skin cancer research paper example

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Review Article
• Open access
• Published: 09 April 2024

Human-AI interaction in skin cancer diagnosis: a systematic review and meta-analysis

• Isabelle Krakowski 1 , 2 , 3   na1 ,
• Jiyeong Kim   ORCID: orcid.org/0000-0002-2869-5751 1 , 3   na1 ,
• Zhuo Ran Cai 1 , 3 ,
• Roxana Daneshjou   ORCID: orcid.org/0000-0001-7988-9356 4 ,
• Jan Lapins 5 ,
• Hanna Eriksson 2 , 6 ,
• Anastasia Lykou 7 &
• Eleni Linos   ORCID: orcid.org/0000-0002-5856-6301 1 , 3  

npj Digital Medicine volume  7 , Article number:  78 ( 2024 ) Cite this article

455 Accesses

7 Altmetric

Metrics details

• Skin cancer
• Skin manifestations

The development of diagnostic tools for skin cancer based on artificial intelligence (AI) is increasing rapidly and will likely soon be widely implemented in clinical use. Even though the performance of these algorithms is promising in theory, there is limited evidence on the impact of AI assistance on human diagnostic decisions. Therefore, the aim of this systematic review and meta-analysis was to study the effect of AI assistance on the accuracy of skin cancer diagnosis. We searched PubMed, Embase, IEE Xplore, Scopus and conference proceedings for articles from 1/1/2017 to 11/8/2022. We included studies comparing the performance of clinicians diagnosing at least one skin cancer with and without deep learning-based AI assistance. Summary estimates of sensitivity and specificity of diagnostic accuracy with versus without AI assistance were computed using a bivariate random effects model. We identified 2983 studies, of which ten were eligible for meta-analysis. For clinicians without AI assistance, pooled sensitivity was 74.8% (95% CI 68.6–80.1) and specificity was 81.5% (95% CI 73.9–87.3). For AI-assisted clinicians, the overall sensitivity was 81.1% (95% CI 74.4–86.5) and specificity was 86.1% (95% CI 79.2–90.9). AI benefitted medical professionals of all experience levels in subgroup analyses, with the largest improvement among non-dermatologists. No publication bias was detected, and sensitivity analysis revealed that the findings were robust. AI in the hands of clinicians has the potential to improve diagnostic accuracy in skin cancer diagnosis. Given that most studies were conducted in experimental settings, we encourage future studies to further investigate these potential benefits in real-life settings.

Similar content being viewed by others

Deep learning-aided decision support for diagnosis of skin disease across skin tones

Matthew Groh, Omar Badri, … Rosalind Picard

Human–computer collaboration for skin cancer recognition

Philipp Tschandl, Christoph Rinner, … Harald Kittler

Systematic review of deep learning image analyses for the diagnosis and monitoring of skin disease

Shern Ping Choy, Byung Jin Kim, … Satveer K. Mahil

Introduction

As a result of increasing data availability and computational power, artificial intelligence (AI) algorithms—have reached a level of sophistication that enables them to take on complex tasks previously only conducted by human beings 1 . Several AI algorithms are now approved by the United States Food and Drug Administration (FDA) for medical use 2 , 3 , 4 . Though there are currently no image-based dermatology AI applications that have FDA approval, several are in development 2 .

Skin cancer diagnosis relies heavily on the interpretation of visual patterns, making it a complex task that requires extensive training in dermatology and dermatoscopy 5 , 6 . However, AI algorithms have been shown to accurately diagnose skin cancers, even outperforming experienced dermatologists in image classification tasks in constrained settings 7 , 8 , 9 . However, these algorithms can be sensitive to data distribution shifts. Therefore, AI-human partnerships could provide performance improvements that surmount the limitations of both human clinicians or AI alone. Notably, Tschandl et al. demonstrated in their 2020 paper that the accuracy of clinicians supported by AI algorithms surpassed that of either clinicians or AI algorithms working separately 10 . This approach of an AI-clinician partnership is considered the most likely clinical use of AI in dermatology, given the ethical and legal concerns of automated diagnosis alone. Therefore, there is an urgent need to better understand how the use of AI by clinicians affects decision making 11 . The goal of this study was to evaluate the diagnostic accuracy of clinicians with vs. without AI assistance using a systematic review and meta-analysis of the available literature.

Literature search and screening

For this systematic review and meta-analysis, 2983 records were initially retrieved, of which, 1972 abstracts were screened after the automatic duplicate removal by Covidence (Fig. 1 ). As 1936 articles were considered irrelevant and further excluded, the full text of 36 articles was reviewed. A total of 12 studies were included in the systematic review 10 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 and ten studies were included in the meta-analysis 10 , 12 , 13 , 14 , 15 , 17 , 19 , 20 , 21 , 22 , whereas the information needed to create contingency tables of AI-assisted and un-assisted medical professionals was unavailable in two studies 16 , 18 .

Flow diagram of the study selection process.

Study characteristics

Tables 1 and 2 presents the characteristics of the included studies. Half of the studies were conducted in Asia (50%, South Korea=5, China=1) and the other half was done in North/South America (25%, USA = 1, Argentina=1, Chile=1), and Europe (25%, Austria=1, Germany=1, Switzerland=1). More studies were performed in experimental (67%, n  = 8) than clinical settings (33% n  = 4). A quarter of studies included only dermatologists (25%, n  = 3), more than a half (58%, n  = 7) included a combination of dermatology specialists (e.g., dermatologist and dermatology residents) and non-dermatology medical professionals (e.g., primary care physicians, nurse practitioners, medical students) and among these, two studies included lay persons, but this data was not included for meta-analysis. In two studies (17%), only non-dermatology medical professionals were included. The median number of study participants was 18.5, ranging from 7 to 302.

Clinical information was provided to study participants in addition to images or in-patient visits in half of the studies (50%, n  = 6). For diagnosis, outpatient clinical images were most frequently provided (42%, n  = 5), followed by dermoscopic images (33%, n  = 4) and in-patient visits (25%, n  = 3). Diagnostic task was either choosing the most likely diagnosis (58%, n  = 7) or rating the lesion as malignant vs. benign (42%, n  = 5). Most studies (75%, n  = 9) used a paired design with the same reader diagnosing the same case first without, then with AI assistance, whereas two studies provided different images between the two tasks. A fully crossed design (i.e., all readers diagnosing all cases in both modalities) was performed in four studies. One study only reported diagnosis with AI support, thus did not allow to analyze the effect of AI 16 . Studies included a reference standard that was either varying combinations of either histopathology, a dermatologist panel’s diagnosis or the treating physician, from medical records, clinical follow-up or in vivo confocal microscopy (75%, n  = 9) or histopathologic diagnosis on all images (17%, n  = 2). One study considered either histopathology or the study participant being in concordance with two AI tools that were studied as the reference standard 17 . Most AI algorithms did not provide explanation for their outputs or presentation beyond the top-1 or top-3 diagnoses along with their respective probabilities or a binary malignancy score. Content-based image retrieval (CBIR) was the only explainability method that was used, namely in two of the studies (17%) and Tschandl et al. 10 was the only study that delved into the effects of various representation of AI output on the diagnostic performance of physicians. Definition of target condition varied across studies, but all studies included at least one skin cancer among the differential diagnoses. The summary of methodological quality assessments can be found in Supplementary Table 1 . Although κ was low (κ = 0.33), the Bowker’s Test of Symmetry 23 was not significant, hence two raters were considered having the same propensity to select categories. All three assessors agreed with the final quality assessments.

