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.

ORF1p Is a Potential Novel Diagnostic Marker for Differentiated Vulvar Intraepithelial Neoplasia.

Accurate diagnosis of differentiated vulvar intraepithelial neoplasia (dVIN) can be challenging as histomorphologic features may be subtle and overlap with nondysplastic lesions. In practice, aberrant p53 expression supports the diagnosis, but a substantial percentage retains wild-type p53. Recently, the retrotransposon long interspersed nuclear element 1 has been detected in distinct cancer types. We have now investigated the expression of the long interspersed nuclear element 1 encoded protein ORF1p in dysplastic and nondysplastic vulvar samples to assess its diagnostic value. Specimens of dVIN (n=29), high-grade squamous intraepithelial lesions (n=26), inflammatory vulvar lesions (n=20), lichen sclerosus (n=22), and normal vulvar epithelia (n=29) were included. ORF1p and p53 expression was determined using immunohistochemistry. The majority of dVIN [27/29 (93%)] and high-grade squamous intraepithelial lesions [20/26 (77%)] showed distinct (i.e. moderate or strong) ORF1p expression in the basal and suprabasal or all epithelial layers, respectively. Of note, ORF1p was present in all 4 cases of dVIN with wild-type p53 staining pattern. In contrast, ORF1p was negative or weakly expressed in most inflammatory lesions [14/20 (70%)] and lichen sclerosus [18/22 (82%), P <0.001]. Normal control epithelium exhibited negative staining in all cases. In conclusion, ORF1p might be a useful diagnostic marker for dVIN, especially in cases with retained wild-type p53.

Integrated diagnosis of nevi with severely atypical features and impact of second opinions.

BACKGROUND: A subset of melanocytic proliferations is difficult to classify by dermatopathology alone and their management is challenging. OBJECTIVE: To explore the value of correlation with dermatoscopy and to evaluate the utility of second opinions by additional pathologists. METHODS: For this single center retrospective study we collected 122 lesions that were diagnosed as atypical melanocytic proliferations, we reviewed dermatoscopy and asked two experienced pathologists to reassess the slides independently. RESULTS: For the binary decision of nevus versus melanoma the diagnostic consensus among external pathologists was only moderate (kappa 0.43; 95% CI 0.25-0.61). If ground truth were defined such that both pathologists had to agree on the diagnosis of melanoma, 13.1% of cases would have been diagnosed as melanoma. If one pathologist were sufficient to call it melanoma 29.5% of cases would have been diagnosed as melanoma. In either case, the presence of dermatoscopic white lines was associated with the diagnosis of melanoma. In lesions with peripheral dots and clods, melanoma was not jointly diagnosed by the two pathologists if the patient was younger than 45 years. CONCLUSIONS: A considerable number of atypical melanocytic proliferations may be diagnosed as melanoma if revised by other pathologists. The presence of white lines on dermatoscopy increases the likelihood of revision towards melanoma. Peripheral clods indicate growth but are not a melanoma clue if patients are younger than 45 years.

Development of a localization-based algorithm for the prediction of leg ulcer etiology.

BACKGROUND: Diagnostic work-up of leg ulcers is time- and cost-intensive. This study aimed at evaluating ulcer location as a diagnostic criterium and providing a diagnostic algorithm to facilitate differential diagnosis. PATIENTS AND METHODS: The study consisted of 277 patients with lower leg ulcers. The following five groups were defined: Venous leg ulcer, arterial ulcers, mixed ulcer, arteriolosclerosis, and vasculitis. Using computational surface rendering, predilection sites of different ulcer types were evaluated. The results were integrated in a multinomial logistic regression model to calculate the likelihood of a specific diagnosis depending on location, age, bilateral involvement, and ulcer count. Additionally, neural network image analysis was performed. RESULTS: The majority of venous ulcers extended to the medial malleolar region. Arterial ulcers were most frequently located on the dorsal aspect of the forefoot. Arteriolosclerotic ulcers were distinctly localized at the middle third of the lower leg. Vasculitic ulcers appeared to be randomly distributed and were markedly smaller, multilocular and bilateral. The multinomial logistic regression model showed an overall satisfactory performance with an estimated accuracy of 0.68 on unseen data. CONCLUSIONS: The presented algorithm based on ulcer location may serve as a basic tool to narrow down potential diagnoses and guide further diagnostic work-up.

Reproduction of patterns in melanocytic proliferations by agent-based simulation and geometric modeling.

