Early Detection Of Melanoma Research Articles

Melanoma arising from pre-existing naevus in carriers of a germline CDKN2A pathogenic variant.

In this study we used total body photography (TBP) to determine whether melanoma emerges from a preexisting naevi or normal-appearing skin in individuals with pathogenic CDKN2a variant. In total, 60 melanoma cases occurred in 44 CDKN2a germline carriers. In addition, we observed that 78% of the 60 included melanoma cases developed from preexisting naevi, significantly higher than the 30% reported for sporadic melanoma. This high incidence may be due to the numerous atypical naevi and early bi-allelic loss of CDKN2A. Our findings underscore the importance of routine TBP in these patients for early detection of melanoma. More than previously appreciated, the surgical removal of melanocytic naevi in these high-risk patients may prevent melanoma development.

Participant Motivators and Expectations in the MEL-SELF Randomized Clinical Trial of Patient-Led Surveillance for Recurrent Melanoma: Content Analysis of Survey Responses.

Limited data exist on the motivations and expectations of participants when enrolling in dermatology clinical trials, including melanoma early detection trials. Understanding participant motivators for research engagement has been identified as a prioritized area for trial methodology research. The study aimed to determine motivators of participation and expectations from trial involvement among patients enrolled in the MEL-SELF randomized clinical trial of patient-led surveillance for new or recurrent melanoma. The MEL-SELF trial is recruiting patients previously treated for localized melanoma, who own a smartphone, have a partner to assist with skin self-examination (SSE), and attend routinely scheduled follow-up at specialist and primary care skin clinics in Australia. We evaluated responses from the first 100 randomized participants to 2 open-ended questions about their motivations and expectations for participating in the trial, administered through the internet-based baseline questionnaire. A total of 3 coders independently coded the free-text responses and resolved discrepancies through consensus. Qualitative content analysis by an iterative process was used to group responses into themes. Responses from potential participants who were not randomized and the 404 participants randomized subsequently into the trial, were also checked for new themes. Coding and analysis were conducted in Microsoft Excel. Out of the 100 survey participants, 98 (98%) answered at least 1 of the 2 questions. Overall, responses across the motivation and expectation items indicated 3 broad themes: community benefit, perceived personal benefit, and trusting relationship with their health care provider. The most common motivators for participation were related to community benefit. These included progressing medical research, benefitting future melanoma patients who may have similar experiences, and broader altruistic sentiments such as "helping others" or "giving back." The most common expectations from the trial related to personal benefit. These included perceived improved outcomes such as earlier diagnosis and treatment, access to additional care, and increased self-empowerment to take actions themselves that benefit their health. Patients expressed a desire to gain health-related knowledge and skills and were interested in the potential advantages of teledermatology. There were no new themes in responses from those who were not randomized or were randomized subsequent to the first 100. We report a tailorable, patient-focused approach to identify drivers of research engagement in clinical research. Clinical trials offer an opportunity to collate a substantial evidence base on determinants of research participation and to identify context-specific factors. Results from the MEL-SELF trial emphasized notable altruism, self-empowerment, and perceived advantages of teledermatology as specific motivators. These findings informed consent processes, recruitment, retention, response to trial tasks, and intervention adherence for the MEL-SELF host trial. Australian New Zealand Clinical Trials Registry (ANZCTR): ACTRN12621000176864. https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=379527.

Hybrid Deep Learning Framework for Melanoma Diagnosis Using Dermoscopic Medical Images

Background: Melanoma, or skin cancer, is a dangerous form of cancer that is the major cause of the demise of thousands of people around the world. Methods: In recent years, deep learning has become more popular for analyzing and detecting these medical issues. In this paper, a hybrid deep learning approach has been proposed based on U-Net for image segmentation, Inception-ResNet-v2 for feature extraction, and the Vision Transformer model with a self-attention mechanism for refining the features for early and accurate diagnosis and classification of skin cancer. Furthermore, in the proposed approach, hyperparameter tuning helps to obtain more accurate and optimized results for image classification. Results: Dermoscopic shots gathered by the worldwide skin imaging collaboration (ISIC2020) challenge dataset are used in the proposed research work and achieved 98.65% accuracy, 99.20% sensitivity, and 98.03% specificity, which outperforms the other existing approaches for skin cancer classification. Furthermore, the HAM10000 dataset is used for ablation studies to compare and validate the performance of the proposed approach. Conclusions: The achieved outcome suggests that the proposed approach would be able to serve as a valuable tool for assisting dermatologists in the early detection of melanoma.