Meta-analyses results

The summary estimate of sensitivity for clinicians overall was 74.8% (95% CI 68.6–80.1) and specificity 81.5% (73.9–87.3). The overall diagnostic accuracy increased with AI assistance to a pooled sensitivity and specificity of 81.1% (74.4–86.5) and 86.1% (79.2–90.9), respectively. The SROC curves and forest plots of ten studies for clinicians without vs. with AI assistance each are shown in Figs. 2 and 3 , respectively, where less heterogeneity is observed in the sensitivity of clinicians overall compared to clinicians with AI assistance.

SE sensitivity, SP specificity. Performance of clinicians with no AI assistance ( a ) compared to AI-assisted clinicians ( b ) in the included studies.

Forest plots. Meta-analysis results of the diagnostic performance of clinicians without ( a ) or with ( b ) AI assistance.

To investigate the effect of AI assistance in more detail, we conducted subgroup analyses based on clinical experience level, test task and image type (Table 3 ). We observed that dermatologists had the highest diagnostic accuracy in terms of sensitivity and specificity. Residents (including dermatology residents and interns) were the second most accurate group, followed by non-dermatologists (including primary care providers, nurse practitioners and medical students). Notably, AI assistance significantly improved the sensitivity and specificity of all groups of clinicians. The non-dermatologist group appeared to benefit the most from AI assistance regarding improvement of pooled sensitivity (+13 points) and specificity (+11 points). For classification task, the sensitivity of both binary classification (malignant vs. benign) and top diagnosis improved with AI assistance. Meanwhile, AI assistance significantly improved pooled specificity only for top classification, reaching a specificity of 88.8%, (86.5–90.8). No significant difference was observed for image type.

There was no evidence of a small-study effect in regression test asymmetry for both humans without ( p  = 0.33) and with AI assistance ( p  = 0.23). Please see Supplementary Fig. 1 for funnel plots. The Spearman correlation test found that the presence of positive threshold effect was low likely for both groups. Sensitivity analyses revealed that excluding outliers slightly increased the pooled sensitivity and specificity in both groups while the pooled sensitivity and specificity mostly remained unchanged when excluding the low-quality study (Supplementary Table 2 ).

This systematic review and meta-analysis included 12 studies and 67,700 diagnostic evaluations of potential skin cancer by clinicians with and without AI assistance. Our findings highlight the potential of AI-assisted decision-making in skin cancer diagnosis. All clinicians, regardless of their training level, showed improved diagnostic performance when assisted by AI algorithms. The degree of improvement, however, varied across specialties, with dermatologists exhibiting the smallest increase in diagnostic accuracy and non-dermatologists, including primary care providers, demonstrating the largest improvement. These results suggest that AI assistance may be especially beneficial for clinicians without extensive training in dermatology. Given that many dermatological AI devices have recently obtained regulatory approval in Europe, including some CE marked algorithms utilized in the analyzed studies 24 , 25 , AI assistance may soon be a standard part of a dermatologist’s toolbox. It is therefore important to better understand the interaction between human and AI in clinical decision-making.

While several studies have been conducted to evaluate the dermatologic use of new AI tools, our review of published studies found that most have only compared human clinician performance with that of AI tools, without considering how clinicians interact with these tools. Two of the studies in this systematic review and meta-analysis reported that clinicians perform worse when the AI tool provides incorrect recommendations 10 , 19 . This finding underscores the importance of accurate and reliable algorithms in ensuring that AI implementation enhances clinical outcomes, and highlights the need for further research to validate AI-assisted decision-making in medical practice. Notably, in a recent study by Barata et al. 26 , the authors demonstrated that a reinforcement learning model that incorporated human preferences outperformed a supervised learning model. Furthermore, it improved the performance of participating dermatologists in terms of both diagnostic accuracy and optimal management decisions of potential skin cancer when compared to either a supervised learning model or no AI assistance at all. Hence, the development of algorithms in collaboration with clinicians appears to be important for optimizing clinical outcomes.

Only two studies explored the impact of one explainability technique (CBIR) on physician’s diagnostic accuracy or perceived usefulness. The real clinical utility of explainability methods needs to be further examined, and current methods should be viewed as tools to interrogate and troubleshoot AI models 27 . Additionally, prior research has shown that human behavioral traits can affect trust and reliance on AI assistance in general 28 , 29 . For example, a clinician’s perception and confidence in the AI’s performance on a given task may influence whether they decide to incorporate AI advice in their decision 30 . Moreover, research has also shown that the human’s confidence in their decision, the AI’s confidence level, and whether the human and AI agree all influence if the human incorporates the AI’s advice 30 . To ensure that AI assistance supports and improves diagnostic accuracy, future research should investigate how factors such as personality traits 29 , cognitive style 28 and cognitive biases 31 affect diagnostic performance in real clinical situations. Such research would help inform the integration of AI into clinical practice.

Our findings suggest that AI assistance may be particularly beneficial for less experienced clinicians, consistent with prior studies of human-AI interaction in radiology 32 . This highlights the potential of AI assistance as an educational tool for non-dermatologists and for improving diagnostic performance in settings such as primary care or for dermatologists in training. In a subgroup analysis, we observed no significant difference between AI-assisted other medical professionals vs. unassisted dermatologists (data not shown). However, this area warrants further research.

Some limitations need to be considered when interpreting the findings. First, among the ten studies that provided sufficient data to conduct meta-analysis, there were differences in design, number and experience level of participants, target condition definition, classification task, and algorithm output and training. Taken together, this heterogeneity implies that direct comparisons should be interpreted carefully. Furthermore, caution is warranted for the interpretation of the subgroup analyses due to the small sample size of the subgroups (up to seven) and the data structure (i.e., repeated measures) since the same participants examined the clinical images both without and with AI assistance in most studies. Given the low number of studies, we refrained from performing further subgroup analyses, such as, comparing specific cancer diagnoses in the subset of articles where these are available. Despite these limitations, our results from this meta-analysis support the notion that AI assistance can yield a positive effect on clinician diagnostic performance. We were able to adjust for potential sources of heterogeneity, including diagnostic task and clinician experience level when comparing the diagnostic accuracy of clinicians with vs. without AI assistance. Moreover, no signs of publication bias and low likelihood of threshold effects were observed. Lastly, the findings were robust such that the pooled sensitivity and specificity nearly stayed the same after excluding outliers or low-quality studies.

Of note, few studies provided participating clinicians with both clinical data and dermoscopic images, which would be available in a real-life clinical situation. Previous research has shown that the use of dermoscopy enables a relative improvement of diagnostic accuracy of melanoma by almost 50% compared to the naked eye 5 . In one of such study, participants were explicitly not allowed to use dermoscopy during the patient examination 19 . Overall, only four studies were conducted in a prospective clinical setting, and three of these could be included for meta-analysis. Thus, most diagnostic ratings in this meta-analysis were made in experimental settings and do not necessarily reflect the decisions made in a clinical real-world situation.

One of the main concerns regarding the accuracy of AI tools rely on the quality of the data it has been trained on 33 . As only three studies used publicly available datasets, evaluation of the data quality is difficult. Furthermore, darker skin tones were underrepresented in the datasets of the included studies, which is a known problem in the field, as most papers do not report skin tone outputs 34 . However, datasets with diverse skin tones have been developed and made publicly available as an effort to reduce disparity in AI performance in dermatology 35 , 36 . Moreover, few studies provided detailed information about the origin and number of images that had been used for training, validation, and testing of the AI tool and different definitions of these terms were used across studies. There is a need for better transparency guidelines for AI tool reporting to enable users and readers to understand the limits and capabilities of these diagnostic tools. Efforts are being made to develop guidelines that are adapted for this purpose, including the STARD-AI 37 , TRIPOD-AI and, PROBAST-AI 38 guidelines, as well as the dermatology-specific CLEAR Derm guidelines 39 . In addition, PRISMA-AI 40 guidelines for systematic reviews and meta-analyses are being developed. These are promising initiatives that will hopefully make both the reporting and evaluation of AI diagnostic tool research more transparent.

