Position statement of the EADV Artificial Intelligence (AI) Task Force on AI-assisted smartphone apps and web-based services for skin disease.

BACKGROUND: As the use of smartphones continues to surge globally, mobile applications (apps) have become a powerful tool for healthcare engagement. Prominent among these are dermatology apps powered by Artificial Intelligence (AI), which provide immediate diagnostic guidance and educational resources for skin diseases, including skin cancer. OBJECTIVE: This article, authored by the EADV AI Task Force, seeks to offer insights and recommendations for the present and future deployment of AI-assisted smartphone applications (apps) and web-based services for skin diseases with emphasis on skin cancer detection. METHODS: An initial position statement was drafted on a comprehensive literature review, which was subsequently refined through two rounds of digital discussions and meticulous feedback by the EADV AI Task Force, ensuring its accuracy, clarity and relevance. RESULTS: Eight key considerations were identified, including risks associated with inaccuracy and improper user education, a decline in professional skills, the influence of non-medical commercial interests, data security, direct and indirect costs, regulatory approval and the necessity of multidisciplinary implementation. Following these considerations, three main recommendations were formulated: (1) to ensure user trust, app developers should prioritize transparency in data quality, accuracy, intended use, privacy and costs; (2) Apps and web-based services should ensure a uniform user experience for diverse groups of patients; (3) European authorities should adopt a rigorous and consistent regulatory framework for dermatology apps to ensure their safety and accuracy for users. CONCLUSIONS: The utilisation of AI-assisted smartphone apps and web-based services in diagnosing and treating skin diseases has the potential to greatly benefit patients in their dermatology journeys. By prioritising innovation, fostering collaboration and implementing effective regulations, we can ensure the successful integration of these apps into clinical practice.

Analysis of dermoscopic changes of blue nevi on digital follow-up: A 21-year retrospective cohort study.

BACKGROUND: Blue nevi are benign dermal melanocytic proliferations that are often easy to recognize clinically. Rarely, these lesions can display atypical features, suggesting the presence of a malignant blue nevus or mimicking cutaneous metastases of melanoma. OBJECTIVE: To describe the clinical evolution of blue nevi over time and to assess the need for monitoring these lesions. METHODS: We conducted a retrospective cohort study of 103 patients who were followed between December 1998 and November 2019. An artificial intelligence algorithm was used to identify blue nevi from the databases of two digital epiluminescence devices. Changes in the area of each lesion were calculated with a segmentation neural network. RESULTS: We included 123 blue nevi from 103 patients. Most of the lesions segmented, 99 (91.7%), were considered stable. Of the 9 (8.3%) growing blue nevi identified, 2 (1.85%) showed significant growth. The studied growing blue nevi turned out to be cellular blue nevi, presented with a low tumour mutation burden and GNAQ c.626A>T alteration was identified in both lesions. LIMITATIONS: Some clinical variants of blue nevi might not be included. CONCLUSIONS: Most blue nevi remain stable during their evolution. Rarely, they can show progressive growth, although histopathological or molecular signs of malignancy have not been identified.

Common genetic variants associated with melanoma risk or naevus count in patients with wildtype MC1R melanoma.