Spatio-temporal patterns of melanocytic proliferations observed in vivo are important for diagnosis but the mechanisms that produce them are poorly understood. Here we present an agent-based model for simulating the emergence of the main biologic patterns found in melanocytic proliferations. Our model portrays the extracellular matrix of the dermo-epidermal junction as a two-dimensional manifold and we simulate cellular migration in terms of geometric translations driven by adhesive, repulsive and random forces. Abstracted cellular functions and melanocyte-matrix interactions are modeled as stochastic events. For identification and validation we use visual renderings of simulated cell populations in a horizontal perspective that reproduce growth patterns observed in vivo by sequential dermatoscopy and corresponding vertical views that reproduce the arrangement of melanocytes observed in histopathologic sections. Our results show that a balanced interplay of proliferation and migration produces the typical reticular pattern of nevi, whereas the globular pattern involves additional cellular mechanisms. We further demonstrate that slight variations in the three basic cellular properties proliferation, migration, and adhesion are sufficient to produce a large variety of morphological appearances of nevi. We anticipate our model to be a starting point for the reproduction of more complex scenarios that will help to establish functional connections between abstracted microscopic behavior and macroscopic patterns in all types of melanocytic proliferations including melanoma.

Artificial neural networks and pathologists recognize basal cell carcinomas based on different histological patterns.

Recent advances in artificial intelligence, particularly in the field of deep learning, have enabled researchers to create compelling algorithms for medical image analysis. Histological slides of basal cell carcinomas (BCCs), the most frequent skin tumor, are accessed by pathologists on a daily basis and are therefore well suited for automated prescreening by neural networks for the identification of cancerous regions and swift tumor classification.In this proof-of-concept study, we implemented an accurate and intuitively interpretable artificial neural network (ANN) for the detection of BCCs in histological whole-slide images (WSIs). Furthermore, we identified and compared differences in the diagnostic histological features and recognition patterns relevant for machine learning algorithms vs. expert pathologists.An attention-ANN was trained with WSIs of BCCs to identify tumor regions (n = 820). The diagnosis-relevant regions used by the ANN were compared to regions of interest for pathologists, detected by eye-tracking techniques.This ANN accurately identified BCC tumor regions on images of histologic slides (area under the ROC curve: 0.993, 95% CI: 0.990-0.995; sensitivity: 0.965, 95% CI: 0.951-0.979; specificity: 0.910, 95% CI: 0.859-0.960). The ANN implicitly calculated a weight matrix, indicating the regions of a histological image that are important for the prediction of the network. Interestingly, compared to pathologists' eye-tracking results, machine learning algorithms rely on significantly different recognition patterns for tumor identification (p < 10-4).To conclude, we found on the example of BCC WSIs, that histopathological images can be efficiently and interpretably analyzed by state-of-the-art machine learning techniques. Neural networks and machine learning algorithms can potentially enhance diagnostic precision in digital pathology and uncover hitherto unused classification patterns.

The dermoscopic inverse approach significantly improves the accuracy of human readers for lentigo maligna diagnosis.

BACKGROUND: A recently introduced dermoscopic method for the diagnosis of early lentigo maligna (LM) is based on the absence of prevalent patterns of pigmented actinic keratosis and solar lentigo/flat seborrheic keratosis. We term this the inverse approach. OBJECTIVE: To determine whether training on the inverse approach increases the diagnostic accuracy of readers compared to classic pattern analysis. METHODS: We used clinical and dermoscopic images of histopathologically diagnosed LMs, pigmented actinic keratoses, and solar lentigo/flat seborrheic keratoses. Participants in a dermoscopy masterclass classified the lesions at baseline and after training on pattern analysis and the inverse approach. We compared their diagnostic performance among the 3 timepoints and to that of a trained convolutional neural network. RESULTS: The mean sensitivity for LM without training was 51.5%; after training on pattern analysis, it increased to 56.7%; and after learning the inverse approach, it increased to 83.6%. The mean proportions of correct answers at the 3 timepoints were 62.1%, 65.5, and 78.5%. The percentages of readers outperforming the convolutional neural network were 6.4%, 15.4%, and 53.9%, respectively. LIMITATIONS: The experimental setting and the inclusion of histopathologically diagnosed lesions only. CONCLUSIONS: The inverse approach, added to the classic pattern analysis, significantly improves the sensitivity of human readers for early LM diagnosis.

Differentiating Arteriolosclerotic Ulcers of Martorell from Other Types of Leg Ulcers Based on Vascular Histomorphology.