The results of this systematic review and meta-analysis indicate that clinicians benefit from AI assistance in skin cancer diagnosis regardless of their experience level. Clinicians with the least experience in dermatology may benefit the most from AI assistance. Our findings are timely as AI is expected to be widely implemented in clinical work globally in the near future. Notably, only four of the identified studies were conducted in clinical settings, three of which could be included in the meta-analysis. Therefore, there is an urgent need for more prospective clinical studies conducted in real-life settings where AI is intended to be used, in order to better understand and anticipate the effect of AI on clinical decision making.

Search strategy and selection criteria

We searched four electronic databases, including PubMed, Embase, Institute of Electrical and Electronics Engineers Xplore (IEE Xplore) and Scopus for peer-reviewed articles of AI-assisted skin cancer diagnosis without language restriction from January 1, 2017, until November 8, 2022. Search terms were combined for four key concepts: (1) AI, (2) skin cancer, (3) diagnosis, (4) doctors. The full search strategy is available in the Supplementary material (Supplementary Table 3 ). We chose 2017 as the cutoff for this review since this was the year when deep learning was first reported as performing at a level comparable to dermatologists, notably in the seminal study by Esteva et al 9 , which suggested that AI technology had reached a clinically useful level in assisting skin cancer diagnosis.

We applied Google Translate software for abstract screening of non-English articles. Manual searches were performed for conference proceedings, including NeurIPS, HICSS, ICML, ICLR, AAAI, CVPR, CHIL and ML4Health, and to identify additional relevant articles by reviewing bibliographies and citations of the screened papers and searching Google Scholar.

We included studies comparing diagnostic accuracy of clinicians detecting skin cancer with and without AI assistance. If studies provided diagnostic data from medical professionals other than physicians this data was also included for analysis, as long as the study also included physicians. However, we excluded studies if (1) diagnosis was not made from either images of skin lesions or in-person visits (e.g., pathology slides), (2) diagnostic accuracy was only compared between clinicians and an AI algorithm, (3) non-deep learning techniques were used, or (4) the articles were editorials, reviews, and case reports. We did not limit participants’ expertise, study design or sample size, reference standard, or skin diagnosis if at least one skin malignancy was included in the study. We contacted nine authors to request additional data and clarifications required for the meta-analysis and received data from five of them 10 , 12 , 13 , 14 , 15 and clarifications from two 16 , 17 . In four studies 10 , 14 , 15 , 17 raw data was not available for all experiments or lesions, and our meta-analysis included the data that was available. Studies with insufficient data to construct contingency tables 16 , 18 were included in the systematic review but not in the meta-analysis.

Three reviewers performed eligibility assessment, data extraction, and study quality evaluations (IK, JK, ZRC). Commonly used standardized programs were employed for duplicate removal, title and abstract screening, and full-text review (Covidence) and data extraction (Microsoft Excel). Paired reviewers independently screened the titles and abstracts using predefined criteria and extracted data. Disagreement was resolved by discussions with the third reviewer. IK imported the extracted data into the summary table for systematic review and two reviewers (JK and ZRC) verified it. JK imported the extracted data and prepared it for meta-analysis and two reviewers (ZRC and IK) verified it. Biostatistician (AL) reviewed and confirmed the final data for meta-analysis. All co-authors reviewed the final tables and figures. This systematic review and meta-analysis followed the PRISMA DTA guidelines 41 and the study protocol was registered with PROSPERO, CRD42023391560.

Data analysis

We extracted key information, including true positive, false positive, false negative, and true negative information among clinicians with and without AI assistance. We generated contingency tables, where possible, to estimate diagnostic test accuracy in terms of pooled sensitivity and specificity. Additional information about the AI algorithm (e.g., architecture, image sources, validation and AI assistance method), participants, patients, target condition, reference standard, study setting and design, and funding were extracted.

A revised tool for the methodological quality assessment of diagnostic accuracy studies (QUADAS-2) 42 was used to assess risk of bias and concerns of applicability of each study in four domains, patient selection, index test, reference standard, and flow and timing (Supplementary Table 1 ). A pair of reviewers independently evaluated the domains, compared the ratings, and, if conflicted, reconciled the discrepancies through discussions led by the third reviewer (IK, JK, ZRC).

We used the Metandi package 43 for Stata 17 (College Station, TX) to compute summary estimates of sensitivity and specificity with 95% confidence intervals (95% CI) of humans with AI-assistance compared to humans without AI assistance using a bivariate model 44 . Summary Receiver Operating Characteristics (SROC) curves were plotted to visually present the summary estimates of sensitivity and specificity with 95% confidence region and the 95% prediction region, which refers to the confidence areas that the sensitivity and specificity of future studies likely fall into. The Bivariate models were performed separately for clinicians with vs. without AI assistance because the Metandi package could not handle the paired design of the data. We applied a random effects model to account for the anticipated heterogeneity across studies, potentially due to the variance of the data, including the use of different AI algorithms, medical professionals, and study settings. Heterogeneity was assessed by visual inspection of graphics, including SROC curve and forest plots 45 , 46 . Additionally, we conducted bivariate meta-regression analysis using the Meqrlogit package (Stata 17, College Station, TX) by the presence of AI assistance or not, for each experience level in dermatology (dermatologists, residents, non-dermatology medical professionals), type of diagnostic task (binary classification or top diagnosis) and type of image (clinical or dermoscopic) separately, to compare diagnostic accuracy by AI assistance and adjust for the potential heterogeneity caused by these factors 47 . To investigate the presence of a positive threshold effect, Spearman correlation coefficient was computed between sensitivity and specificity 48 . Pre-planned sensitivity analyses were conducted by excluding potential outliers, 49 studies with poor methodology (where at least three domains were rated as unclear or high bias), and studies with reference standards other than only histopathology. We examined publication bias using Deeks’ Funnel Plot Asymmetry Test, which ran a regression on the effective sample size funnel plots vs. diagnostic odds ratio 50 . We calculated κ statistics to evaluate the agreements between QUADAS-2 assessors. All statistical significance was determined at p  < 0.05.

Data availability

E.L. has full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. All study materials are available from the corresponding author upon reasonable request.

Code availability

The codes used in the analysis of this study will be made available from the corresponding author upon reasonable request.

Brynjolfsson, E. & Mitchell, T. What can machine learning do? Workforce implications. Science 358 , 1530–1534 (2017).

Article   CAS   PubMed   Google Scholar  

Wu, E. et al. How medical AI devices are evaluated: limitations and recommendations from an analysis of FDA approvals. Nat. Med. 27 , 582–584 (2021).

Yu, K.-H., Beam, A. L. & Kohane, I. S. Artificial intelligence in healthcare. Nat. Biomed. Eng. 2 , 719–731 (2018).

Article   PubMed   Google Scholar  

Topol, E. J. High-performance medicine: the convergence of human and artificial intelligence. Nat. Med. 25 , 44–56 (2019).

Kittler, H., Pehamberger, H., Wolff, K. & Binder, M. Diagnostic accuracy of dermoscopy. Lancet Oncol. 3 , 159–165 (2002).

Marghoob, A. A. & Scope, A. The complexity of diagnosing melanoma. J. Investig. Dermatol. 129 , 11–13 (2009).

Tschandl, P. et al. Comparison of the accuracy of human readers versus machine-learning algorithms for pigmented skin lesion classification: an open, web-based, international, diagnostic study. Lancet Oncol. 20 , 938–947 (2019).

Article   PubMed   PubMed Central   Google Scholar  

Haenssle, H. A. et al. Man against machine reloaded: performance of a market-approved convolutional neural network in classifying a broad spectrum of skin lesions in comparison with 96 dermatologists working under less artificial conditions. Ann. Oncol. 31 , 137–143 (2020).

