dermatology nursing, education, and research. To address the gaps in the literature, I share suggestions for DNP projects and nursing leadership in dermatology with an emphasis on how AI-driven solutions can enhance both patient care and professional development. The recommendations for research can apply to students as well as DNP graduates who typically have more time to dedicate than we experience during our doctoral program. In an upcoming article, I'll share recommendations specifically for entrepreneurs.
"To celebrate Article #100, I'm writing about one of my favorite topics: technology that can be used to increase access to Dermatology Nursing | Education | Research! Thank you for taking this journey with me. I have learned that the journey is the destination. This era of my life, started with a dream to write. To be writing two years later and acknowledged for my contributions to dermatology beyond this blog, has been a surprise every step of the way. The best part? Hearing the impact it's had on you. Thank you for taking the time to share that with me."
Digital Dermatology: Looking Ahead
The field of dermatology is undergoing a digital transformation, driven by advancements in artificial intelligence, telemedicine, and data-driven care. These innovations have the potential to expand access, improve diagnostic accuracy, and enhance patient outcomes. However, they also present challenges that must be addressed to ensure ethical and effective implementation. Artificial intelligence (AI) and machine learning are at the forefront of this transformation.
AI-powered algorithms are increasingly being used for image recognition and diagnostic support, particularly for conditions such as skin cancer and inflammatory diseases. Deep learning models trained on vast image datasets offer the potential for early and more accurate detection, but these tools require careful validation to ensure their effectiveness across diverse skin tones. Teledermatology has also seen rapid growth, with virtual consultations providing access to dermatologic care for patients in underserved and rural communities.
While these digital platforms have expanded patient reach, they work best in hybrid models that balance the efficiency of remote care with the hands-on precision of in-person assessments. Similarly, wearable technology and mobile health apps are enabling continuous skin monitoring, helping both patients and dermatology nurse practitioners track disease progression and treatment responses. As patient-generated data becomes more integrated into electronic health records, personalized dermatologic care is moving closer to reality.
Big data and precision dermatology are shaping a future where treatment plans are tailored to an individual’s genetic, environmental, and lifestyle factors. However, the rapid adoption of these technologies brings ethical considerations, including concerns about data privacy, the security of patient information, and the risk of bias in AI models. Many existing AI algorithms in dermatology have been trained on predominantly lighter skin tones, raising concerns about accuracy and equity in diagnosing and treating patients with skin of color. Addressing these disparities requires intentional efforts to diversify datasets and ensure AI models work effectively across all populations. Regulatory oversight and clear implementation guidelines will be essential in determining how AI and digital dermatology tools are safely and ethically integrated into clinical practice.
In addition to these regulatory challenges, provider and patient adoption remains a key consideration. While AI has the potential to augment clinical decision-making, digital literacy and trust in technology continue to be barriers to widespread implementation. Education, transparency, and collaboration among healthcare professionals, technologists, and policymakers will be critical in addressing these concerns. As digital dermatology continues to evolve, nurse practitioners, researchers, and entrepreneurs play a pivotal role in shaping its ethical and practical application.
By advocating for inclusive AI training data, expanding access to telehealth, and influencing policy development, we can help guide this transformation to ensure that technology amplifies—rather than replaces—clinical expertise. The shift toward digital dermatologic care is not about eliminating the need for human providers but about leveraging technology to deliver more accessible, efficient, and personalized patient care. The future of dermatology is digital, and the time to establish authority is now.
The Ascending Role of Artificial Intelligence in Dermatology
The integration of AI into dermatology is rapidly evolving, driven by advancements in machine learning (ML), deep learning (DL), and natural language processing (NLP) – see definitions below. The visually intensive nature of dermatology, coupled with the increasing availability of clinical images, dermoscopy data, and electronic health records (EHRs), creates a fertile ground for the application of AI technologies. Further, the existing shortage of dermatologists and limited access to dermatological services in many areas underscore the compelling need for AI-augmented solutions to bridge this gap and an opportunity for dermatology nurses and nurse practitioners in clinical practice and business (Glines et al., 2020; Omiye et al., 2023).
Fundamental Principles for Dermatology Nurse Practitioners
Artificial Intelligence at its Core:
At its most fundamental level, AI refers to computer systems designed to mimic human cognitive functions. Think of abilities like learning, problem-solving, decision-making, and understanding language – AI aims to replicate these in machines. In the context of dermatology, instead of a human doctor looking at a skin lesion and deciding if it's cancerous, an AI system can be trained to do the same.