Clinical differential diagnosis of arteriolosclerotic ulcers of Martorell is challenging due to the lack of clearly affirmative instrument-based diagnostic criteria. The aim of this study was to develop vascular histomorphological diagnostic criteria differentiating Martorell ulcers from other types of leg ulcers. The histomorphology of patients diagnosed with arteriolosclerotic ulcers of Martorell (n = 67) was compared with that of patients with venous leg ulcers, necrotizing leukocytoclastic vasculitis, pyoderma gangrenosum, and non-ulcerative controls (n = 15 each). In a multivariable logistic regression model, the rates of arteriolar calcification (odds ratio (OR) 42.71, 95% confidence interval (CI) 7.43-443.96, p < 0.001) and subendothelial hyalinosis (OR 29.28, 95% CI 4.88-278.21, p <0.001) were significantly higher in arteriolosclerotic ulcers of Martorell. Arteriolar cellularity was significantly lower in Martorell ulcers than in controls (OR 0.003, 95 CI < 0.001-0.97, p = 0.05). However, the wall-to-lumen ratio was similar in all ulcers (OR 0.975, 95% CI 0.598-2.04, p =0.929). Based on the Youden index, a wall cellularity of < 0.24 cells/100 µm2 was determined as the optimum cut-off point (sensitivity 0.955, specificity 0.944). Thus, arteriolar calcification, subendothelial hyalinosis, and arteriolar cellularity revealed high discriminatory power for arteriolosclerotic ulcers of Martorell.

Number needed to biopsy ratio and diagnostic accuracy for melanoma detection.

BACKGROUND: The number needed to biopsy (NNB) ratio for melanoma diagnosis is calculated by dividing the total number of biopsies by the number of biopsied melanomas. It is the inverse of positive predictive value (PPV), which is calculated by dividing the number of biopsied melanomas by the total number of biopsies. NNB is increasingly used as a metric to compare the diagnostic accuracy of health care practitioners. OBJECTIVE: To investigate the association of NNB with the standard statistical measures of sensitivity and specificity. METHODS: We extracted published diagnostic accuracy data from 5 cross-sectional skin cancer reader studies (median [min-max] readers/study was 29 [8-511]). Because NNB is a ratio, we converted it to PPV. RESULTS: Four studies showed no association and 1 showed a negative association between PPV and sensitivity. All 5 studies showed a positive association between PPV and specificity. LIMITATIONS: Reader study data. CONCLUSIONS: An individual health care practitioner with a lower NNB is likely to have a higher specificity than one with a higher NNB, assuming they practice under similar conditions; no conclusions can be made about their relative sensitivities. We advocate for additional research to define quality metrics for melanoma detection and caution when interpreting NNB.

Using content-based image retrieval of dermoscopic images for interpretation and education: A pilot study.

BACKGROUND: Dermoscopic content-based image retrieval (CBIR) systems provide a set of visually similar dermoscopic (magnified and illuminated) skin images with a pathology-confirmed diagnosis for a given dermoscopic query image of a skin lesion. Although recent advances in machine learning have spurred novel CBIR algorithms, we have few insights into how end users interact with CBIRs and to what extent CBIRs can be useful for education and image interpretation. MATERIALS AND METHODS: We developed an interactive user interface for a CBIR system with dermoscopic images as a decision support tool and investigated users' interactions and decisions with the system. We performed a pilot experiment with 14 non-medically trained users for a given set of annotated dermoscopic images. RESULTS: Our pilot showed that the number of correct classifications and users' confidence levels significantly increased with the CBIR interface compared with a non-CBIR interface, although the timing also increased significantly. The users found the CBIR interface of high educational value, engaging and easy to use. CONCLUSION: Overall, users became more accurate, found the CBIR approach provided a useful decision aid, and had educational value for learning about skin conditions.

Analysis of Collective Human Intelligence for Diagnosis of Pigmented Skin Lesions Harnessed by Gamification Via a Web-Based Training Platform: Simulation Reader Study.