Esteva, A. et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature 542 , 115–118 (2017).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Tschandl, P. et al. Human–computer collaboration for skin cancer recognition. Nat. Med. 26 , 1229–1234 (2020).

Ngiam, K. Y. & Khor, I. W. Big data and machine learning algorithms for health-care delivery. Lancet Oncol. 20 , e262–e273 (2019).

Lee, S. et al. Augmented decision-making for acral lentiginous melanoma detection using deep convolutional neural networks. J. Eur. Acad. Dermatol. Venereol. 34 , 1842–1850 (2020).

Cho, S. I. et al. Dermatologist-level classification of malignant lip diseases using a deep convolutional neural network. Br. J. Dermatol. 182 , 1388–1394 (2020).

Han, S. S. et al. Augmented intelligence dermatology: deep neural networks empower medical professionals in diagnosing skin cancer and predicting treatment options for 134 skin disorders. J. Investig. Dermatol. 140 , 1753–1761 (2020).

Jain, A. et al. Development and assessment of an artificial intelligence–based tool for skin condition diagnosis by primary care physicians and nurse practitioners in teledermatology practices. JAMA Netw. Open 4 , e217249–e217249 (2021).

Muñoz-López, C. et al. Performance of a deep neural network in teledermatology: a single-centre prospective diagnostic study. J. Eur. Acad. Dermatol. Venereol. 35 , 546–553 (2021).

Jahn, A. S. et al. Over-detection of melanoma-suspect lesions by a CE-certified smartphone app: performance in comparison to dermatologists, 2D and 3D convolutional neural networks in a prospective data set of 1204 pigmented skin lesions involving patients’ perception. Cancers 14 , 3829 (2022).

Lucius, M. et al. Deep neural frameworks improve the accuracy of general practitioners in the classification of pigmented skin lesions. Diagnostics 10 , 969 (2020).

Han, S. S. et al. Evaluation of artificial intelligence-assisted diagnosis of skin neoplasms: a single-center, paralleled, unmasked Randomized Controlled Trial. J. Investig. Dermatol. 142 , 2353–2362.e2352 (2022).

Kim, Y. J. et al. Augmenting the accuracy of trainee doctors in diagnosing skin lesions suspected of skin neoplasms in a real-world setting: a prospective controlled before-and-after study. PLoS One 17 , e0260895 (2022).

Ba, W. et al. Convolutional neural network assistance significantly improves dermatologists’ diagnosis of cutaneous tumours using clinical images. Eur. J. Cancer 169 , 156–165 (2022).

Maron, R. C. et al. Artificial intelligence and its effect on dermatologists’ accuracy in dermoscopic melanoma image classification: web-based survey study. J. Med. Internet Res. 22 , e18091 (2020).

Bowker, A. H. A test for symmetry in contingency tables. J. Am. Stat. Assoc. 43 , 572–574 (1948).

Beltrami, E. J. et al. Artificial intelligence in the detection of skin cancer. J. Am. Acad. Dermatol. 87 , 1336–1342 (2022).

Young, A. T., Xiong, M., Pfau, J., Keiser, M. J. & Wei, M. L. Artificial intelligence in dermatology: a primer. J. Investig. Dermatol. 140 , 1504–1512 (2020).

Barata, C. et al. A reinforcement learning model for AI-based decision support in skin cancer. Nat. Med. 29 , 1941–1946 (2023).

Ghassemi, M., Oakden-Rayner, L. & Beam, A. L. The false hope of current approaches to explainable artificial intelligence in health care. Lancet Digital Health 3 , e745–e750 (2021).

Krakowski, S. M., Haftor, D., Luger, J., Pashkevich, N. & Raisch, S. Humans and algorithms in organizational decision making: evidence from a field experiment. Acad. Manag. Proc. 2019 , 16633 (2019).

Article   Google Scholar  

Park, J. & Woo, S. E. Who likes artificial intelligence? personality predictors of attitudes toward artificial intelligence. J. Psychol. 156 , 68–94 (2022).

Vodrahalli, K., Daneshjou, R., Gerstenberg, T. & Zou, J. Do humans trust advice more if it comes from AI? An analysis of human-ai interactions. In Proc. 2022 AAAI/ACM Conference on AI, Ethics, and Society 763–777 (Association for Computing Machinery, Oxford, United Kingdom, 2022).

Ludolph, R. & Schulz, P. J. Debiasing health-related judgments and decision making: a systematic review. Med. Decis. Mak. 38 , 3–13 (2018).

Gaube, S. et al. Non-task expert physicians benefit from correct explainable AI advice when reviewing X-rays. Sci. Rep. 13 , 1383 (2023).

Breck, E., Polyzotis, N., Roy, S., Whang, S. & Zinkevich, M. Data validation for machine learning. In Proceedings of the Conference on Systems and Machine Learning, (2019)

Daneshjou, R., Smith, M. P., Sun, M. D., Rotemberg, V. & Zou, J. Lack of transparency and potential bias in artificial intelligence data sets and algorithms: a scoping review. JAMA Dermatol. 157 , 1362–1369 (2021).

Daneshjou, R. et al. Disparities in dermatology AI performance on a diverse, curated clinical image set. Sci. Adv. 8 , eabq6147 (2022).

Groh, M. et al. Evaluating Deep Neural Networks Trained on Clinical Images in Dermatology with the Fitzpatrick 17k Dataset. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) , 1820-1828 (2021).

Sounderajah, V. et al. Developing a reporting guideline for artificial intelligence-centred diagnostic test accuracy studies: the STARD-AI protocol. BMJ Open 11 , e047709 (2021).

Collins, G. S. et al. Protocol for development of a reporting guideline (TRIPOD-AI) and risk of bias tool (PROBAST-AI) for diagnostic and prognostic prediction model studies based on artificial intelligence. BMJ Open 11 , e048008 (2021).

Daneshjou, R. et al. 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. JAMA Dermatol. 158 , 90–96 (2022).

Cacciamani, G. E. et al. PRISMA AI reporting guidelines for systematic reviews and meta-analyses on AI in healthcare. Nat. Med. 29 , 14–15 (2023).

McInnes, M. D. F. et al. Preferred reporting items for a systematic review and meta-analysis of diagnostic test accuracy studies: the PRISMA-DTA statement. JAMA 319 , 388–396 (2018).

Whiting, P. F. et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann. Intern. Med. 155 , 529–536 (2011).

Harbord, R. M. & Whiting, P. metandi: meta–analysis of diagnostic accuracy using hierarchical logistic regression. Stata J. 9 , 211–229 (2009).

Reitsma, J. B. et al. Bivariate analysis of sensitivity and specificity produces informative summary measures in diagnostic reviews. J. Clin. Epidemiol. 58 , 982–990 (2005).

Macaskill P, T. Y., et al. editor(s). In Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy 1–46 (Cochrane, London, 2022).

Kim, K. W., Lee, J., Choi, S. H., Huh, J. & Park, S. H. Systematic review and meta-analysis of studies evaluating diagnostic test accuracy: a practical review for clinical researchers-Part I. General Guidance and Tips. Korean J. Radio. 16 , 1175–1187 (2015).

Takwoingi, Y. et al. Chapter 10: Undertaking meta-analysis. Draft version (4 October 2022) for inclusion in: Deeks, J. J., Bossuyt, P. M., Leeflang, M. M., Takwoingi, Y. In Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy 1–77 (Cochrane, London, 2022).

Zamora, J., Abraira, V., Muriel, A., Khan, K. & Coomarasamy, A. Meta-DiSc: a software for meta-analysis of test accuracy data. BMC Med. Res. Methodol. 6 , 31–31 (2006).

Harrer, M., Cuijpers, P., Furukawa, T. A. & Ebert, D. D. Doing Meta-Analysis With R: A Hands-On Guide , (Chapman & Hall/CRC Press, Boca Raton, FL and London, 2021).