Three-Dimensional Total Body Photography, Digital Dermoscopy, and in vivo Reflectance Confocal Microscopy for Follow-Up Assessments of High-Risk Patients for Melanoma: A Prospective, Controlled Study

Plain Language SummaryThe combination of total body photography (TBP) and digital dermoscopy (DD) for the monitoring of patients with high risk for melanoma leads to early detection of melanoma. The aim of the present study was to examine if regular monitoring of patients with high risk for melanoma with 3D-TBP, DD, and confocal microscopy (RCM) leads to a benefit in comparison to a control group (CG). The intervention group (IG) was examined with 3D-TBP at every visit. If necessary, DD and/or RCM were used for diagnosis and/or monitoring of pigmented lesions. The CG was examined clinically with dermoscopy only. A total of 600 patients (324 male and 276 female) were followed up between April 21, 2021, and January 8, 2024. Lesions (n = 166) were monitored with DD and 105 lesions were examined with RCM. The number of lesions that the dermatologists had to cut out to find a melanoma was 2.83, when RCM was used. There were more melanomas (22 vs. 1, p = 0.022) and more nevi excised in the IG (186 vs. 10, p < 0.001), among them more dysplastic nevi (137 vs. 2, p < 0.001). Out of the melanomas diagnosed in the IG, three melanomas were diagnosed with RCM directly, nine melanomas with the combination of 3D-TBP and RCM, and 10 melanomas were diagnosed clinically with dermoscopy only. The follow-up with the combination of 3D-TBP, DD and RCM led to detection of more melanomas in comparison to a CG. RCM led to a reduced NNT for melanocytic lesions.

Creating an effective curriculum for cosmetology students to promote early detection of melanoma in salons and barber shops: a scoping review

This scoping review aims to summarize available research literature on previous skin cancer educational programs among hair professionals or cosmetology students to inform the creation of an effective cosmetology school curriculum that will promote early detection of melanoma on the scalp and neck. Cutaneous melanoma on the scalp and neck is a significant cause of morbidity and mortality in the United States, and the number of new cases continues to rise. Hair professionals can detect this cancer by regularly inspecting the skin under these hair-bearing areas. Providing skin cancer education to cosmetology students can potentially affect significant changes in screening and referral practices, improving the stage at which this cancer is treated. All articles in this study involved hair professionals or cosmetology students in the context of hair salons, barbershops, or cosmetology schools. These articles included a focus on skin cancer or melanoma education. A systematic search of electronic databases (CINAHL, MEDLINE, Science Direct, and Google Scholar) for full-length, peer-reviewed journal articles written in English in the last 20 years was performed in August 2022 and January 2023. Articles were initially screened by title, abstract, and keywords, followed by full-length review and reference scanning. Data were systematically extracted from included articles, summarized, and presented in narrative and table form. Search terms included “hairdressers, barbers, cosmetology students, salons, barbershops, cosmetology schools, education, curriculum, intervention, knowledge, skin cancer, and melanoma." Thirteen sources of evidence were included in the review, all of which were empirical research studies. Nine studies evaluated the effectiveness of a skin cancer educational intervention for hair professionals or cosmetology students, and 4 assessed skin cancer knowledge, attitudes, or beliefs in this population. All interventions improved knowledge, confidence, screening behaviors, or referral practices in hair professionals or cosmetology students. No specific educational modality was more effective than another, no apparent difference was seen with different intervention lengths, and the type of information included within each educational modality was rarely discussed. Most hair professionals believe they lack adequate training or confidence in skin cancer detection and are interested in becoming educated on this subject. Short videos or live lectures were the preferred skin cancer educational modalities among hair professionals and cosmetology students. Brief skin cancer educational programs improve cosmetology students and hair professionals’ knowledge about melanoma and screening practices. This population has a perceived and measured need for skin cancer education. Including this type of program in the cosmetology school curriculum can effectively disseminate information to these individuals, leading to earlier diagnosis of melanomas on the scalp and neck, thus decreasing the mortality rate of this disease. Further research, including larger, diverse samples and more rigorous study designs, is needed to bolster the results of the review findings.