BACKGROUND: The diagnosis of pigmented skin lesion is error prone and requires domain-specific expertise, which is not readily available in many parts of the world. Collective intelligence could potentially decrease the error rates of nonexperts. OBJECTIVE: The aim of this study was to evaluate the feasibility and impact of collective intelligence for the detection of skin cancer. METHODS: We created a gamified study platform on a stack of established Web technologies and presented 4216 dermatoscopic images of the most common benign and malignant pigmented skin lesions to 1245 human raters with different levels of experience. Raters were recruited via scientific meetings, mailing lists, and social media posts. Education was self-declared, and domain-specific experience was tested by screening tests. In the target test, the readers had to assign 30 dermatoscopic images to 1 of the 7 disease categories. The readers could repeat the test with different lesions at their own discretion. Collective human intelligence was achieved by sampling answers from multiple readers. The disease category with most votes was regarded as the collective vote per image. RESULTS: We collected 111,019 single ratings, with a mean of 25.2 (SD 18.5) ratings per image. As single raters, nonexperts achieved a lower mean accuracy (58.6%) than experts (68.4%; mean difference=-9.4%; 95% CI -10.74% to -8.1%; P<.001). Collectives of nonexperts achieved higher accuracies than single raters, and the improvement increased with the size of the collective. A collective of 4 nonexperts surpassed single nonexperts in accuracy by 6.3% (95% CI 6.1% to 6.6%; P<.001). The accuracy of a collective of 8 nonexperts was 9.7% higher (95% CI 9.5% to 10.29%; P<.001) than that of single nonexperts, an improvement similar to single experts (P=.73). The sensitivity for malignant images increased for nonexperts (66.3% to 77.6%) and experts (64.6% to 79.4%) for answers given faster than the intrarater mean. CONCLUSIONS: A high number of raters can be attracted by elements of gamification and Web-based marketing via mailing lists and social media. Nonexperts increase their accuracy to expert level when acting as a collective, and faster answers correspond to higher accuracy. This information could be useful in a teledermatology setting.

Comparison of the accuracy of human readers versus machine-learning algorithms for pigmented skin lesion classification: an open, web-based, international, diagnostic study.

BACKGROUND: Whether machine-learning algorithms can diagnose all pigmented skin lesions as accurately as human experts is unclear. The aim of this study was to compare the diagnostic accuracy of state-of-the-art machine-learning algorithms with human readers for all clinically relevant types of benign and malignant pigmented skin lesions. METHODS: For this open, web-based, international, diagnostic study, human readers were asked to diagnose dermatoscopic images selected randomly in 30-image batches from a test set of 1511 images. The diagnoses from human readers were compared with those of 139 algorithms created by 77 machine-learning labs, who participated in the International Skin Imaging Collaboration 2018 challenge and received a training set of 10 015 images in advance. The ground truth of each lesion fell into one of seven predefined disease categories: intraepithelial carcinoma including actinic keratoses and Bowen's disease; basal cell carcinoma; benign keratinocytic lesions including solar lentigo, seborrheic keratosis and lichen planus-like keratosis; dermatofibroma; melanoma; melanocytic nevus; and vascular lesions. The two main outcomes were the differences in the number of correct specific diagnoses per batch between all human readers and the top three algorithms, and between human experts and the top three algorithms. FINDINGS: Between Aug 4, 2018, and Sept 30, 2018, 511 human readers from 63 countries had at least one attempt in the reader study. 283 (55·4%) of 511 human readers were board-certified dermatologists, 118 (23·1%) were dermatology residents, and 83 (16·2%) were general practitioners. When comparing all human readers with all machine-learning algorithms, the algorithms achieved a mean of 2·01 (95% CI 1·97 to 2·04; p<0·0001) more correct diagnoses (17·91 [SD 3·42] vs 19·92 [4·27]). 27 human experts with more than 10 years of experience achieved a mean of 18·78 (SD 3·15) correct answers, compared with 25·43 (1·95) correct answers for the top three machine algorithms (mean difference 6·65, 95% CI 6·06-7·25; p<0·0001). The difference between human experts and the top three algorithms was significantly lower for images in the test set that were collected from sources not included in the training set (human underperformance of 11·4%, 95% CI 9·9-12·9 vs 3·6%, 0·8-6·3; p<0·0001). INTERPRETATION: State-of-the-art machine-learning classifiers outperformed human experts in the diagnosis of pigmented skin lesions and should have a more important role in clinical practice. However, a possible limitation of these algorithms is their decreased performance for out-of-distribution images, which should be addressed in future research. FUNDING: None.

A prospective diagnostic study on povidone-iodine retention in lesions suspected to be squamous cell carcinoma or keratoacanthoma.