Deeks, J. J., Macaskill, P. & Irwig, L. The performance of tests of publication bias and other sample size effects in systematic reviews of diagnostic test accuracy was assessed. J. Clin. Epidemiol. 58 , 882–893 (2005).

Download references

Acknowledgements

This project received no specific funding. E.L. is supported by the National Institutes of Health: Mid-career Investigator Award in Patient-Oriented Research (K24AR075060) and Research Project Grant (R01AR082109). I.K. received research funding from Radiumhemmet Research Funds (009614) and H.E. received funding from Radiumhemmet Research Funds (211063, 181083), Region Stockholm (FoUI-962339, FoUI-972654), the Swedish Cancer Society (2111617Pj, 210406JCIA01) and the Swedish Research Council (202201534). The funders played no role in study design, data collection, analysis and interpretation of data, or the writing of this manuscript.

Author information

These authors jointly supervised this work: Isabelle Krakowski, Jiyeong Kim.

Authors and Affiliations

Center for Digital Health, Stanford University School of Medicine, Stanford, CA, USA

Isabelle Krakowski, Jiyeong Kim, Zhuo Ran Cai & Eleni Linos

Department of Oncology and Pathology, Karolinska Institutet, Stockholm, Sweden

Isabelle Krakowski & Hanna Eriksson

Department of Dermatology, Stanford, Stanford University, Stanford, CA, USA

Department of Dermatology, Department of Biomedical Data Science, Stanford School of Medicine, Stanford, CA, USA

Roxana Daneshjou

Department of Dermatology, Theme Inflammation, Karolinska University Hospital, Stockholm, Sweden

Theme Cancer, Unit of Head-Neck-, Lung- and Skin Cancer, Skin Cancer Center, Karolinska University Hospital, Stockholm, Sweden

Hanna Eriksson

Department of Education, University of Nicosia, Nicosia, Cyprus

Anastasia Lykou

You can also search for this author in PubMed   Google Scholar

Contributions

IK and JK contributed equally as joint first authors. Concept and design: EL, RD and IK. Literature search, screening process, data extraction and bias assessment: IK, JK and ZRC. Data analysis and interpretation: JK, AL, IK and EL. Drafting of the manuscript: IK and JK. Critical revision for important intellectual content and approval of the manuscript: All authors. Obtained funding: EL, HE and IK. Supervision: EL and AL.

Corresponding author

Correspondence to Eleni Linos .

Ethics declarations

Competing interests.

H.E. has served in advisory roles and delivered presentations for Novartis, BMS, GSK and Pierre Fabre and has obtained industry-sponsored research funding from SkylineDx. RD is an AAD AI committee member and Associate Editor at the Journal of Investigative Dermatology, has received consulting fees from Pfizer, L’Oreal, Frazier Healthcare Partners, and has stock options in Revea and MDAlgorithms. All other authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary information, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Krakowski, I., Kim, J., Cai, Z.R. et al. Human-AI interaction in skin cancer diagnosis: a systematic review and meta-analysis. npj Digit. Med. 7 , 78 (2024). https://doi.org/10.1038/s41746-024-01031-w

Download citation

Received : 22 September 2023

Accepted : 05 February 2024

Published : 09 April 2024

DOI : https://doi.org/10.1038/s41746-024-01031-w

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: Cancer newsletter — what matters in cancer research, free to your inbox weekly.

Skin Cancer Detection using Deep Learning

Ieee account.

• Change Username/Password
• Update Address

Purchase Details

• Payment Options
• Order History
• View Purchased Documents

Profile Information

• Communications Preferences
• Profession and Education
• Technical Interests
• US & Canada: +1 800 678 4333
• Worldwide: +1 732 981 0060
• Contact & Support
• About IEEE Xplore
• Accessibility
• Terms of Use
• Nondiscrimination Policy
• Privacy & Opting Out of Cookies

A not-for-profit organization, IEEE is the world's largest technical professional organization dedicated to advancing technology for the benefit of humanity. © Copyright 2024 IEEE - All rights reserved. Use of this web site signifies your agreement to the terms and conditions.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• J Med Internet Res
• v.23(11); 2021 Nov

Artificial Intelligence for Skin Cancer Detection: Scoping Review

Abdulrahman takiddin.

1 Department of Electrical and Computer Engineering, Texas A&M University, College Station, TX, United States

2 College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar

Jens Schneider

Alaa abd-alrazaq, mowafa househ, associated data.

Search query.

Data extraction form.

Study characteristics.

Data and deployment characteristics.

Technical details.

Data, model, and evaluation.

Skin cancer is the most common cancer type affecting humans. Traditional skin cancer diagnosis methods are costly, require a professional physician, and take time. Hence, to aid in diagnosing skin cancer, artificial intelligence (AI) tools are being used, including shallow and deep machine learning–based methodologies that are trained to detect and classify skin cancer using computer algorithms and deep neural networks.

The aim of this study was to identify and group the different types of AI-based technologies used to detect and classify skin cancer. The study also examined the reliability of the selected papers by studying the correlation between the data set size and the number of diagnostic classes with the performance metrics used to evaluate the models.

We conducted a systematic search for papers using Institute of Electrical and Electronics Engineers (IEEE) Xplore, Association for Computing Machinery Digital Library (ACM DL), and Ovid MEDLINE databases following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) guidelines. The studies included in this scoping review had to fulfill several selection criteria: being specifically about skin cancer, detecting or classifying skin cancer, and using AI technologies. Study selection and data extraction were independently conducted by two reviewers. Extracted data were narratively synthesized, where studies were grouped based on the diagnostic AI techniques and their evaluation metrics.

We retrieved 906 papers from the 3 databases, of which 53 were eligible for this review. Shallow AI-based techniques were used in 14 studies, and deep AI-based techniques were used in 39 studies. The studies used up to 11 evaluation metrics to assess the proposed models, where 39 studies used accuracy as the primary evaluation metric. Overall, studies that used smaller data sets reported higher accuracy.

Conclusions

This paper examined multiple AI-based skin cancer detection models. However, a direct comparison between methods was hindered by the varied use of different evaluation metrics and image types. Performance scores were affected by factors such as data set size, number of diagnostic classes, and techniques. Hence, the reliability of shallow and deep models with higher accuracy scores was questionable since they were trained and tested on relatively small data sets of a few diagnostic classes.

Introduction

Skin cancer is the most common cancer type that affects humans [ 1 ]. Melanoma and nonmelanoma are the two main types of skin cancer [ 2 ]. Nonmelanoma is of lesser concern since it usually can be cured by surgery and is nonlethal. Melanoma, however, is the most dangerous skin cancer type, with a high mortality rate, although it represents less than 5% of all skin cancer cases [ 1 ]. The World Health Organization (WHO) estimated 132,000 yearly melanoma cases globally. In 2015, 60,000 cases caused death [ 2 ].

Traditional methods of early detection of skin cancer include skin self-examination and skin clinical examination (screening) [ 3 ]. However, skin self-examination, where the patient or a family member notices a lesion, is a random method as people might overreact or underact. In addition, clinical examination using expensive, specialized medical tools, such as a dermoscope, microspectroscopy, and laser-based tools, requires training, effort to operate, time, and regular follow-ups [ 4 ]. Thus, patients have started using mobile technologies, such as smartphones, to share images with their doctors to get faster diagnoses. However, sharing images over the internet may compromise privacy. Worse yet, the image quality may not be sufficient, which may lead to inaccurate diagnoses. With evolvement, artificial intelligence (AI), which is the human-like intelligence exhibited by trained machines [ 5 ], has become so pervasive that most humans interact with AI-based tools daily, which assists physicians in decision making and decreases the decision variations among physicians. It is worth mentioning that even with the presence of such AI technologies, the role of an expert dermatologist is vital for diagnosis and treatment.

