A protocol for annotation of total body photography for machine learning to analyze skin phenotype and lesion classification.

Introduction: Artificial Intelligence (AI) has proven effective in classifying skin cancers using dermoscopy images. In experimental settings, algorithms have outperformed expert dermatologists in classifying melanoma and keratinocyte cancers. However, clinical application is limited when algorithms are presented with 'untrained' or out-of-distribution lesion categories, often misclassifying benign lesions as malignant, or misclassifying malignant lesions as benign. Another limitation often raised is the lack of clinical context (e.g., medical history) used as input for the AI decision process. The increasing use of Total Body Photography (TBP) in clinical examinations presents new opportunities for AI to perform holistic analysis of the whole patient, rather than a single lesion. Currently there is a lack of existing literature or standards for image annotation of TBP, or on preserving patient privacy during the machine learning process. Methods: This protocol describes the methods for the acquisition of patient data, including TBP, medical history, and genetic risk factors, to create a comprehensive dataset for machine learning. 500 patients of various risk profiles will be recruited from two clinical sites (Australia and Spain), to undergo temporal total body imaging, complete surveys on sun behaviors and medical history, and provide a DNA sample. This patient-level metadata is applied to image datasets using DICOM labels. Anonymization and masking methods are applied to preserve patient privacy. A two-step annotation process is followed to label skin images for lesion detection and classification using deep learning models. Skin phenotype characteristics are extracted from images, including innate and facultative skin color, nevi distribution, and UV damage. Several algorithms will be developed relating to skin lesion detection, segmentation and classification, 3D mapping, change detection, and risk profiling. Simultaneously, explainable AI (XAI) methods will be incorporated to foster clinician and patient trust. Additionally, a publicly released dataset of anonymized annotated TBP images will be released for an international challenge to advance the development of new algorithms using this type of data. Conclusion: The anticipated results from this protocol are validated AI-based tools to provide holistic risk assessment for individual lesions, and risk stratification of patients to assist clinicians in monitoring for skin cancer.

Reflectance confocal microscopy - Consensus terminology glossary in Brazilian Portuguese for normal skin, melanocytic and non-melanocytic lesions

Abstract Background: Currently, there is no uniform and official terminology in Portuguese for reflectance confocal microscopy analysis, despite the increasing number of Brazilian dermatologists using this new tool. Objective: To present the terminology in Brazilian Portuguese for the description of reflectance confocal microscopy and establish the first Brazilian consensus on terms related to normal skin and cutaneous tumors. Methods: 10 Brazilian specialists from different institutions and states of Brazil were selected to evaluate the best corresponding terms in Portuguese for normal skin, melanocytic and non-melanocytic tumors. The terms used were translated from international consensuses in the English language. The modified Delphi method was used to create the consensus in 3 steps. Results: The terms considered the most appropriate in the Portuguese language to describe the findings of normal skin, melanocytic and non-melanocytic lesions in the reflectance confocal microscopy analysis were presented. Study limitations: The limitations of the present study include the number of participants and limited regional representation (only two of the five Brazilian regions were represented). Conclusion: This Brazilian consensus represents an opportunity for dermatologists and physicians specializing in cutaneous oncology to become familiar with reflectance confocal microscopy, propagating the technique in clinical and research environments to stimulate national and international publications on this subject.

A Narrative Review: Opportunities and Challenges in Artificial Intelligence Skin Image Analyses Using Total Body Photography.

Artificial intelligence (AI) algorithms for skin lesion classification have reported accuracy at par with and even outperformance of expert dermatologists in experimental settings. However, the majority of algorithms do not represent real-world clinical approach where skin phenotype and clinical background information are considered. We review the current state of AI for skin lesion classification and present opportunities and challenges when applied to total body photography (TBP). AI in TBP analysis presents opportunities for intrapatient assessment of skin phenotype and holistic risk assessment by incorporating patient-level metadata, although challenges exist for protecting patient privacy in algorithm development and improving explainable AI methods.

Reflectance confocal microscopy - Consensus terminology glossary in Brazilian Portuguese for normal skin, melanocytic and non-melanocytic lesions.

