Ethical considerations for artificial intelligence in dermatology: a scoping review

Applications, Ethical Considerations, and Patient Perspectives on Artificial Intelligence in Dermatologic Surgery: Focus on Mohs Surgery

BACKGROUND

Artificial intelligence (AI) is increasingly used in health care and has the potential to revolutionize Mohs micrographic surgery (MMS) by improving diagnostic accuracy and workflow efficiency.

OBJECTIVE

To provide an overview of AI in MMS, including its principles, applications, limitations, ethical considerations, and patient perceptions.

MATERIALS AND METHODS

A comprehensive search of PubMed was conducted for studies published from 2010 onward using the following key terms: ("Artificial Intelligence" OR "Machine Learning") AND ("Dermatologic Surgery" OR "Dermatology" OR "Mohs Surgery") AND ("Ethics" OR "Patient Perspectives"), along with related MeSH terms.

RESULTS

AI applications, including diagnostic support, histologic classification, and workflow optimization, demonstrate potential to enhance efficiency and clinical outcomes. Studies report high accuracy in tumor detection, case complexity prediction, and surgical planning, yet challenges remain, particularly in data privacy, transparency, informed consent, and algorithmic bias. Patient attitudes are mixed, with most preferring AI as an assistive tool rather than a standalone diagnostic system. Although AI has been shown to streamline workflows and improve procedural efficiency, its clinical adoption is still in early stages, requiring further validation, regulatory approval, and standardization.

CONCLUSION

AI can enhance efficiency, accuracy, and outcomes in MMS, but its implementation must prioritize patient welfare, transparency, and accountability.

Evaluating the Diagnostic and Treatment Capabilities of GPT-4 Vision in Dermatology: A Pilot Study

BACKGROUND

The integration of generative artificial intelligence within dermatology presents a new frontier for enhancing diagnostic accuracy and treatment planning.

OBJECTIVE

This research evaluates Generative Pre-trained Transformer-4 Vision's (GPT-4V) performance in accurately diagnosing and generating treatment plans for common dermatological conditions, comparing its assessment of textual versus image data and its performance with multimodal inputs.

METHODS

A dataset of 102 images representing 9 common dermatological conditions was compiled from dermatlas.org and dermnet.nz. Images were screened by 2 board-certified dermatologists and were excluded if they did not represent a classic presentation of the respective conditions. Fifty-four images were included in the final analysis. In addition, 9 text-based clinical scenarios corresponding to each condition were developed. GPT-4V's diagnostic capabilities were assessed across 3 setups: Image Prompt, Scenario Prompt, and Image + Scenario Prompt.

RESULTS

In the Image Prompt setup, GPT-4V correctly identified the primary diagnosis for 54% of the images. The Scenario Prompt and the Image + Scenario Prompt setups, respectively, both achieved an 89% accuracy rate in identifying the primary diagnosis. Treatment recommendations were evaluated using a modified Entrustment Scale, showing competent but not expert-level performance. A Wilcoxon signed-rank test demonstrated a statistically significant difference in treatment recommendations based on the Entrustment Score, with the model performing better in the Image + Scenario setup (P < .01).

CONCLUSION

GPT-4V demonstrates the potential to augment dermatological diagnosis and treatment recommendations, particularly in text-based scenarios. However, its underwhelming performance in image-based diagnosis and integration of multimodal data highlights important areas for improvement.

The use of ChatGPT in the dermatological field: a narrative review

Artificial intelligence (AI) encompasses the development of computer systems capable of tasks typically requiring human intelligence, such as visual perception, speech recognition, decision-making and language translation. Over time, numerous applications have emerged, with the integration of AI into medicine marking a significant leap forward in healthcare delivery, diagnosis and treatment. Among medical specialties, dermatology stands at the forefront of AI advancements, leveraging machine learning and deep learning to enhance dermatologists' abilities and improve patient care. ChatGPT is an advanced language model by OpenAI, originally designed for conversations, which has expanded its utility into diverse fields, including healthcare and dermatology. In this context, the aim of this review article was to explore the synergistic relationship between ChatGPT and dermatology, examining how this innovative AI model is reshaping skin health management, its potential applications, preliminary data on its efficiency and accuracy, as well as ethical and legal concerns related to the use of its tool.