How AI Manifests in Dermatology: Computational Subfields
Within dermatology, AI's capabilities are primarily driven by two major computational subfields: Machine Learning (ML) and Natural Language Processing (NLP). Increasingly, multimodal approaches that combine different types of data are also becoming significant.
Machine Learning (ML): Learning from Data
The Basic Idea: ML algorithms enable computers to learn from data without being explicitly programmed for each specific task. Instead of someone writing a rigid set of rules to identify melanoma, an ML algorithm is fed thousands of images of skin lesions (some melanoma, some not) and it learns to identify the patterns and features that distinguish them.
Deep Learning (DL), Advanced Pattern Recognition: Deep learning (DL) is a more advanced subset of ML that uses artificial neural networks, inspired by the structure of the human brain. These networks consist of multiple layers that can learn very complex patterns and relationships directly from raw data, like the pixels of a skin image.
Convolutional Neural Networks (CNNs), For Images: CNNs are a type of DL particularly well-suited for analyzing images. In dermatology, CNNs are used extensively to process clinical photographs and dermoscopy images to identify and classify skin lesions, like distinguishing melanoma from benign nevi.
Transformer Models, Understanding Sequences: Transformer models are another type of neural network that excel at understanding context and relationships in sequential data. While traditionally used for language, they can also be applied to analyze sequences of medical events or even features extracted from images over time (Omiye et al., 2023).
How ML Algorithms Learn:
Supervised Learning, Learning with Labels: This is the most common type of ML used in image-based dermatology AI. In supervised learning, the algorithm is trained on a labeled dataset, meaning each image or data point is paired with the correct answer (e.g., an image labeled as "melanoma" or "benign nevus"). The algorithm learns to map the input data (the image) to the correct output (the diagnosis) and can then make predictions on new, unseen images. Most AI models that classify skin cancers from images use supervised learning.
Unsupervised Learning, Finding Hidden Patterns: Unsupervised learning involves training models on unlabeled data to discover inherent patterns and structures. For example, it could be used to group similar-looking skin lesions together without prior knowledge of what those groups represent.
Reinforcement Learning, Learning by Trial and Error: Reinforcement learning involves an "agent" (the algorithm) interacting with an "environment" and learning optimal behavior through feedback (rewards or penalties). While less common in direct diagnostic tasks in dermatology currently, it could potentially be used in areas like optimizing treatment plans over time (Omiye et al., 2023).
Natural Language Processing: Understanding and Generating Language
Focus on Text Data: NLP is the branch of AI dedicated to enabling computers to understand, interpret, and generate human language. In dermatology, this is crucial for analyzing the vast amounts of unstructured text data found in EHRs, such as clinical notes, patient complaints, and research papers.
Understanding and Generating: NLP encompasses both natural language understanding (NLU), which focuses on deciphering the meaning of text, and natural language generation (NLG), which focuses on creating new text.
Large Language Models (LLMs): The Power of Text: Recent advancements in NLP have led to the development of LLMs, such as ChatGPT. These models have the ability to understand and generate human-like text with remarkable fluency. In dermatology, LLMs can be used to analyze dermatology nurse practitioner progress notes to identify key discussion points, answer patient inquiries about conditions like melanoma, and even generate patient education materials for acne (Omiye et al., 2023).
Multimodal Approaches: Combining Different Data Types
Integrating Images and Text (and More): Multimodal AI models go beyond analyzing just images or just text; they can take multiple types of data as input and learn from their relationships. In dermatology, this is particularly relevant because diagnosis and treatment often involve considering both visual information (from images) and textual information (like patient history, symptoms, and lab results).
More Robust and Clinically Relevant AI: By integrating different data types, multimodal models can potentially build a more comprehensive understanding of a patient's condition, leading to more accurate diagnoses and personalized treatment plans. Examples of emerging multimodal models include Med-PaLM M and LLaVa-Med.
Foundation Models (FMs), Learning Broadly, Adapting Specifically: FMs are AI models trained on a massive amount of diverse, unlabeled data. Once trained on this broad data, they can be fine-tuned (adapted) for specific downstream tasks, such as image classification or text analysis in dermatology. This approach allows these models to leverage vast amounts of information and then apply that knowledge to specific medical applications, showing great promise for learning from the rich diversity of medical data (Omiye et al., 2023).