BACKGROUND: Currently, there is no uniform and official terminology in Portuguese for reflectance confocal microscopy analysis, despite the increasing number of Brazilian dermatologists using this new tool. OBJECTIVE: To present the terminology in Brazilian Portuguese for the description of reflectance confocal microscopy and establish the first Brazilian consensus on terms related to normal skin and cutaneous tumors. METHODS: 10 Brazilian specialists from different institutions and states of Brazil were selected to evaluate the best corresponding terms in Portuguese for normal skin, melanocytic and non-melanocytic tumors. The terms used were translated from international consensuses in the English language. The modified Delphi method was used to create the consensus in 3 steps. RESULTS: The terms considered the most appropriate in the Portuguese language to describe the findings of normal skin, melanocytic and non-melanocytic lesions in the reflectance confocal microscopy analysis were presented. STUDY LIMITATIONS: The limitations of the present study include the number of participants and limited regional representation (only two of the five Brazilian regions were represented). CONCLUSION: This Brazilian consensus represents an opportunity for dermatologists and physicians specializing in cutaneous oncology to become familiar with reflectance confocal microscopy, propagating the technique in clinical and research environments to stimulate national and international publications on this subject.

Melanoma with peripheral globules: Clinical and dermatoscopic features.

BACKGROUND: The morphology and biology of nevi with peripheral globules are well known, whereas those of melanomas with peripheral globules remain unclear. OBJECTIVE: Comparing the dermatoscopic characteristics of nevi and melanomas with peripheral globules. METHODS: A total of 401 melanocytic lesions with peripheral globules were included in this retrospective study. Dermatoscopic patterns and structures, including those of peripheral globules, were evaluated. A generalized estimating equation model with a binomial distribution dependent variable and logit link function was fitted to the dataset to identify features with the highest odds of differentiating melanoma from nevi. RESULTS: Of the 401 lesions, 179 (44.64%) were excised, 41 (10.22%) of which were melanomas. Melanomas were most common in the lower extremities (P < .01), with a disorganized pattern, whereas melanocytic nevi were most common on the trunk, with an organized pattern. In addition, the presence of blotches, atypical dots and globules, or atypical vessels was associated with melanomas (P < .01). LIMITATIONS: The retrospective design of the study may have caused an inclusion bias. CONCLUSION: Melanocytic lesions displaying peripheral globules are at the greatest risk of melanoma if located on the lower extremity and if lesions reveal any of the following structures: blotch, atypical dots and globules, or atypical vessels.

Hotspot analysis by confocal microscopy can help to differentiate challenging melanocytic skin lesions.

Some melanocytic lesions do not present enough clinical and dermoscopic features to allow ruling out a possible melanoma diagnosis. These "doubtful melanocytic lesions" pose a very common and challenging scenario in clinical practice and were selected at this study for reflectance confocal microscopy evaluation and subsequent surgical excision for histopathological diagnosis. The study included 110 lesions and three confocal features were statistically able to distinguish benign melanocytic lesions from melanomas: "peripheral hotspot at dermo-epidermal junction", "nucleated roundish cells at the dermo-epidermal junction" and "sheet of cells". The finding of a peripheral hotspot (atypical cells in 1mm2) at the DEJ is highlighted because has not been previously reported in the literature as a confocal feature related to melanomas.

Pigmented lesion on the face: which is the chance of being melanoma using reflectance confocal microscopy features?

Facial melanoma presents itself as a brownish macula, being difficult to differentiate it from benign pigmented lesions of the face on clinical examination. Reflectance confocal microscopy (RCM) assists in diagnosing facial lesions in which dermoscopy has limitations, allowing to increase the diagnostic accuracy. The study aimed to analyze the RCM features of pigmented isolated lesions of the face for diagnosing melanoma. Also, we sought to establish the chance of a pigmented lesion on the face being a melanoma using RCM criteria. In this retrospective and prospective study, 105 clinical pigmented lesions on the face underwent RCM, and cytoarchitectural features in the epidermis, the dermo-epidermal junction (DEJ), and dermis were described. For statistical analysis, the exact chi-square test was applied to the RCM criteria. The odds ratio was estimated using univariate logistic regression. Finally, we used the multiple logistic regression method for creating a nomogram to predict the chance of a lesion being a melanoma. After univariate and multivariate logistic regression, atypical round nucleated cells within the epidermis, pagetoid spread, and follicular dendritic cells presented as statistically significant features. Then, a complex nomogram was created to give the chance of a pigmented lesion on the face being a melanoma. The presence of these three features resulted in a 98% probability for melanoma. This study allowed to estimate the diagnosis of melanoma on the face, using RCM, practicable and feasible in the daily routine, through the presence of some RCM nomogram criteria.