Advancing Sports Cardiology: Integrating Artificial Intelligence with Wearable Devices for Cardiovascular Health Management

Sports cardiology focuses on athletes' cardiovascular health, yet sudden cardiac death remains a significant concern despite preventative measures. Prolonged physical activity leads to notable cardiovascular adaptations, known as the athlete's heart, which can resemble certain pathological conditions, complicating accurate diagnoses and potentially leading to serious consequences such as unnecessary exclusion from sports or missed treatment opportunities. Wearable devices, including smartwatches and smart glasses, have become prevalent for monitoring health metrics, offering potential clinical applications for sports cardiologists. These gadgets are capable of spotting exercise-induced arrhythmias, uncovering hidden heart problems, and offering crucial information for training and recovery, to minimize exercise-related cardiac incidents and enhance heart health care. However, concerns about data accuracy and the actionable value of the obtained information persist. A major challenge lies in the integration of artificial intelligence with wearables, research gaps remain regarding their ability to provide real-time, reliable, and clinically relevant insights. Combining artificial intelligence with wearable devices can improve how data is managed and used in sports cardiology. Artificial intelligence, particularly machine learning, can classify, predict, and draw inferences from the data collected by wearables, revolutionizing patient data usage. Despite artificial intelligence's proven effectiveness in managing chronic conditions, the limited research on its application in sports cardiology, particularly regarding wearables, creates a critical gap that needs to be addressed. This review examines commercially available wearables and their applications in sports cardiology, exploring how artificial intelligence can be integrated into wearable technology to advance the field.

Grading facial aging: Comparing the clinical assessments made by three dermatologists with those obtained by an AI-based scoring system that analyses selfie pictures. A focus on Chinese subjects of both genders

OBJECTIVE

The objective of this study is to assess the correspondence, in live conditions, between clinical gradings of facial aging signs by three dermatologists and those afforded by an automatic AI-based algorithm that analyses smartphones' selfie images of Chinese subjects.

METHODS

In total, 125 Chinese subjects of both genders, aged 18-62y, took a selfie using their own smartphones and were immediately viewed by three dermatologists. The latter graded the severity of 15 facial signs in women and 9 in men, using the standardized values afforded by a Skin Aging Atlas referential dedicated to Asian skin. The data issued by both methodologies were then statistically compared.

RESULTS

The absolute gradings of the automatic system were found highly correlated with clinical assessments, with lower values in most cases. In women, large differences in absolute values were found on the gradings for size of isolated spot, cheek fold, spread macules, and texture of mouth contour women. Analysis of the Mean Absolute Errors (M.A.E) revealed that these rarely exceed 0.6 grading units in women and to a lesser extent in men.

CONCLUSION

The present study confirmed the value of the automatic system towards an extended use towards large human cohorts as a surrogate of clinical evaluations and allowed to detect the points where improvements must be brought to the system.

Knowledge, attitude, and perceptions of MENA researchers towards the use of ChatGPT in research: A cross-sectional study

Background

Artificial intelligence (AI) technologies are increasingly recognized for their potential to revolutionize research practices. However, there is a gap in understanding the perspectives of MENA researchers on ChatGPT. This study explores the knowledge, attitudes, and perceptions of ChatGPT utilization in research.

Methods

A cross-sectional survey was conducted among 369 MENA researchers. Participants provided demographic information and responded to questions about their knowledge of AI, their experience with ChatGPT, their attitudes toward technology, and their perceptions of the potential roles and benefits of ChatGPT in research.