The focus of this review is on the use of AI as a tool that helps in the process of skin cancer diagnostics. Herein, AI-based skin cancer diagnostic tools use either shallow or deep AI methodologies. Both involve customizing computer algorithms through a process called training to learn from data formed by predefined features. The difference is that shallow methods tend to not use multilayer neural networks at all or use such networks limited to a minimum of layers [ 6 ]. In contrast, deep methodologies involve training large, deep multilayer neural networks with many hidden layers, typically ranging from dozens to hundreds [ 7 ].

Research Problem

Detecting skin cancer can be challenging, time consuming, and relatively expensive [ 4 ]. For example, Figure 1 shows two lesions that superficially seem identical [ 8 ]. However, the left image is of a normal benign lesion, whereas the right image shows a melanoma lesion. As AI technologies are becoming smarter and faster [ 5 ], it is hardly surprising that they are being used to assist in diagnosing skin cancer and suggesting courses of action. This is due to the fact that AI-based methods are considered to be relatively cheap, easy to use, and accessible [ 5 ]. Thus, they offer the potential to overcome the issues inherent in the aforementioned existing skin cancer detection methods. However, as the literature on the medical use of AI quickly grows and continues to report findings using incompatible performance metrics, direct comparison between prior work becomes more challenging and threatens to hamper future research. This study seeks to address this issue by performing a rigorous and transparent review of the existing literature. We aim to answer the research question, What are the existing AI-based tools that are used to detect and classify skin cancer?

Similarity of normal lesion (left) and melanoma (right).

This scoping review analyzes papers from different online databases. We defined strict inclusion and exclusion criteria to decide which papers to include. We then grouped the papers by the methodology used and analyzed the ground covered in the papers. Finally, we identified gaps in the literature and discussed how these gaps can be filled by future work. We developed a protocol before commencing the review. To ensure that this scoping review is transparent and replicable, we followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) instructions and guidelines [ 9 ].

Search Strategy

We conducted a systematic search on July 15, 2020. We identified articles from Institute of Electrical and Electronics Engineers (IEEE) Xplore, Association for Computing Machinery Digital Library (ACM DL), and Ovid MEDLINE databases. The terms used for searching the bibliographic databases were identified based on the target population (eg, “skin neoplasms,” “skin cancer,” “skin lesion”), intervention (eg, “artificial intelligence,” “machine learning,” “deep learning”), and outcome (“diagnosis,” “screening,” “detection,” “classification”). We derived the search terms from previous literature studies and reviews. For practical reasons, we did not conduct backward or forward reference list checking, and we also did not contact experts. Multimedia Appendix 1 shows the search strategy used for searching Ovid MEDLINE, where “skin neoplasms,” “artificial intelligence,” “machine learning,” and “deep learning” were used as MESH terms. Multimedia Appendix 1 also shows the search query for IEEE Xplore and ACM DL.

Study Eligibility Criteria

We included studies fulfilling the following criteria:

• Studies published between January 1, 2009, and July 15, 2020.
• Studies written in English.
• Population: studies discussing only skin cancer. Studies discussing other diseases or forms of cancer were excluded.
• Intervention: studies discussing only AI-based applications. Studies that discussed skin cancer–related applications or systems, including theoretical, statistical, or mathematical approaches, were excluded.
• Studies discussing the specific use of AI for detecting, classifying, or diagnosing skin cancer. Studies discussing only the general use of AI in a clinical setting were excluded.
• Studies proposing a new AI-based method. Case studies, surveys, review or response papers, or papers that reviewed, assessed, analyzed, evaluated, or compared existing methods were excluded.

No restrictions on the country of publication, study design, comparator, or outcomes were enforced.

Study Selection

Authors Abdulrahman Takiddin (AT) and Alaa Abd-Alrazaq (AA) independently screened the titles and abstracts of all retrieved studies. Following the written protocol, they independently read the full texts of the papers included in this study after reading their titles and abstracts. Any disagreements between both reviewers were resolved by discussion. We assessed the intercoder agreement by calculating the Cohen kappa (κ), which was 0.86 and 0.93 for screening titles and abstracts and for reading full texts, respectively, indicating good agreement.

Data Extraction

For reliable and accurate data extraction from the included studies, a data extraction form was developed and piloted using eight included studies ( Multimedia Appendix 2 ). The data extraction process was independently conducted by AT and AA. Any disagreements were resolved by discussion with good intercoder agreement (Cohen κ=0.88) between the reviewers.

Data Synthesis

A narrative approach was used to synthesize the extracted data. Specifically, we first grouped the included studies by diagnostic techniques based on complexity. Then, we discussed the evaluation metrics used in each study. Next, we grouped the studies based on the used evaluation metrics. In addition, we took into consideration the used data set in terms of the number of images, types of images, and number of diseases (diagnostic classes) that the data set contained. We assessed the correlation between the accuracy score and the number of images and diagnostic classes of the data set.

Search Results

After searching the 3 online databases, we retrieved a total of 906 studies. We then started excluding papers in three phases. As shown in Figure 2 , in the first phase, “identification,” we excluded 42 papers. In the second phase, “screening,” we excluded 711 papers. In the last phase, “eligibility,” we included 153 papers for a full-text review. After reviewing the full text of the papers, we excluded 100 papers. The specific reasons behind excluding the papers in each phase are mentioned in Figure 2 . Hence, the total number of included papers in this scoping review was 53.

PRISMA approach. ACM DL: Association for Computing Machinery Digital Library; AI: artificial intelligence; IEEE: Institute of Electrical and Electronics Engineers; PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Study Characteristics

Table 1 summarizes the characteristics of the selected studies. Figure 3 shows the number of papers published per year: 4 of 53 studies (7.6%) were published before 2016 [ 10 - 13 ], 26 studies (49.1%) were published in 2016, 2017, and 2018 [ 14 - 39 ], and 23 studies (43.4%) were published in 2019 and 2020 [ 40 - 62 ]. Although our selection criteria included papers published between 2009 and July 2020, the oldest published paper included after the full-text review was published in 2011. We observed that the number of papers sharply increased in 2018 and 2019.

Study characteristics (N=53).

Number of published papers by year.

Figure 4 shows the region of publication of the included studies. The studies included were published in different parts of the world. In Southern Asia, 22 studies (41.5%) were conducted in China, India, Bangladesh, Indonesia, Pakistan, Singapore, South Korea, and Thailand; 10 studies (18.9%) were conducted in North America, specifically the United States and Canada; 10 studies were conducted in Europe, including Austria, Poland, Germany, France, the United Kingdom, and Russia; 5 studies (9.4%) were conducted in the Middle East, including Lebanon, Turkey, Iran, and Saudi Arabia; 3 studies (5.7%) were conducted in Africa, specifically Egypt, South Africa, and Nigeria; and in Oceania, 3 studies were concluded in New Zealand and Australia.

Number of published papers by region.

The selected studies were either published in conference proceedings or journals: 31 of 53 studies (58.5%) were published in conference proceedings, and the rest of the papers (22/53, 41.5%) were published in journals. Multimedia Appendix 3 displays the characteristics of each included study.

Data Characteristics

Table 2 summarizes the characteristics of the used data in the selected studies. The studies used different sizes of data sets to train their models. The average number of used images in the selected studies was around 7800. The lowest number of images used was 40 [ 24 ], whereas the highest number of images used was 129,450 [ 23 ]. We categorized these data set sizes into three groups, depending on the number of images used. The first category contained small data sets that had fewer than 1000 images (21/53, 39.6%). The second category used medium-size data sets consisting of 1000-10,000 images (25/53, 47.2%). The last category contained large data sets that included more than 10,000 images (7/53, 13.2%).