Abstract B031: Investigating the associations of self-efficacy and social factors among Hispanics at Risk for Skin Cancer

Abstract Background: Skin cancer ranks as the most prevalent cancer in the United States. Over the past two decades, the incidence of melanoma, the deadliest form of skin cancer, among Hispanics has risen by 20%. Melanoma mortality rates are higher in Hispanics than in non-Hispanic whites. Early detection of melanoma via skin self-examination may lead to diagnosis of melanomas at an earlier stage, when they are thinner, less invasive, and more easily treatable, resulting in improved survival rates. Given the gap in research addressing culturally-relevant factors related to skin cancer prevention and detection among Hispanics and informed by the Preventive Health Model, this study tested the associations between social factors and self-efficacy with skin self- examination (SSE) intentions and queried participants about their preferences for skin cancer prevention-related interventions. Methods: Amazon Mechanical Turk (MTurk) was utilized for participant recruitment, and respondents who were at risk for skin cancer (finished a short screening survey) were invited to complete a Qualtrics survey. Upon finishing the Qualtrics survey, participants were given a completion code to obtain an incentive ($3.50) directly from Amazon. The entire study lasted approximately 20–30 minutes. A multivariable regression analysis was performed to explore the associations between descriptive norms, injunctive norms, provider-patient communication, self-efficacy and SSE intentions. PROCESS macro model 4 was used to test the mediation associations. Results: Among the respondents (n = 79), 55.7% were female (n = 44), and 89.9% held a college or higher degree (n = 71). Self-efficacy fully mediated the effects of descriptive norms, injunctive norms, and provider-patient communication on SSE intentions among Hispanics. There was a significant indirect effect of descriptive norms on SSE intentions via its prior effect on self-efficacy (Bindirect = 0.43, SE = 0.14, 95% CI = 0.158, 0.698). Similarly, there was a significant indirect effect of injunctive norms on SSE intentions via its prior effect on self-efficacy (Bindirect = 0.45, SE = 0.12, 95% CI = 0.217, 0.692). Moreover, an indirect effect of provider-patient communication via self-efficacy on SSE intentions was also found (Bindirect = 0.66, SE = 0.26, 95% CI = 0.211, 1.196). On average, respondents demonstrated considerable interest in participating in a skin cancer-related behavioral intervention using a mobile application (75.6%) and/or using WhatsApp (71.8%). Conclusions: These preliminary findings provide new insights for development of future digital skin cancer intervention programs among Hispanics targeting social factors, including social norms and provider-patient communication, and utilizing preferred digital tools. Citation Format: Zhaomeng Niu, Carolyn Heckman. Investigating the associations of self-efficacy and social factors among Hispanics at Risk for Skin Cancer [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr B031.

Prevention and early detection of the most aggressive skin cancer: melanoma

Skin cancers are the most common cancers and they are divided into two types: melanoma and non-melanoma skin cancers (NMSCs). Melanoma is the most lethal and aggressive type of skin cancer. The incidence of this type of cancer is increasing. Mortality from melanoma can be reduced by avoidance of exposure to ultraviolet radiation and regular skin examination. Materials and methods of research English-language scientific literature available in the PubMed and Google Scholar databases was reviewed using the following keywords: ”melanoma”; ”early-detection”; ”self-screening”; ”tanning”; ”sunscreen”; ”dermatoscopy”; ”mortality”; ”prevention”; ”risk-factors”; ”UV protection”; ”epidemiology”. Aim of the study The purpose of this narrative review is to present risk factors, prevention and the importance of early detection of melanoma. This review also shows that it is possible to increase people's awareness of skin self-examination and the importance of UV protection through the use of modern technology (text messaging). Conclusions There are many risk factors for melanoma. People's awareness of the risks of UV exposure from both sunbathing and tanning beds is increasing. More and more people are starting to protect themselves from the sun and regularly check their skin. This behavior reduces the risk of developing melanoma and increases the chances of detecting melanoma in its early stages.