These principles underpin a wide array of applications in dermatology:
Skin Malignancies: AI excels in identifying and distinguishing between benign nevi and melanoma with high sensitivity and specificity by analyzing skin lesion images at the pixel level using Deep Learning techniques. Models can also classify various malignancies and even detect metastases in lymph nodes.
Inflammatory Skin Diseases: AI is being applied to the diagnosis and management of conditions like psoriasis, dermatitis, and acne through image analysis of skin, nails, and scalp. Machine Learning (ML) techniques can also predict the risk of associated conditions, the efficacy of biologic therapies, and identify drug targets and biomarkers. Similarly, AI aids in diagnosing dermatitis, predicting disease severity, and even predicting skin sensitization potential. Acne lesion segmentation and severity grading from smartphone images are also being explored.
Ulcer Assessment: AI-powered image segmentation techniques are used to identify and measure wound boundaries, with applications in predicting and preventing pressure ulcers through analysis of images and even body heat maps.
Dermatopathology: ML techniques are being implemented to classify conditions like basal cell carcinoma in digitized histology slides, aiding in melanoma diagnoses and interpreting immunofluorescence microscopies.
Miscellaneous Multiclass Classification and Text-Based Analysis: AI systems are being developed for multi-class classification problems, providing ranked lists of potential diagnoses and creating body distribution maps for various skin conditions. NLP is used to analyze patient experiences on social media and identify key discussion topics in clinical notes, offering valuable insights into patient perceptions and the impact of diseases. LLMs like ChatGPT are being explored for patient guidance, administrative assistance for clinicians, and trainee education (Omiye et al., 2023).
The Emergence of Human-AI Collaboration
The performance of AI algorithms in dermatology is increasingly being compared to that of clinicians, with some studies showing AI models matching or even exceeding the diagnostic accuracy of dermatologists. This has led to the development of AI-based assistive tools designed to augment clinical decision-making in real-world settings, with promising results in pilot studies and randomized trials.
Navigating Limitations and Ethical Considerations
Despite the significant progress, several limitations and ethical considerations must be addressed for the widespread adoption of AI in dermatology:
Datasets: Biases and confounders in training datasets can compromise the validity and fairness of AI algorithms. Issues like surgical pen markers leading to incorrect malignancy classifications and underperformance on darker skin tones due to lack of representation highlight the critical need for diverse and equitable data.
Image Quality and Image Capturing Modalities: Heterogeneity in image sources, quality, and capture settings can significantly impact AI model performance. Establishing robust image-capturing standards and data formats like DICOM is being recommended.
Black Box Problem: The lack of transparency in the reasoning of many AI algorithms makes it difficult for clinicians to understand and trust their outputs. Efforts are underway to improve interpretability through techniques like saliency maps.
Implementation Challenges: Integrating AI into existing clinical workflows faces medical-legal hurdles related to patient consent, data privacy, and liability. The continuous learning nature of AI models necessitates vigilant monitoring and ongoing validation after regulatory approval. A lack of high-quality prospective randomized controlled trials in diverse real-world settings also hinders the validation and generalizability of AI models. Building trust among all stakeholders is essential for successful integration (Omiye et al., 2023).
AI for Streamlining Clinical Trials
Beyond direct patient care, AI is also demonstrating its value in optimizing clinical trial processes. Tools like the Retrieval Augmented Generation Enabled Clinical Trial Infrastructure for Inclusion Exclusion Review (RECTIFIER), an LLM-based tool, can significantly reduce the time required for screening patient eligibility by efficiently parsing unstructured EHR data and automating the comparison against inclusion and exclusion criteria. This AI-assisted screening can lead to faster trial completion and earlier access to novel therapies (Hswen & Collins, 2025).
In a recent study by Hswen and Collins (2025), the authors highlighted that AI-assisted screening, while still requiring human oversight ("human-in-the-loop"), dramatically reduced the number of patients needing manual screening and accelerated the identification of eligible participants. Importantly, the AI tool did not appear to increase false-positive eligibility assessments. While initial results are promising and the technology is being beta-tested for broader implementation, further validation across multiple sites and disease areas is necessary to fully realize its potential (Hswen & Collins, 2025).