BACKGROUND: Hypomorphic MC1R variants are the most prevalent genetic determinants of melanoma risk in the white population. However, the genetic background of patients with wildtype (WT) MC1R melanoma is poorly studied. OBJECTIVES: To analyse the role of candidate common genetic variants on the melanoma risk and naevus count in Spanish patients with WT MC1R melanoma. METHODS: We examined 753 individuals with WT MC1R from Spain (497 patients and 256 controls). We used OpenArray reverse-transcriptase polymerase chain reaction to genotype a panel of 221 common genetic variants involved in melanoma, naevogenesis, hormonal pathways and proinflammatory pathways. Genetic models were tested using multivariate logistic regression models. Nonparametric multifactor dimensionality reduction (MDR) was used to detect gene-gene interactions within each biological subgroup of variants. RESULTS: We found that variant rs12913832 in the HERC2 gene, which is associated with blue eye colour, increased melanoma risk in individuals with WT MC1R [odds ratio (OR) 1·97, 95% confidence interval (CI) 1·48-2·63; adjusted P < 0·001; corrected P < 0·001]. We also observed a trend between the rs3798577 variant in the oestrogen receptor alpha gene (ESR1) and a lower naevus count, which was restricted to female patients with WT MC1R (OR 0·51, 95% CI 0·33-0·79; adjusted P = 0·002; corrected P = 0·11). This sex-dependent association was statistically significant in a larger cohort of patients with melanoma regardless of their MC1R status (n = 1497; OR 0·71, 95% CI 0·57-0·88; adjusted P = 0·002), reinforcing the hypothesis of an association between hormonal pathways and susceptibility to melanocytic proliferation. Last, the MDR analysis revealed four genetic combinations associated with melanoma risk or naevus count in patients with WT MC1R. CONCLUSIONS: Our data suggest that epistatic interaction among common variants related to melanocyte biology or proinflammatory pathways might influence melanocytic proliferation in individuals with WT MC1R. What is already known about this topic? Genetic variants in the MC1R gene are the most prevalent melanoma genetic risk factor in the white population. Still, 20-40% of cases of melanoma occur in individuals with wildtype MC1R. Multiple genetic variants have a pleiotropic effect in melanoma and naevogenesis. Additional variants in unexplored pathways might also have a role in melanocytic proliferation in these patients. Epidemiological evidence suggests an association of melanocytic proliferation with hormonal pathways and proinflammatory pathways. What does this study add? Variant rs12913832 in the HERC2 gene, which is associated with blue eye colour, increases the melanoma risk in individuals with wildtype MC1R. Variant rs3798577 in the oestrogen receptor gene is associated with naevus count regardless of the MC1R status in female patients with melanoma. We report epistatic interactions among common genetic variants with a role in modulating the risk of melanoma or the number of naevi in individuals with wildtype MC1R. What is the translational message? We report a potential role of hormonal signalling pathways in melanocytic proliferation, providing a basis for better understanding of sex-based differences observed at the epidemiological level. We show that gene-gene interactions among common genetic variants might be responsible for an increased risk for melanoma development in individuals with a low-risk phenotype, such as darkly pigmented hair and skin.

BCN20000: Dermoscopic Lesions in the Wild.

Advancements in dermatological artificial intelligence research require high-quality and comprehensive datasets that mirror real-world clinical scenarios. We introduce a collection of 18,946 dermoscopic images spanning from 2010 to 2016, collated at the Hospital Clínic in Barcelona, Spain. The BCN20000 dataset aims to address the problem of unconstrained classification of dermoscopic images of skin cancer, including lesions in hard-to-diagnose locations such as those found in nails and mucosa, large lesions which do not fit in the aperture of the dermoscopy device, and hypo-pigmented lesions. Our dataset covers eight key diagnostic categories in dermoscopy, providing a diverse range of lesions for artificial intelligence model training. Furthermore, a ninth out-of-distribution (OOD) class is also present on the test set, comprised of lesions which could not be distinctively classified as any of the others. By providing a comprehensive collection of varied images, BCN20000 helps bridge the gap between the training data for machine learning models and the day-to-day practice of medical practitioners. Additionally, we present a set of baseline classifiers based on state-of-the-art neural networks, which can be extended by other researchers for further experimentation.

Expert Agreement on the Presence and Spatial Localization of Melanocytic Features in Dermoscopy.

Dermoscopy aids in melanoma detection; however, agreement on dermoscopic features, including those of high clinical relevance, remains poor. In this study, we attempted to evaluate agreement among experts on exemplar images not only for the presence of melanocytic-specific features but also for spatial localization. This was a cross-sectional, multicenter, observational study. Dermoscopy images exhibiting at least 1 of 31 melanocytic-specific features were submitted by 25 world experts as exemplars. Using a web-based platform that allows for image markup of specific contrast-defined regions (superpixels), 20 expert readers annotated 248 dermoscopic images in collections of 62 images. Each collection was reviewed by five independent readers. A total of 4,507 feature observations were performed. Good-to-excellent agreement was found for 14 of 31 features (45.2%), with eight achieving excellent agreement (Gwet's AC >0.75) and seven of them being melanoma-specific features. These features were peppering/granularity (0.91), shiny white streaks (0.89), typical pigment network (0.83), blotch irregular (0.82), negative network (0.81), irregular globules (0.78), dotted vessels (0.77), and blue-whitish veil (0.76). By utilizing an exemplar dataset, a good-to-excellent agreement was found for 14 features that have previously been shown useful in discriminating nevi from melanoma. All images are public (www.isic-archive.com) and can be used for education, scientific communication, and machine learning experiments.