Cirurgia micrográfica no carcinoma espinocelular recidivante: relato de caso / Squamous cell carcinoma and micrographic surgery: a case report

O carcinoma espinocelular (CEC) recidivante necessita de abordagem adequada, visando à diminuição do seu risco de recorrência. Relata-se o caso de paciente com histórico de CEC na região frontal esquerda, apresentando placa infiltrada de limites imprecisos, cuja biópsia confirmou o diagnóstico de recidiva. Fez remoção cirúrgica e congelação intraoperatória convencional. Após dois meses, apresentou nova área suspeita, com biópsia evidenciando carcinoma pouco diferenciado infiltrativo, que foi tratado com cirurgia micrográfica (CM) sem novo acometimento. Ressaltamos a importância do uso de métodos de análise periférica das margens cirúrgicas na ocorrência de recidiva local ou em paciente com carcinoma de alto risco

The recurrent squamous cell carcinoma (SCC) needs an appropriate approach to decrease the risk of recurrences. We report the case of a man diagnosed with SCC on the left temporal region treated by conventional surgery, with infiltration and erythema in the scar area. Reflectance confocal microscopy suggested the diagnosis of recurrent SCC confirmed by histopathology. New surgery and traditional intraoperative frozen section were performed. After two months, the patient had a new relapse, finally treated with Micrographic surgery without new involvement. This article highlights the importance of micrographic surgery in treating SCC, aiming to decrease new recurrences.

Microscopia confocal de reflectância como ferramenta para avaliar os efeitos causados pelo microagulhamento: uma série de cinco casos / Reflectance confocal microscopy as a tool for assessing the injury caused by microneedling: a serie of five cases

O microagulhamento é um procedimento cirúrgico ambulatorial que pode ser utilizado para diferentes indicações com o objetivo de estimular a produção de colágeno. Foram avaliados 5 casos no transcorrer das 72 horas após o procedimento, por meio da Microscopia Confocal de Reflectância, com o objetivo de avaliar a vida útil dos orifícios.

Microneedling is an ambulatory surgical procedure that can be used for different indications with the objective of stimulating the production of collagen. Five cases were evaluated in the first 72 hours after the procedure by reflectance confocal microscopy in order to evaluate the pores lifetime.

Reflectance confocal microscopy as a tool for screening surgical margins of basal cell carcinoma.

Surgical excision of basal cell carcinoma with minimum margins requires serial assessment of layers by frozen histopathology in the case of Mohs micrographic surgery. Evaluation of presurgical tumor margins by in vivo reflectance confocal microscopy is a potential alternative. We selected 12 basal cell carcinoma lesions that were analyzed by confocal microscopy to define margins. The lesions were excised by Mohs surgery. Six tumors showed negative margins in the first phase of Mohs micrographic surgery. We concluded that reflectance confocal microscopy can be useful in the preoperative definition of basal cell carcinoma margins.

Reflectance confocal microscopy as a tool for screening surgical margins of basal cell carcinoma

Abstract: Surgical excision of basal cell carcinoma with minimum margins requires serial assessment of layers by frozen histopathology in the case of Mohs micrographic surgery. Evaluation of presurgical tumor margins by in vivo reflectance confocal microscopy is a potential alternative. We selected 12 basal cell carcinoma lesions that were analyzed by confocal microscopy to define margins. The lesions were excised by Mohs surgery. Six tumors showed negative margins in the first phase of Mohs micrographic surgery. We concluded that reflectance confocal microscopy can be useful in the preoperative definition of basal cell carcinoma margins.

A utilidade da microscopia confocal nas lesões de lentigo maligno / The utility of confocal microscopy in lentigo maligna lesions

O diagnóstico das lesões pigmentadas da face é considerado desafiador uma vez que lesões benignas e malignas podem compartilhar características clínicas e dermatoscópicas semelhantes, principalmente em lesões iniciais, sendo muitas vezes difícil de identificar as lesões de lentigo maligno da face. Assim, a microscopia confocal de reflectância pode se tornar uma ferramenta útil no diagnóstico dos lentigos malignos bem como para o seu planejamento cirúrgico.

The diagnosis of pigmented facial lesions is considered challenging since benign and malignant lesions might have similar clinical and dermoscopic features ­ especially in the early stages of the lesion ­ entailing that it is often difficult to identify lentigo maligna lesions in the face. In this way, confocal reflectance microscopy has the potential to become a useful tool both in the diagnosis and surgical planning of lentigo maligna.

Reflectance confocal microscopy features of BRAF V600E mutated thin melanomas detected by immunohistochemistry.