Data and deployment characteristics (N=53).

We divided the papers into two groups based on the classification type. We found that more than half of the papers (31/53, 58.5%) built models to classify whether the lesion was benign or malignant (two-class/binary classification). The rest of the papers (22/53, 41.5%) presented models in which skin lesions were classified using three or more diagnostic classes (multiclass classification). Figure 5 shows the number of papers using different diagnostic classes. In the multiclass classification, 8 studies used 3 diagnostic classes, 1 study used 4 classes, 2 studies used 5 classes, 10 studies used 7 classes, and 1 study used 9 classes. The benign classes included benign keratosis, melanocytic nevus, and dermatofibroma. The malignant classes included melanoma and basal cell carcinoma. Other lesions, such as vascular lesions, actinic keratosis, genodermatosis, and tumors, could be either benign or malignant.

Number of published papers by number of diagnostic classes used.

With regard to the type of images used to train, test, and validate the models, 43 of 53 studies (81.1%) used dermoscopic images; 5 studies (9.4%) used clinical images that were taken using a normal camera; and 4 studies (7.5%) used high-quality images that were taken with a professional camera. The remaining study used spectroscopic images requiring a specialized system taking images of a lesion from three different spots using polarized and unpolarized light.

The majority of the studies (45/53, 84.9%) presented technologies that are still in the development phase. The rest of the studies (8/53, 15.1%) have been deployed into a usable form: 3 studies developed a health care system, 3 studies deployed the model into a mobile application, and 2 studies transferred the model into a web application. Multimedia Appendix 4 displays the data and deployment characteristics of each included study.

Diagnostic Techniques

We categorized the papers into two groups based on the AI technique used in detecting and classifying skin cancer. The groups were shallow techniques and deep techniques. These two groups differed mainly in the complexity of the AI architecture underlying the model. Shallow techniques use either simple machine learning algorithms, such as a support vector machine (SVM), or only a couple of layers of neural networks [ 63 ]. If, in contrast, the AI architecture is a neural network that consists of at least three layers, it is categorized as a deep technique [ 19 ]. It turns out that around a quarter of the studies (14/53, 26.4%) used shallow techniques, while the rest (39/53, 73.6%) used deep techniques. Within each of the groups, studies may have used different models or algorithms, and some studies proposed multiple methods or provided testing data using multiple methods. In this study, we only considered the model that had the best-reported performance in each paper.

As shown in Table 3 , most studies that used shallow techniques adopted an SVM (9/14, 64.3%), which is a common two-class classifier that uses a hyperplane as a decision boundary [ 6 ]. The rest of the studies (5/14, 35.7%) adopted the naive Bayes (NB) algorithm (1/14, 7.1%), which is a probabilistic classifier that assumes conditional independence among the features [ 6 ]; logistic regression (LR; 1/14), which uses probability for prediction; k-nearest neighbors (kNNs; 1/14), which classify a sample based on samples close to it; and random forests (RFs; 1/14), which classify using decision trees [ 6 ]. A hybrid model (1/14) classified images through multiple iteratives using Adaboost and an SVM.

Techniques used in included studies using shallow techniques (N=14).

a SVM: support vector machine.

b NB: naive Bayes.

c LR: logistic regression.

d kNN: k-nearest neighbor.

e RF: random forest.

The majority of the studies that used deep techniques ( Table 4 ) adopted different types of convolutional neural networks (CNNs; 36/39, 92.3%), which assign importance to parts of images using ImageNet-pretrained architectures (18/39, 46.2%), including the residual network (ResNet), Inception, AlexNet, MobileNet, Visual Geometry Group (VGG), Xception, DenseNet, and GoogleNet. In addition, some of the CNN-based studies (11/39, 28.2%) built customized CNNs or ResNets. Moreover, some studies adopted different combinations of CNNs along with other models (hybrid models; 5/39, 12.8%), as well as using ensemble models (4/39, 10.3%); the remaining study (1/39, 2.6%) used the OpenCV library. Multimedia Appendix 5 provides further details regarding each of the models in terms of the method used, the number of layers (ranging from 1 to 121 layers), the method used for selecting the hyperparameters, and the performance of the proposed model with respect to other reported models within the study.

Techniques used in included studies using deep techniques (N=39).

a CNN: convolutional neural network.

b ResNet: residual network.

c VGG: Visual Geometry Group.

Evaluation Metrics

The studies included in this scoping review used different evaluation metrics to assess their proposed models. In the studies, the following five primary evaluation metrics were used to assess the built models: accuracy, sensitivity and specificity, positive predictive value (PPV) or precision, area under the curve (AUC), and F1-score. All five metrics ranged from 0% to 100%; the higher the score, the better the model performance. To compute the different evaluation metrics, the following types of samples were identified: First, true positives (TPs), which are malignant samples that the AI tool also detected as malignant; second, false positives (FPs), which are benign samples that the AI tool detected as malignant; third, true negatives (TNs), which are benign samples that were also detected as benign by the AI tool; and fourth, false negatives (FNs), which are malignant samples that were detected as benign by the AI tool. It is worth mentioning that more than half of the studies (33/53, 62.3%) reported multiple evaluation metrics, in addition to the primary metric.

Accuracy = (TP + TN)/(TP + TN + FP + FN), which implies how well the model detects the diagnostic classes, was reported in the majority of the papers (44/53, 83%). Sensitivity or recall = TP/(TP + FN), which is the probability of the model, given only malignant samples, to correctly diagnose them as malignant, was reported in 30 (56.6%) papers. Specificity = TN/(TN + FP), which determines the proportion of negative samples that are correctly detected, was reported in 24 (45.3%) papers. The PPV or precision = TP/(TP + FP) was reported in 13 (24.5%) papers. The AUC, which is the area of the receiver operating characteristic (ROC) curve and plots the TP against the FP, was reported in 11 (20.8%) papers. The F1-score, which is the harmonic mean of recall and precision, was reported in 9 (16.9%) papers. In addition, the dice coefficient = 4TP/(FN + 2TP + FP) was reported in 4 (7.5%) papers. The negative predictive value (NPV) = TN/(TN + FN) was reported in 2 (3.8%) papers. The Jaccard index = 2TP/(TP + FN + FP) was reported in 2 papers. The Cohen κ was also reported in 2 papers. Finally, the Youden index = sensitivity + specificity – 1 was reported in 1 (1.9%) paper.

Herein, we conducted our analysis of each paper based on the best-performing experiment in case multiple experiments were conducted. In addition, if multiple evaluation metrics were used, we used the primary evaluation metric score that was reported by the authors in the abstract or conclusion as the main focus of the paper or the used average score of each of the diagnostic classes for multiclass classification papers. Of the aforementioned metrics, accuracy, AUC, sensitivity and specificity, and the F1-score were used as the primary evaluation metrics. Around 73% (39/53) of the papers used accuracy as their primary evaluation metric to assess the trained models. The average accuracy value was 86.8%, with a maximum of 98.8% [ 60 ] and a minimum of 67% [ 10 ]. The AUC was reported in 9 studies, with an average score of 87.2%; the highest AUC score was 91.7% [ 41 ], whereas the lowest AUC score was 82.0% [ 26 ]. Sensitivity and specificity were used in 4 studies, and the F1-score was reported in 1 study. Multimedia Appendix 6 shows the data characteristics, used model, and evaluation scores for each included study ( Table 5 ).

Primary evaluation metrics and scores reported by included studies (N=53).

a AUC: area under the curve.