Deep learning approach for skin melanoma and benign classification using empirical wavelet decomposition.

Melanoma is a malignant skin cancer that causes high mortality. Early detection of melanoma can save patients' lives. The features of the skin lesion images can be extracted using computer techniques to differentiate early between melanoma and benign skin lesions. A new model of empirical wavelet decomposition (EWD) based on tan hyperbolic modulated filter banks (THMFBs) (EWD-THMFBs) was used to obtain the features of skin lesion images by MATLAB software. The EWD-THMFBs model was compared with the empirical short-time Fourier decomposition method based on THMFBs (ESTFD-THMFBs) and the empirical Fourier decomposition method based on THMFBs (EFD-THMFBs). The accuracy rates obtained for EWD-THMFBs, ESTFD-THMFBs, and EFD-THMFBs models were 100%, 98.89%, and 83.33%, respectively. The area under the curve (AUC) was 1, 0.97, and 0.91, respectively. The EWD-THMFBs model performed best in extracting features from skin lesion images. This model can be programmed on a mobile to detect skin lesions in rural areas by a nurse before consulting a dermatologist.

Melanoma skin cancer detection based on deep learning methods and binary Harris Hawk optimization

AbstractThe issue of skin cancer has garnered significant attention from the scientific community worldwide, with melanoma being the most lethal and uncommon form of the disease. Melanoma occurs due to the uncontrolled growth of melanocyte cells, which are responsible for imparting color to the skin. If left untreated, melanoma can spread throughout the body and cause death. Early detection of melanoma can lower its mortality rate. In this study, we propose a robust Convolutional Neural Network (CNN)-based method for classifying melanoma images as healthy or non-healthy. To train and test the model, we utilized public datasets from International Skin Imaging Collaboration (ISIC). Additionally, we compared our method with other classification techniques, including Support Vector Machine (SVM), Decision Tree, and K-Nearest Neighbors (K-NN), using the Harris Hawks Optimization algorithm. The results of our method showed superior performance compared to the other approaches.

Enhancing Melanoma Diagnosis with Advanced Deep Learning Models Focusing on Vision Transformer, Swin Transformer, and ConvNeXt.

Skin tumors, especially melanoma, which is highly aggressive and progresses quickly to other sites, are an issue in various parts of the world. Nevertheless, the one and only way to save lives is to detect it at its initial stages. This study explores the application of advanced deep learning models for classifying benign and malignant melanoma using dermoscopic images. The aim of the study is to enhance the accuracy and efficiency of melanoma diagnosis with the ConvNeXt, Vision Transformer (ViT) Base-16, and Swin Transformer V2 Small (Swin V2 S) deep learning models. The ConvNeXt model, which integrates principles of both convolutional neural networks and transformers, demonstrated superior performance, with balanced precision and recall metrics. The dataset, sourced from Kaggle, comprises 13,900 uniformly sized images, preprocessed to standardize the inputs for the models. Experimental results revealed that ConvNeXt achieved the highest diagnostic accuracy among the tested models. Experimental results revealed that ConvNeXt achieved an accuracy of 91.5%, with balanced precision and recall rates of 90.45% and 92.8% for benign cases, and 92.61% and 90.2% for malignant cases, respectively. The F1-scores for ConvNeXt were 91.61% for benign cases and 91.39% for malignant cases. This research points out the potential of hybrid deep learning architectures in medical image analysis, particularly for early melanoma detection.

The role of physical therapists in the early detection of skin melanoma: insights from an anonymous survey.