DNP Project and PhD Nurse-Led Research Recommendations in AI and Dermatology
The integration of artificial intelligence in dermatology presents unique opportunities for PhD nurse-led research, DNP Projects, and collaborative interventions. Below are key areas where we can lead impactful initiatives to improve patient outcomes and shape AI’s role in dermatologic care.
Implementation of Generalist Medical AI in Dermatology Nursing Practice
DNP Project Idea: Evaluate the effectiveness of AI-driven decision support tools in dermatology clinics to enhance NP diagnostic accuracy, treatment recommendations, and patient education.
PhD Research Focus: Investigate how large language models and multimodal AI influence clinical decision-making, patient safety, and provider trust in dermatology.
Ethical AI Development for Skin of Color
DNP Project Idea: Develop guidelines to improve dermatology AI datasets by integrating images of diverse skin tones and testing AI-based decision support tools for bias detection.
PhD Research Focus: Analyze disparities in AI-driven dermatologic diagnoses and propose methodologies for equitable model training, particularly in underrepresented populations.
Federated Learning and Privacy-Preserving AI for Dermatology Care
DNP Project Idea: Pilot a federated learning framework within a dermatology telehealth practice, evaluating its impact on patient privacy, diagnostic accuracy, and data security.
PhD Research Focus: Assess the feasibility and effectiveness of federated learning in dermatology AI development to create fairer, more generalizable models across diverse patient populations.
AI Model Evaluation Beyond Accuracy: Clinical Utility and Nursing Integration
DNP Project Idea: Develop and implement an AI model evaluation framework tailored to nurse practitioners, focusing on usability, transparency, and clinical value in dermatology settings.
PhD Research Focus: Investigate the role of nurse-led AI validation studies in dermatology, analyzing model performance across different patient subgroups and care settings.
AI-Driven Interventions for Dermatologic Care in Resource-Limited Settings
DNP Project Idea: Implement an AI-powered diagnostic tool in a rural or underserved dermatology clinic, measuring its impact on diagnostic confidence and treatment planning by NPs.
PhD Research Focus: Explore the barriers and facilitators of AI adoption in dermatology within resource-limited settings, with a focus on health equity and accessibility.
Regulatory and Ethical Considerations for AI Deployment in Dermatology Nursing
DNP Project Idea: Develop a nurse-led policy framework for the ethical integration of AI in dermatology practices, addressing safety, bias mitigation, and regulatory compliance.
PhD Research Focus: Examine the ethical implications of AI-assisted dermatologic diagnosis, focusing on provider liability, patient perceptions, and informed consent practices.
Call to Action for DNP and PhD Nurses
The future of AI in dermatology is shaped not just by technology but by the ethical, clinical, and systemic frameworks that guide its use. DNP and PhD-trained nurses and nurse practitioners have a critical role in researching, implementing, and evaluating AI-driven dermatology solutions to enhance patient care, improve diversity and transparency, and ensure responsible AI deployment.
Conclusion
In conclusion, AI holds immense promise for transforming dermatology, from enhancing diagnostic accuracy and personalizing treatment approaches to improving efficiency in clinical research. As AI continues to mature, a focus on addressing limitations, upholding ethical standards, and fostering interdisciplinary collaboration will be essential to ensure the responsible and effective integration of AI into dermatological practice, ultimately benefiting both clinicians and patients. In an upcoming article, I'll share recommendations specifically for entrepreneurs to address gaps in nursing, education, and research.
Hswen, Y. & Collins N. (2025). Study Finds AI Can Quickly Prescreen Patients for Clinical Trials, Speeding Enrollment. JAMA, 333(15), 1275–1277. PMID: 40085112
Dr. Kimberly Madison, DNP, AGPCNP-BC, WCC, is a Board-Certified, Doctorally-prepared Nurse Practitioner, educator, and author dedicated to advancing dermatology nursing education and research with an emphasis on skin of color. As the founder of Mahogany Dermatology Nursing | Education | Research™ and the Alliance of Cosmetic Nurse Practitioners™, she expands access to dermatology research, business acumen, and innovation while also leading professional groups and mentoring clinicians. Through her engaging and informative social media content and peer-reviewed research, Dr. Madison empowers nurses and healthcare professionals to excel in dermatology and improve patient care.
.svg)
.svg)
.svg)