Monitoring multidimensional aspects of quality of life after cancer immunotherapy: protocol for the international multicentre, observational QUALITOP cohort study.

INTRODUCTION: Immunotherapies, such as immune checkpoint inhibitors and chimeric antigen receptor T-cell therapy, have significantly improved the clinical outcomes of various malignancies. However, they also cause immune-related adverse events (irAEs) that can be challenging to predict, prevent and treat. Although they likely interact with health-related quality of life (HRQoL), most existing evidence on this topic has come from clinical trials with eligibility criteria that may not accurately reflect real-world settings. The QUALITOP project will study HRQoL in relation to irAEs and its determinants in a real-world study of patients treated with immunotherapy. METHODS AND ANALYSIS: This international, observational, multicentre study takes place in France, the Netherlands, Portugal and Spain. We aim to include about 1800 adult patients with cancer treated with immunotherapy in a specifically recruited prospective cohort, and to additionally obtain data from historical real-world databases (ie, databiobanks) and medical administrative registries (ie, national cancer registries) in which relevant data regarding other adult patients with cancer treated with immunotherapy has already been stored. In the prospective cohort, clinical health status, HRQoL and psychosocial well-being will be monitored until 18 months after treatment initiation through questionnaires (at baseline and 3, 6, 12 and 18 months thereafter), and by data extraction from electronic patient files. Using advanced statistical methods, including causal inference methods, artificial intelligence algorithms and simulation modelling, we will use data from the QUALITOP cohort to improve the understanding of the complex relationships among treatment regimens, patient characteristics, irAEs and HRQoL. ETHICS AND DISSEMINATION: All aspects of the QUALITOP project will be conducted in accordance with the Declaration of Helsinki and with ethical approval from a suitable local ethics committee, and all patients will provide signed informed consent. In addition to standard dissemination efforts in the scientific literature, the data and outcomes will contribute to a smart digital platform and medical data lake. These will (1) help increase knowledge about the impact of immunotherapy, (2) facilitate improved interactions between patients, clinicians and the general population and (3) contribute to personalised medicine. TRIAL REGISTRATION NUMBER: NCT05626764.

Validation of artificial intelligence prediction models for skin cancer diagnosis using dermoscopy images: the 2019 International Skin Imaging Collaboration Grand Challenge.

BACKGROUND: Previous studies of artificial intelligence (AI) applied to dermatology have shown AI to have higher diagnostic classification accuracy than expert dermatologists; however, these studies did not adequately assess clinically realistic scenarios, such as how AI systems behave when presented with images of disease categories that are not included in the training dataset or images drawn from statistical distributions with significant shifts from training distributions. We aimed to simulate these real-world scenarios and evaluate the effects of image source institution, diagnoses outside of the training set, and other image artifacts on classification accuracy, with the goal of informing clinicians and regulatory agencies about safety and real-world accuracy. METHODS: We designed a large dermoscopic image classification challenge to quantify the performance of machine learning algorithms for the task of skin cancer classification from dermoscopic images, and how this performance is affected by shifts in statistical distributions of data, disease categories not represented in training datasets, and imaging or lesion artifacts. Factors that might be beneficial to performance, such as clinical metadata and external training data collected by challenge participants, were also evaluated. 25 331 training images collected from two datasets (in Vienna [HAM10000] and Barcelona [BCN20000]) between Jan 1, 2000, and Dec 31, 2018, across eight skin diseases, were provided to challenge participants to design appropriate algorithms. The trained algorithms were then tested for balanced accuracy against the HAM10000 and BCN20000 test datasets and data from countries not included in the training dataset (Turkey, New Zealand, Sweden, and Argentina). Test datasets contained images of all diagnostic categories available in training plus other diagnoses not included in training data (not trained category). We compared the performance of the algorithms against that of 18 dermatologists in a simulated setting that reflected intended clinical use. FINDINGS: 64 teams submitted 129 state-of-the-art algorithm predictions on a test set of 8238 images. The best performing algorithm achieved 58·8% balanced accuracy on the BCN20000 data, which was designed to better reflect realistic clinical scenarios, compared with 82·0% balanced accuracy on HAM10000, which was used in a previously published benchmark. Shifted statistical distributions and disease categories not included in training data contributed to decreases in accuracy. Image artifacts, including hair, pen markings, ulceration, and imaging source institution, decreased accuracy in a complex manner that varied based on the underlying diagnosis. When comparing algorithms to expert dermatologists (2460 ratings on 1269 images), algorithms performed better than experts in most categories, except for actinic keratoses (similar accuracy on average) and images from categories not included in training data (26% correct for experts vs 6% correct for algorithms, p<0·0001). For the top 25 submitted algorithms, 47·1% of the images from categories not included in training data were misclassified as malignant diagnoses, which would lead to a substantial number of unnecessary biopsies if current state-of-the-art AI technologies were clinically deployed. INTERPRETATION: We have identified specific deficiencies and safety issues in AI diagnostic systems for skin cancer that should be addressed in future diagnostic evaluation protocols to improve safety and reliability in clinical practice. FUNDING: Melanoma Research Alliance and La Marató de TV3.