Skin melanoma is the most severe form of skin cancer. Recent years have seen an increase in melanoma incidence. Melanomas often appear on the back, a hidden area, leading to late diagnoses. Physical therapists, who frequently treat patients with lower back pain, could play a crucial role in early melanoma detection. An anonymous online survey was conducted among 254 Israeli physical therapists to assess their perspectives on melanoma detection, their knowledge in identifying suspicious lesions, and their referral patterns. The survey included sections on demographics, professional perspectives, melanoma knowledge, personal or family experiences with melanoma, and referral frequency for dermatological evaluation. The survey revealed that 75.2% of physical therapists viewed melanoma detection as part of their professional duties, yet 59.1% reported insufficient knowledge in identifying suspicious lesions. Despite this, 94.1% expressed a desire to improve their knowledge. However, only 44.1% actively referred patients with suspicious lesions to dermatologists. There was a significant positive correlation between professional experience and referral rates (p-value < 0.001), indicating that more experienced therapists were more likely to refer patients. The study highlights the critical need for incorporating melanoma detection training into physical therapy education and professional development. Enhancing physical therapists' knowledge and skills in this area could improve early detection and patient outcomes. Despite the potential role of physical therapists in melanoma detection, current training programs lack emphasis on this aspect, underscoring the importance of revising educational curricula to include skin cancer detection techniques. The findings suggest that while physical therapists recognize their role in melanoma detection, there is a significant knowledge gap. Addressing this through targeted education and training could enhance early detection efforts and improve patient care.

Three-Point Checklist Dermoscopy for Melanoma Screening: Experience in Medical Students New to Dermatology.

Early detection of melanoma and optimal referral to the specialist starts in primary care. Medical education is usually deficient in training general physicians in early detection and risk management for most skin malignancies. A three-point dermoscopy checklist is used as a screening tool for differentiating malignant and benign pigmented lesions in non-expert clinicians using dermoscopy. To evaluate the impact of brief medical training on the three-point dermoscopy algorithm in third-year medical students new to dermatology and to determine the levels of sensitivity and specificity to differentiate malignant and benign pigmented lesions. Optional dermoscopy lecture for third-year medical students new to dermatology in the context of general medical semiology courses, with case discussion and evaluation of 50 dermoscopy cases (25 benign and 25 malignant). Students were asked to classify malignant versus benign pathology based on the three-point dermoscopy algorithm discussed. Sensitivity, specificity, and predictive values were calculated according to the students' responses. Sixty-five students provided 3250 responses. Malignant pathology was misclassified as benign in 154 responses, while benign pathology was misclassified as malignant in 668 responses. Sensitivity and specificity for differentiating malignant lesions were 89.70% and 61.99%, respectively. Moderate interobserver agreement was found (Kappa value = 0.50; [CI: 0.47-0.54]). When evaluating melanocytic lesions, the focus of primary healthcare and general medical education should emphasize the correct determination of malignant or benign pathology. Teaching the three-point dermoscopy rule to medical students new to dermatology yields satisfactory levels of sensitivity and specificity, comparable to general physicians.

Associations of Social Factors and Self-Efficacy with Skin-Self Examination Intentions Among Hispanics at Risk for Skin Cancer and Their Preferences for Digital Interventions.

Skin cancer ranks as the most prevalent cancer in the United States. Over the past two decades, the incidence of melanoma, the deadliest form of skin cancer, among Hispanics has risen by 20%. Melanoma mortality rates are higher in Hispanics than in non-Hispanic Whites (NHW). Early detection of melanoma via skin self-examination may lead to diagnosis of melanoma at an earlier stage, when they are thinner, less invasive, and more easily treatable, resulting in improved survival rates. Given the gap in research addressing culturally relevant factors related to skin cancer prevention and detection among Hispanics and informed by the Preventive Health Model, this study tested the associations between social and normative factors and self-efficacy with thorough skin self-examination (TSSE) intentions and queried participants about their preferences for skin cancer-related interventions. Among respondents (n = 79), 55.7% were female (n = 44), and 89.9% held a college or higher degree (n = 71). Self-efficacy fully mediated the effects of descriptive norms, injunctive norms, and provider-patient communication on TSSE intentions among Hispanics. On average, respondents demonstrated considerable interest in participating in a skin cancer-related behavioral intervention using a mobile application (75.6%) and/or using WhatsApp (71.8%). These preliminary findings provide new insights for development of future digital skin cancer intervention programs among Hispanics targeting social factors, including social norms and provider-patient communication, and utilizing preferred digital tools.