Results

The results indicate a moderate level of knowledge about ChatGPT, with a total score of 58.3 ± 19.6. Attitudes towards its use were generally positive, with a total score of 68.1 ± 8.1 expressing enthusiasm for integrating ChatGPT into their research workflow. About 56 % of the sample reported using ChatGPT for various applications. In addition, 27.6 % expressed their intention to use it in their research, while 17.3 % have already started using it in their research. However, perceptions varied, with concerns about accuracy, bias, and ethical implications highlighted. The results showed significant differences in knowledge scores based on gender (p < 0.001), working country (p < 0.05), and work field (p < 0.01). Regarding attitude scores, there were significant differences based on the highest qualification and the employment field (p < 0.05). These findings underscore the need for targeted training programs and ethical guidelines to support the effective use of ChatGPT in research.

Conclusion

MENA researchers demonstrate significant awareness and interest in integrating ChatGPT into their research workflow. Addressing concerns about reliability and ethical implications is essential for advancing scientific innovation in the MENA region.

Chatting ethically: practical recommendations for ethical use of large language models in dermatology practice, research and education

As large language model (LLM) use increases, developing ethical standards for their applications to supplement dermatologists is critically important. Here, we summarize the primary applications of LLMs in dermatology, their ethical implications and recommendations for appropriate use.

Artificial intelligence and machine learning in dermatological research and healthcare: British Society for Investigative Dermatology Skin Club Report, Southampton, April 2024

The British Society of Investigative Dermatology is the annual meeting of the UK’s skin research community. At this year’s meeting in Southampton, there was a discussion on the history and potential of artificial intelligence (‘AI’) in health care.

The four experts who spoke at the meeting have summarized their lectures in this paper. There is a piece on Alan Turing, who proposed the ‘Turing Test’ to find out if a machine could think like a human. Concepts like ‘machine learning’ (a key tool in AI) are explained. Next, there is a piece on the challenges of using AI decision-making tools in the skin cancer pathway. We discuss AI/machine learning approaches to grouping patients and choosing the best treatments for people with ‘atopic dermatitis’ (or ‘eczema’). Finally, potential pitfalls in AI are highlighted, including the need to explain how AI makes decisions and approaches to achieving this.

There is much excitement about AI, and this paper captures the discussion from the meeting of the current state of AI in dermatology health care.

Perceived Worries in the Adoption of Artificial Intelligence Among Healthcare Professionals in Saudi Arabia: A Cross-Sectional Survey Study

The use of AI in the healthcare sector is facing some formidable concerns raised by the practitioners themselves. This study aimed to establish the concerns that surround the adoption of AI among Saudi Arabian healthcare professionals. Materials and methods: This was a cross-sectional study using stratified convenience sampling from September to November 2024 across health facilities. This study included all licensed healthcare professionals practicing for at least one year, whereas interns and administrative staff were excluded from the research. Data collection was conducted through a 33-item validated questionnaire that was provided in paper form and online. The questionnaire measured AI awareness with eight items, past experience with five items, and concerns in four domains represented by 20 items. Four hundred questionnaires were distributed, and the response rate was 78.5% (n = 314). The majority of the participants were females (52.5%), Saudis (89.2%), and employees of MOH (77.1%). The mean age for the participants was 35.6 ± 7.8 years. Quantitative analysis revealed high AI awareness scores with a mean of 3.96 ± 0.167, p < 0.001, and low previous experience scores with a mean of 2.65 ± 0.292. Data management-related worries came out as the top worry, with a mean of 3.78 ± 0.259, while the poor data entry impact topped with a mean of 4.15 ± 0.801; healthcare provider-related worries with a mean of 3.71 ± 0.182; and regulation/ethics-related worries with a mean of 3.67 ± 0.145. Health professionals' main concerns about AI adoption were related to data reliability and impacts on clinical decision-making, which significantly hindered successful AI integration in healthcare. These are the particular concerns that, if addressed through robust data management protocols and enhanced processes for clinical validation, will afford the best implementation of AI technology in an optimized way to bring better quality and safety to healthcare. Quantitative validation of AI outcomes and the development of standardized integration frameworks are subjects for future research.