Checklist for Evaluation of Image-Based Artificial Intelligence Reports in Dermatology: CLEAR Derm Consensus Guidelines From the International Skin Imaging Collaboration Artificial Intelligence Working Group.

IMPORTANCE: The use of artificial intelligence (AI) is accelerating in all aspects of medicine and has the potential to transform clinical care and dermatology workflows. However, to develop image-based algorithms for dermatology applications, comprehensive criteria establishing development and performance evaluation standards are required to ensure product fairness, reliability, and safety. OBJECTIVE: To consolidate limited existing literature with expert opinion to guide developers and reviewers of dermatology AI. EVIDENCE REVIEW: In this consensus statement, the 19 members of the International Skin Imaging Collaboration AI working group volunteered to provide a consensus statement. A systematic PubMed search was performed of English-language articles published between December 1, 2008, and August 24, 2021, for "artificial intelligence" and "reporting guidelines," as well as other pertinent studies identified by the expert panel. Factors that were viewed as critical to AI development and performance evaluation were included and underwent 2 rounds of electronic discussion to achieve consensus. FINDINGS: A checklist of items was developed that outlines best practices of image-based AI development and assessment in dermatology. CONCLUSIONS AND RELEVANCE: Clinically effective AI needs to be fair, reliable, and safe; this checklist of best practices will help both developers and reviewers achieve this goal.

Influence of germline genetic variants on dermoscopic features of melanoma.

Nevus count is highly determined by inherited variants and has been associated with the origin of melanoma. De novo melanomas (DNMMs) are more prevalent in patients with a low nevus count and have distinctive dermoscopic features than nevus-associated melanomas. We evaluated the impact of nine single nucleotide polymorphisms (SNPs) of MTAP (rs10811629, rs2218220, rs7023329 and rs751173), PLA2G6 (rs132985 and rs2284063), IRF4 (rs12203592), and PAX3 (rs10180903 and rs7600206) genes associated with nevus count and melanoma susceptibility, and the MC1R variants on dermoscopic features of 371 melanomas from 310 patients. All MTAP variants associated with a low nevus count were associated with regression structures (peppering and mixed regression), blue-whitish veil, shiny white structures, and pigment network. SNPs of PLA2G6 (rs132985), PAX3 (rs7600206), and IRF4 (rs12203592) genes were also associated with either shiny white structures or mixed regression (all corrected p-values ≤ .06). Melanomas from red hair color MC1R variants carriers showed lower total dermoscopy score (p-value = .015) and less blotches than melanomas from non-carriers (p-value = .048). Our results provide evidence that germline variants protective for melanoma risk and/or associated with a low nevus count are associated with certain dermoscopic features, more characteristic of de novo and worse prognosis melanomas.

Deep learning automated pathology in ex vivo microscopy.