Melanoma classification using generative adversarial network and proximal policy optimization.

In oncology, melanoma is a serious concern, often arising from DNA changes caused mainly by ultraviolet radiation. This cancer is known for its aggressive growth, highlighting the necessity of early detection. Our research introduces a novel deep learning framework for melanoma classification, trained and validated using the extensive SIIM-ISIC Melanoma Classification Challenge-ISIC-2020 dataset. The framework features three dilated convolution layers that extract critical feature vectors for classification. A key aspect of our model is incorporating the Off-policy Proximal Policy Optimization (Off-policy PPO) algorithm, which effectively handles data imbalance in the training set by rewarding the accurate classification of underrepresented samples. In this framework, the model is visualized as an agent making a series of decisions, where each sample represents a distinct state. Additionally, a Generative Adversarial Network (GAN) augments training data to improve generalizability, paired with a new regularization technique to stabilize GAN training and prevent mode collapse. The model achieved an F-measure of 91.836% and a geometric mean of 91.920%, surpassing existing models and demonstrating the model's practical utility in clinical environments. These results demonstrate its potential in enhancing early melanoma detection and informing more accurate treatment approaches, significantly advancing in combating this aggressive cancer.

Clinicopathologically Defined Nevus Subtypes and Melanoma Risk

Early detection of melanoma is a major determinant in disease outcome and drives the number of (over)excised nevi in clinical practice. This study aimed to evaluate demographic features and melanoma risk of clinically suspicious, mainly flat nevus subtypes. Based on the methodology of exvivo dermoscopy and derm dotting, the 12 most prevalent nevus subtypes were identified in a collection of over 7000 nevi excised for medical reasons. Dermoscopical, histopathological and clinical features of these subtypes were described. In addition, the association with melanoma history, histopathological atypia and melanoma occurrence within nevi was compared. Nearly half of the nevi removed for medical reasons were of the hypermelanotic subtype with no or mild histopathological atypia and low melanoma association, suggesting overtreatment in daily practice. Contrarily, the subtypes atypical lentiginous nevus and orange pulverocytic flat nevus were associated with higher proportions of (severe) atypia and melanoma (history). We believe these subtypes may reflect different tumoral and/or (germline) genetic entities with different melanoma risk. The data from this study may direct further prospective research on specific nevus subtypes in order to obtain better insights in associated clinical/genetic factors and melanoma risk.

SKIN CANCER DETECTION USING CNN

The skin is the main component of a human body and serves to cover all of the body's muscles, bones, and other structures. Many people these days are afflicted with skin cancer. Skin cancer that is most deadly is called malignant melanoma. Melanoma is the most dangerous kind of cancer. It is a massive malignant tumour on the skin, and its spread is growing daily. If melanoma is discovered early on, treatment is straightforward. Melanoma diagnosis involves both automated and clinical procedures. Early detection of malignant melanoma has significant potential with image-based computer aided diagnosis methods. Automatically identifying the type of skin cancer from the pictures can help with a speedier diagnosis and improved accuracy.

Automated segmentation for early detection of uveal melanoma

ObjectiveUveal melanoma is the most common intraocular malignancy in adults. Current screening and triaging methods for melanocytic choroidal tumours face inherent limitations, particularly in regions with limited access to specialized ocular oncologists. This study explores the potential of machine learning to automate tumour segmentation. We develop and evaluate a machine-learning model for lesion segmentation using ultra-wide-field fundus photography. MethodA retrospective chart review was conducted of patients diagnosed with uveal melanoma, choroidal nevi, or congenital hypertrophy of the retinal pigmented epithelium at a tertiary academic medical centre. Included patients had a single ultra-wide-field fundus photograph (Optos PLC, Dunfermline, Fife, Scotland) of adequate quality to visualize the lesion of interest, as confirmed by a single ocular oncologist. These images were used to develop and test a machine-learning algorithm for lesion segmentation. ResultsA total of 396 images were used to develop a machine-learning algorithm for lesion segmentation. Ninety additional images were used in the testing data set along with images of 30 healthy control individuals. Of the images with successfully detected lesions, the machine-learning segmentation yielded Dice coefficients of 0.86, 0.81, and 0.85 for uveal melanoma, choroidal nevi, and congenital hypertrophy of the retinal pigmented epithelium, respectively. Sensitivities for any lesion detection per image were 1.00, 0.90, and 0.87, respectively. For images without lesions, specificity was 0.93. ConclusionOur study demonstrates a novel machine-learning algorithm's performance, suggesting its potential clinical utility as a widely accessible method of screening choroidal tumours. Additional evaluation methods are necessary to further enhance the model's lesion classification and diagnostic accuracy.