The classification of melanoma into four histological subtypes has been questioned regarding its clinical validity in providing relevant information for treatment for metastatic tumors. Specific genetic alterations are associated with particular clinical and histopathological features, suggesting that these could be helpful in refining existing melanoma classification schemes. We analyzed BRAF V600E mutated melanomas to explore the Reflectance confocal microscopy (RCM) utility as a screening aid in the evaluation of the most appropriate patients for genetic testing. Thus, 32 melanomas were assessed regarding their BRAF V600E mutational status. Experts blinded to dermoscopic images and V600E immunohistochemistry results evaluated RCM images regarding previously described melanoma features. BRAF positive melanomas were related to younger age (p = 0.035), invasive melanomas (p = 0.03) and to the presence of hiporreflective cells (p = 0.02), epidermal nests (p = 0.02), dermal-epidermal junction nests (p = 0.05), edged papillae (p = 0.05), and bright dots (p = 0.05), and to absence of junctional thickening due to isolated cells (p = 0.01) and meshwork (p = 0.02). This study can not characterize other mutations in the BRAF, because the immunohistochemistry is specific to the type V600E. The findings should encourage the genetic evaluation of BRAF mutation. This study highlights the potential of RCM as a supplementary tool in the screening of BRAF-mutated melanomas.

Opening a Window into Living Tissue: Histopathologic Features of Confocal Microscopic Findings in Skin Tumors.

The knowledge of histopathology and in vivo reflectance confocal microscopy correlation has several potential applications. Reflectance confocal microscopy can be performed in all skin tumors, and in this article, the most common histopathologic features of confocal microscopic findings in melanocytic skin tumors and nonmelanocytic skin tumors are described.

Characterization of individuals at high risk of developing melanoma in Latin America: bases for genetic counseling in melanoma.

PURPOSE: CDKN2A is the main high-risk melanoma-susceptibility gene, but it has been poorly assessed in Latin America. We sought to analyze CDKN2A and MC1R in patients from Latin America with familial and sporadic multiple primary melanoma (SMP) and compare the data with those for patients from Spain to establish bases for melanoma genetic counseling in Latin America. METHODS: CDKN2A and MC1R were sequenced in 186 Latin American patients from Argentina, Brazil, Chile, Mexico, and Uruguay, and in 904 Spanish patients. Clinical and phenotypic data were obtained. RESULTS: Overall, 24 and 14% of melanoma-prone families in Latin America and Spain, respectively, had mutations in CDKN2A. Latin American families had CDKN2A mutations more frequently (P = 0.014) than Spanish ones. Of patients with SMP, 10% of those from Latin America and 8.5% of those from Spain had mutations in CDKN2A (P = 0.623). The most recurrent CDKN2A mutations were c.-34G>T and p.G101W. Latin American patients had fairer hair (P = 0.016) and skin (P < 0.001) and a higher prevalence of MC1R variants (P = 0.003) compared with Spanish patients. CONCLUSION: The inclusion criteria for genetic counseling of melanoma in Latin America may be the same criteria used in Spain, as suggested in areas with low to medium incidence, SMP with at least two melanomas, or families with at least two cases among first- or second-degree relatives.Genet Med 18 7, 727-736.

Inner gray halo, a novel dermoscopic feature for the diagnosis of pigmented actinic keratosis: clues for the differential diagnosis with lentigo maligna.

BACKGROUND: Pigmented actinic keratosis (PAK) is a frequent simulator of lentigo maligna (LM) on the face upon clinical and dermoscopic examination, leading to misdiagnosis and unnecessary excisions. LM and PAK share dermoscopic features, making it difficult to have a confident diagnosis of PAK only with current dermoscopic knowledge. OBJECTIVE: We sought to evaluate sensitivity, specificity, and interobserver reproducibility of a novel dermoscopic feature, inner gray halo (IGH), and establish its histopathological and confocal correlations. METHODS: Dermoscopists blinded to histopathological diagnosis evaluated 58 PAK and 21 LM for the presence of IGH and dermoscopy parameters. Areas exhibiting IGH were marked and imaged with reflectance confocal microscopy before sampling for histopathologic correlation. Reflectance confocal microscopy and transverse histologic sectioning were performed in 14 of 79 cases. RESULTS: IGH was present in 53 of 58 (94.1%) PAK and in 5 of 21 (23.8%) LM in our series (sensitivity 91.4%; specificity 71.4%; positive predictive value 89.8%). Interobserver agreement was excellent (Kappa 0.846). Through transverse and perpendicular histologic sections, a dermoscopic-histologic-confocal correlation of IGH was established. LIMITATIONS: A larger test set is needed to further validate the use of IGH in the differential diagnosis of PAK and facial pigmented lesions. CONCLUSION: IGH is a novel dermoscopic parameter useful for the differentiation of PAK from LM on the face.