Application of ChatGPT as a content generation tool in continuing medical education: acne as a test topic

The large language model (LLM) ChatGPT can answer open-ended and complex questions, but its accuracy in providing reliable medical information requires a careful assessment. As part of the AICHECK (Artificial Intelligence for CME Health E-learning Contents and Knowledge) Study, aimed at evaluating the potential of ChatGPT in continuous medical education (CME), we compared ChatGPT-generated educational contents to the recommendations of the National Institute for Health and Care Excellence (NICE) guidelines on acne vulgaris. ChatGPT version 4 was exposed to a 23-item questionnaire developed by an experienced dermatologist. A panel of five dermatologists rated the answers positively in terms of "quality" (87.8%), "readability" (94.8%), "accuracy" (75.7%), "thoroughness" (85.2%), and "consistency" with guidelines (76.8%). The references provided by ChatGPT obtained positive ratings for "pertinence" (94.6%), "relevance" (91.2%), and "update" (62.3%). The internal reproducibility was adequate both for answers (93.5%) and references (67.4%). Answers related to issues of uncertainty and/or controversy in the scientific community scored the lowest. This study underscores the need to develop rigorous evaluation criteria for AI-generated medical content and for expert oversight to ensure accuracy and guideline adherence.

Artificial intelligence in assisting pathogenic microorganism diagnosis and treatment: a review of infectious skin diseases

The skin, the largest organ of the human body, covers the body surface and serves as a crucial barrier for maintaining internal environmental stability. Various microorganisms such as bacteria, fungi, and viruses reside on the skin surface, and densely arranged keratinocytes exhibit inhibitory effects on pathogenic microorganisms. The skin is an essential barrier against pathogenic microbial infections, many of which manifest as skin lesions. Therefore, the rapid diagnosis of related skin lesions is of utmost importance for early treatment and intervention of infectious diseases. With the continuous rapid development of artificial intelligence, significant progress has been made in healthcare, transforming healthcare services, disease diagnosis, and management, including a significant impact in the field of dermatology. In this review, we provide a detailed overview of the application of artificial intelligence in skin and sexually transmitted diseases caused by pathogenic microorganisms, including auxiliary diagnosis, treatment decisions, and analysis and prediction of epidemiological characteristics.

Assessing the Impact of ChatGPT in Dermatology: A Comprehensive Rapid Review

Background/Objectives: The use of artificial intelligence (AI) in dermatology is expanding rapidly, with ChatGPT, a large language model (LLM) from OpenAI, showing promise in patient education, clinical decision-making, and teledermatology. Despite its potential, the ethical, clinical, and practical implications of its application remain insufficiently explored. This study aims to evaluate the effectiveness, challenges, and future prospects of ChatGPT in dermatology, focusing on clinical applications, patient interactions, and medical writing. ChatGPT was selected due to its broad adoption, extensive validation, and strong performance in dermatology-related tasks. Methods: A thorough literature review was conducted, focusing on publications related to ChatGPT and dermatology. The search included articles in English from November 2022 to August 2024, as this period captures the most recent developments following the launch of ChatGPT in November 2022, ensuring that the review includes the latest advancements and discussions on its role in dermatology. Studies were chosen based on their relevance to clinical applications, patient interactions, and ethical issues. Descriptive metrics, such as average accuracy scores and reliability percentages, were used to summarize study characteristics, and key findings were analyzed. Results: ChatGPT has shown significant potential in passing dermatology specialty exams and providing reliable responses to patient queries, especially for common dermatological conditions. However, it faces limitations in diagnosing complex cases like cutaneous neoplasms, and concerns about the accuracy and completeness of its information persist. Ethical issues, including data privacy, algorithmic bias, and the need for transparent guidelines, were identified as critical challenges. Conclusions: While ChatGPT has the potential to significantly enhance dermatological practice, particularly in patient education and teledermatology, its integration must be cautious, addressing ethical concerns and complementing, rather than replacing, dermatologist expertise. Future research should refine ChatGPT's diagnostic capabilities, mitigate biases, and develop comprehensive clinical guidelines.