The effect of hair removal and filtering on melanoma detection: a comparative deep learning study with AlexNet CNN.

Melanoma is the most aggressive and prevalent form of skin cancer globally, with a higher incidence in men and individuals with fair skin. Early detection of melanoma is essential for the successful treatment and prevention of metastasis. In this context, deep learning methods, distinguished by their ability to perform automated and detailed analysis, extracting melanoma-specific features, have emerged. These approaches excel in performing large-scale analysis, optimizing time, and providing accurate diagnoses, contributing to timely treatments compared to conventional diagnostic methods. The present study offers a methodology to assess the effectiveness of an AlexNet-based convolutional neural network (CNN) in identifying early-stage melanomas. The model is trained on a balanced dataset of 10,605 dermoscopic images, and on modified datasets where hair, a potential obstructive factor, was detected and removed allowing for an assessment of how hair removal affects the model's overall performance. To perform hair removal, we propose a morphological algorithm combined with different filtering techniques for comparison: Fourier, Wavelet, average blur, and low-pass filters. The model is evaluated through 10-fold cross-validation and the metrics of accuracy, recall, precision, and the F1 score. The results demonstrate that the proposed model performs the best for the dataset where we implemented both a Wavelet filter and hair removal algorithm. It has an accuracy of 91.30%, a recall of 87%, a precision of 95.19%, and an F1 score of 90.91%.

Early detection of skin cancer using AI: Deciphering dermatology images for melanoma detection

This Review explores the transformative impact of artificial intelligence (AI) on the early detection of skin cancer, with a specific focus on melanoma, a potentially lethal form of the disease. Beginning with an overview of traditional diagnostic methods and their limitations, this paper delves into the evolution of AI within dermatology, emphasizing its application in image analysis and pattern recognition. A comprehensive examination of AI algorithms for melanoma detection, including machine learning and deep learning models, is provided. This Review critically assesses the performance metrics, training datasets, and comparative analyses with traditional methods. Addressing challenges such as data quality, interpretability, and ethical considerations, this paper outlines future directions, emphasizing ongoing research, algorithm improvements, and integration with clinical practices. Case studies and success stories highlight the real-world impact of AI in dermatology. This Review concludes by summarizing key findings and underlining the pivotal role of AI in revolutionizing early melanoma detection, with implications for personalized medicine and enhanced patient outcomes.

Site-specific patterns of early-stage cancer diagnosis during the COVID-19 pandemic.

The COVID-19 pandemic caused widespread disruptions in cancer care. We hypothesized that the greatest disruptions in diagnosis occurred in screen-detected cancers. We identified patients (≥18 years of age) with newly diagnosed cancer from 2019 to 2020 in the US National Cancer Database and calculated the change in proportion of early-stage to late-stage cancers using a weighted linear regression. Disruptions in early-stage diagnosis were greater than in late-stage diagnosis (17% vs 12.5%). Melanoma demonstrated the greatest relative decrease in early-stage vs late-stage diagnosis (22.9% vs 9.2%), whereas the decrease was similar for pancreatic cancer. Compared with breast cancer, cervical, melanoma, prostate, colorectal, and lung cancers showed the greatest disruptions in early-stage diagnosis. Uninsured patients experienced greater disruptions than privately insured patients. Disruptions in cancer diagnosis in 2020 had a larger impact on early-stage disease, particularly screen-detected cancers. Our study supports emerging evidence that primary care visits may play a critical role in early melanoma detection.