Standard histopathology is currently the gold standard for assessment of margin status in Mohs surgical removal of skin cancer. Ex vivo confocal microscopy (XVM) is potentially faster, less costly and inherently 3D/digital compared to standard histopathology. Despite these advantages, XVM use is not widespread due, in part, to the need for pathologists to retrain to interpret XVM images. We developed artificial intelligence (AI)-driven XVM pathology by implementing algorithms that render intuitive XVM pathology images identical to standard histopathology and produce automated tumor positivity maps. XVM images have fluorescence labeling of cellular and nuclear biology on the background of endogenous (unstained) reflectance contrast as a grounding counter-contrast. XVM images of 26 surgical excision specimens discarded after Mohs micrographic surgery were used to develop an XVM data pipeline with 4 stages: flattening, colorizing, enhancement and automated diagnosis. The first two stages were novel, deterministic image processing algorithms, and the second two were AI algorithms. Diagnostic sensitivity and specificity were calculated for basal cell carcinoma detection as proof of principal for the XVM image processing pipeline. The resulting diagnostic readouts mimicked the appearance of histopathology and found tumor positivity that required first collapsing the confocal stack to a 2D image optimized for cellular fluorescence contrast, then a dark field-to-bright field colorizing transformation, then either an AI image transformation for visual inspection or an AI diagnostic binary image segmentation of tumor obtaining a diagnostic sensitivity and specificity of 88% and 91% respectively. These results show that video-assisted micrographic XVM pathology could feasibly aid margin status determination in micrographic surgery of skin cancer.

A patient-centric dataset of images and metadata for identifying melanomas using clinical context.

Prior skin image datasets have not addressed patient-level information obtained from multiple skin lesions from the same patient. Though artificial intelligence classification algorithms have achieved expert-level performance in controlled studies examining single images, in practice dermatologists base their judgment holistically from multiple lesions on the same patient. The 2020 SIIM-ISIC Melanoma Classification challenge dataset described herein was constructed to address this discrepancy between prior challenges and clinical practice, providing for each image in the dataset an identifier allowing lesions from the same patient to be mapped to one another. This patient-level contextual information is frequently used by clinicians to diagnose melanoma and is especially useful in ruling out false positives in patients with many atypical nevi. The dataset represents 2,056 patients (20.8% with at least one melanoma, 79.2% with zero melanomas) from three continents with an average of 16 lesions per patient, consisting of 33,126 dermoscopic images and 584 (1.8%) histopathologically confirmed melanomas compared with benign melanoma mimickers.

Skin cancer classification via convolutional neural networks: systematic review of studies involving human experts.

BACKGROUND: Multiple studies have compared the performance of artificial intelligence (AI)-based models for automated skin cancer classification to human experts, thus setting the cornerstone for a successful translation of AI-based tools into clinicopathological practice. OBJECTIVE: The objective of the study was to systematically analyse the current state of research on reader studies involving melanoma and to assess their potential clinical relevance by evaluating three main aspects: test set characteristics (holdout/out-of-distribution data set, composition), test setting (experimental/clinical, inclusion of metadata) and representativeness of participating clinicians. METHODS: PubMed, Medline and ScienceDirect were screened for peer-reviewed studies published between 2017 and 2021 and dealing with AI-based skin cancer classification involving melanoma. The search terms skin cancer classification, deep learning, convolutional neural network (CNN), melanoma (detection), digital biomarkers, histopathology and whole slide imaging were combined. Based on the search results, only studies that considered direct comparison of AI results with clinicians and had a diagnostic classification as their main objective were included. RESULTS: A total of 19 reader studies fulfilled the inclusion criteria. Of these, 11 CNN-based approaches addressed the classification of dermoscopic images; 6 concentrated on the classification of clinical images, whereas 2 dermatopathological studies utilised digitised histopathological whole slide images. CONCLUSIONS: All 19 included studies demonstrated superior or at least equivalent performance of CNN-based classifiers compared with clinicians. However, almost all studies were conducted in highly artificial settings based exclusively on single images of the suspicious lesions. Moreover, test sets mainly consisted of holdout images and did not represent the full range of patient populations and melanoma subtypes encountered in clinical practice.