AI integration in nephrology: evaluating ChatGPT for accurate ICD-10 documentation and coding

Background

Accurate ICD-10 coding is crucial for healthcare reimbursement, patient care, and research. AI implementation, like ChatGPT, could improve coding accuracy and reduce physician burden. This study assessed ChatGPT's performance in identifying ICD-10 codes for nephrology conditions through case scenarios for pre-visit testing.

Methods

Two nephrologists created 100 simulated nephrology cases. ChatGPT versions 3.5 and 4.0 were evaluated by comparing AI-generated ICD-10 codes against predetermined correct codes. Assessments were conducted in two rounds, 2 weeks apart, in April 2024.

Results

In the first round, the accuracy of ChatGPT for assigning correct diagnosis codes was 91 and 99% for version 3.5 and 4.0, respectively. In the second round, the accuracy of ChatGPT for assigning the correct diagnosis code was 87% for version 3.5 and 99% for version 4.0. ChatGPT 4.0 had higher accuracy than ChatGPT 3.5 (p = 0.02 and 0.002 for the first and second round respectively). The accuracy did not significantly differ between the two rounds (p > 0.05).

Conclusion

ChatGPT 4.0 can significantly improve ICD-10 coding accuracy in nephrology through case scenarios for pre-visit testing, potentially reducing healthcare professionals' workload. However, the small error percentage underscores the need for ongoing review and improvement of AI systems to ensure accurate reimbursement, optimal patient care, and reliable research data.

Exploring the Potential of DALL-E 2 in Pediatric Dermatology: A Critical Analysis

INTRODUCTION

Artificial intelligence (AI) is becoming increasingly explored for its potential applications in dermatology. Among various AI models, DALL-E 2 (San Francisco, CA: OpenAI), which generates de novoimages from textual inputs, has garnered significant attention for its remarkable photorealism. In our study, we aimed to analyze the performance of DALL-E 2 in the context of dermatology.

METHODS

The following 12 pediatric dermatological conditions common to ages <15 years were selected as tests: acne, atopic dermatitis, contact dermatitis, vitiligo, congenital melanocytic nevus, warts, molluscum contagiosum, seborrheic dermatitis, alopecia areata, infantile hemangioma, impetigo, and dermatophytosis, specifically tinea corporis. Representative morphological descriptions of each diagnosis, along with their corresponding names, were inputted into DALL-E 2 as textual prompts and subsequently compared. The accuracy of the generated images and their alignment with the intended descriptions were assessed.

RESULTS

Among the total of 24 images reported, 18 were photorealistic and six were cartoons. More cartoons were generated when providing the model with morphological descriptions as textual inputs compared to when diagnoses were inputted. While not entirely accurate, acne stood out as the only diagnosis that was the most consistent and closest to the actual diagnosis. Both images of acne portrayed erythematous papules scattered across the face. However, DALL-E 2 resulted in poor performance for the remaining eleven diagnoses. They did not accurately represent the intended diagnoses nor align with their counterpart image. Moreover, most of the generated images featured lighter skin tones.

CONCLUSION

In assessing DALL-E 2's applications in dermatology, our study highlights the need for more domain-specific and demographically inclusive training data in its algorithms to improve performance.

Applications of Large Language Models in Pathology

Large language models (LLMs) are transformer-based neural networks that can provide human-like responses to questions and instructions. LLMs can generate educational material, summarize text, extract structured data from free text, create reports, write programs, and potentially assist in case sign-out. LLMs combined with vision models can assist in interpreting histopathology images. LLMs have immense potential in transforming pathology practice and education, but these models are not infallible, so any artificial intelligence generated content must be verified with reputable sources. Caution must be exercised on how these models are integrated into clinical practice, as these models can produce hallucinations and incorrect results, and an over-reliance on artificial intelligence may lead to de-skilling and automation bias. This review paper provides a brief history of LLMs and highlights several use cases for LLMs in the field of pathology.