BACKGROUND AND OBJECTIVES: While dermatoscopy improves diagnostic accuracy for raised nonpigmented lesions, those with white surface keratin can be problematical. We investigated whether retention of povidone-iodine by surface keratin provides a clue to benignity. METHODS: We performed a retrospective pilot study (n = 57) followed by a prospective study (n = 117) on raised nonpigmented lesions with white surface keratin. An initial dermatoscopic image was taken of each lesion, povidone-iodine was applied and another image taken. Following lavage with 70% ethanol, a third image was acquired. The percentage surface area of residual povidone-iodine staining after lavage was recorded, and the results analysed. RESULTS: The optimal cut-off point of residual staining was 80%, where values of ≤80% pointed to malignancy. At this cut-off, the OR for lesions with values ≤80% to be truly malignant in the retrospective set was 4.03 (95% CI: 2.1-7.6) and the AUC was 0.7 (95% CI: 0.62-0.78). For the prospective set, the corresponding OR was 2.3 (95% CI: 1.4-3.7) and the AUC was 0.62 (95% CI: 0.55-0.68). CONCLUSIONS: This study presents evidence that povidone-iodine retention may have a degree of efficacy in distinguishing benign from malignant keratotic lesions. Further study is warranted.

Multimodal skin lesion classification using deep learning.

While convolutional neural networks (CNNs) have successfully been applied for skin lesion classification, previous studies have generally considered only a single clinical/macroscopic image and output a binary decision. In this work, we have presented a method which combines multiple imaging modalities together with patient metadata to improve the performance of automated skin lesion diagnosis. We evaluated our method on a binary classification task for comparison with previous studies as well as a five class classification task representative of a real-world clinical scenario. We showed that our multimodal classifier outperforms a baseline classifier that only uses a single macroscopic image in both binary melanoma detection (AUC 0.866 vs 0.784) and in multiclass classification (mAP 0.729 vs 0.598). In addition, we have quantitatively showed the automated diagnosis of skin lesions using dermatoscopic images obtains a higher performance when compared to using macroscopic images. We performed experiments on a new data set of 2917 cases where each case contains a dermatoscopic image, macroscopic image and patient metadata.

Dermatoscopy of Neoplastic Skin Lesions: Recent Advances, Updates, and Revisions.

OPINION STATEMENT: Dermatoscopy (dermoscopy) improves the diagnosis of benign and malignant cutaneous neoplasms in comparison with examination with the unaided eye and should be used routinely for all pigmented and non-pigmented cutaneous neoplasms. It is especially useful for the early stage of melanoma when melanoma-specific criteria are invisible to the unaided eye. Preselection by the unaided eye is therefore not recommended. The increased availability of polarized dermatoscopes, and the extended use of dermatoscopy in non-pigmented lesions led to the discovery of new criteria, and we recommend that lesions should be examined with polarized and non-polarized dermatoscopy. The "chaos and clues algorithm" is a good starting point for beginners because it is easy to use, accurate, and it works for all types of pigmented lesions not only for those melanocytic. Physicians, who use dermatoscopy routinely, should be aware of new clues for acral melanomas, nail matrix melanomas, melanoma in situ, and nodular melanoma. Dermatoscopy should also be used to distinguish between different subtypes of basal cell carcinoma and to discriminate highly from poorly differentiated squamous cell carcinomas to optimize therapy and management of non-melanoma skin cancer. One of the most exciting areas of research is the use of dermatoscopic images for machine learning and automated diagnosis. Convolutional neural networks trained with dermatoscopic images are able to diagnose pigmented lesions with the same accuracy as human experts. We humans should not be afraid of this new and exciting development because it will most likely lead to a peaceful and fruitful coexistence of human experts and decision support systems.

Accuracy of dermatoscopy for the diagnosis of nonpigmented cancers of the skin.

BACKGROUND: Nonpigmented skin cancer is common, and diagnosis with the unaided eye is error prone. OBJECTIVE: To investigate whether dermatoscopy improves the diagnostic accuracy for nonpigmented (amelanotic) cutaneous neoplasms. METHODS: We collected a sample of 2072 benign and malignant neoplastic lesions and inflammatory conditions and presented close-up images taken with and without dermatoscopy to 95 examiners with different levels of experience. RESULTS: The area under the curve was significantly higher with than without dermatoscopy (0.68 vs 0.64, P < .001). Among 51 possible diagnoses, the correct diagnosis was selected in 33.1% of cases with and 26.4% of cases without dermatoscopy (P < .001). For experts, the frequencies of correct specific diagnoses of a malignant lesion improved from 40.2% without to 51.3% with dermatoscopy. For all malignant neoplasms combined, the frequencies of appropriate management strategies increased from 78.1% without to 82.5% with dermatoscopy. LIMITATIONS: The study deviated from a real-life clinical setting and was potentially affected by verification and selection bias. CONCLUSIONS: Dermatoscopy improves the diagnosis and management of nonpigmented skin cancer and should be used as an adjunct to examination with the unaided eye.