Main Findings

We studied multiple characteristic types for the 53 selected studies. First, we included the study characteristics. Most studies were published in 2019, the majority of the studies were published in Southern Asia, and most studies were published in journals. Second, we discussed the data characteristics. For training and testing, most of the studies used medium-size data sets, the majority of the studies built binary classifiers, and dermoscopic images were used the most. Third, we categorized the adopted AI models into shallow and deep. Most shallow models were SVM based, whereas most deep models were CNN-based neural networks. Generally, deep models were adopted more than shallow models. Fourth, we listed the evaluation metrics used along with the reported scores to assess the performance of the models. In total, 11 different evaluation metrics were used, where accuracy was the most commonly used metric, so we focused on accuracy.

Performance Factors

After analyzing the reported performance scores, we concluded that there is a correlation between the performance and the number of classes used. In addition, another factor that affects the performance is the data set size. Next, we study this hypothesis with respect to accuracy since most of the studies (39/53, 73.6%) used it as the primary evaluation metric, although it might not be the most fitted evaluation metric to assess such a task, especially in the case of imbalanced data. We believe that having a confusion matrix or the number of TPs, FPs, TNs, and FNs would avoid bias and give a clearer evaluation of how the model behaves with regard to each of the diagnostic classes. From the studies, the top accuracy scores were ~98% [ 21 , 27 , 60 ]. In studies leading to this accuracy, the authors built a two-class classification (benign vs malignant) model using data sets of 200, 356, and 200 images, respectively. The top 10 accuracy scores (99%-92%) also built two-class classifiers using an average of around 800 images. In addition, 26 studies built two-class classifiers with an average accuracy score of around 88% using an average data set size of around 1000 images, while 17 studies built multiclass classifiers with an average accuracy score of 85%; they used around 15,000 images on average. The second-lowest accuracy score was 72% [ 23 ], in which the authors developed a multiclass classifier using 9 different diagnostic classes and 129,450 images, which is the highest number of classes and the biggest data set size included in this study. Figure 6 plots the logarithmic data set size over accuracy, using colors to indicate the number of diagnostic classes. As can be seen, accuracy increases as the number of diagnostic classes and data set size decreases. Specifically, after the threshold of 90% in accuracy, we can see that the majority of the studies built two-class classifiers. The factors that might be behind such a pattern are further discussed next.

Effect of the number of diagnostic classes and data set size on accuracy.

Classification Type Factor

Binary classifiers tend to have better performance when compared to multiclass classifiers. This seems intuitively right since binary classifiers are less expressive. Instead of distinguishing between several classes, binary classifiers have “less to learn.” To illustrate this point, let us compare limits on the probability of each class for a binary and a five-class classifier. For the five-class classifier, there must be at least one class with a probability of ≤20% (according to the pigeonhole principle [ 64 ]). Predicting this low probability class is, therefore, typically harder than in the case of a binary classifier, for which we know that there exists exactly (and, thus, at most) one class with a probability of ≤50%. Another way of looking at it is to consider an algorithm that performs a random choice assuming perfectly balanced data. In the binary case, the error rate of this algorithm would be 50%, whereas for the five-class classifier, it increases to 80%, a 1.6-fold increase. The problem may be further exacerbated by imbalanced data, which often arises naturally due to differences in the prevalence rates of medical conditions. Therefore, it is also not surprising that binary classifiers work well, given less data for training, since the model may still be fed sufficient numbers of examples for each class.

Data Set Size Factor

However, what is surprising is that Figure 6 suggests that the performance increased with decreasing training data. To this end, we would like to note that the two methods with the best performance used shallow techniques that tend to be far less hungry for data than deep methods, since manual feature engineering is often part of the pipeline. Furthermore, Afifi et al [ 21 ] used clinical image data, which may be of superior quality. In addition, depending on the testing setup, it cannot be ruled out that methods relying on less data lack the generality of models that have been trained using large volumes of data. In such scenarios, the models would be closer to data retrieval machines due to overfitting than general detectors and classifiers. To fully assess apparent issues such as this, it is important not to rely on a single performance metric when reporting results. Especially, sensitivity and specificity can be as important as accuracy in this context since they model FN and FP rates. All considered, we would, therefore, like to reiterate our earlier statement that we believe it is important for any AI to undergo rigorous clinical studies and testing before being deployed in a clinical environment.

Technique Type Factor

With regard to the techniques described in the studies included in this review, deep and shallow models (regardless of the number of layers) have similar performances. For example, within the shallow models, the top five skin cancer detectors were built using an SVM with accuracy scores of 93%-99% using relatively small data sets. The SVM was the most commonly used method among the shallow models. Similarly, within the deep models, the top five CNN-based skin cancer detectors had 94%-96% accuracy using medium-size data sets. CNNs were also the most commonly used method among the deep models. Theoretically, deep neural networks tend to have better performance with regard to image classifications [ 65 ]. One reason is that shallow models are often limited to less expressive functional spaces when compared to deep networks. From a technical perspective, this may well explain their lower performance due to a lack of the ability to fully capture the complex nature of images during training. In contrast, deep networks and CNNs can learn features at multiple scales and complexity to provide fast diagnoses [ 66 ]. Therefore, they not only detect, select, and extract features from medical images but also contribute by enhancing and constructing new features from the medical images [ 67 ]. Such similarities and inconsistencies in the performances of the included studies are due to the diverse evaluation metrics used, the data set size, image types, and the number of diagnostic classes among the studies.

Publication Year

Based on the study characteristics, we noticed that the number of published papers has increased since 2016 and that most papers discuss the use of dermoscopic images, making it the most used image modality for the detection and classification of skin cancer. We believe that this is because the International Skin Imaging Collaboration (ISIC) competition started in 2016 [ 8 ], which offered several medical data sets of dermoscopic images that have ever since been used to build AI-based models. Most of these studies are still in the development stage, and we firmly believe that these models still need to be further validated and tested in hospitals. However, dermatologists and patients are beginning to adapt to the notion of relying on AI to diagnose skin cancer.

Practical and Research Implications

In this scoping review, we summarized the findings in the literature related to diagnosing skin cancer by using AI-based technology. We also categorized the papers included in this review based on the methodology used, the type of AI techniques, and their performance, and found the link between these aspects.

We noted that although all the papers included in this scoping review discuss the application and performance of a specific AI technology, the reporting is performed heterogeneously. A discussion of the relationship between using one specific AI technique and other aspects, such as data set size, or even a discussion of why the evaluation metric used is reasonable is normally not attempted. This, of course, potentially hampers research in this direction, as it becomes harder for future studies to provide a comprehensive comparison with the existing work that follows scientific rigor. This scoping review filled this gap by performing the necessary characterizations and analyses. This was achieved by grouping each of the used AI technologies into shallow and deep approaches, linking each type to the evaluation metrics used, listing and interpreting the number of diagnostic classes used in each study, and highlighting the dependency of performance on data set size and other factors. To the best of our knowledge, no similar work has been performed to fill this gap. In the Conclusion section, we will highlight our main findings.

Limitations

This scoping review examined papers that were published between January 2009 and July 2020, and any published study outside this time line was excluded, which may have excluded older AI-based methods. In addition, we examined papers written in English; other languages were not included, which may have led to the exclusion of some studies conducted in other parts of the world. Another limitation might be the gap between the time the research was performed and the time the work was submitted, which excluded published papers during that period. Although we applied all due diligence, a small residual chance of accidentally having overlooked papers in an academic database cannot be fully ruled out. In addition, although we tried to discuss all findings in the literature, it is beyond the scope of this review to detail every single finding of the papers. Similarly, an investigation into data biases in the literature (imbalanced data with respect to diagnostic classes, patient ethnicity and skin color, gender, etc) is left as a direction for future studies.

The use of AI has high potential to facilitate the way skin cancer is diagnosed. Two main branches of AI are used to detect and classify skin cancer, namely shallow and deep techniques. However, the reliability of such AI tools is questionable since different data set sizes, image types, and number of diagnostic classes are being used and evaluated with different evaluation metrics. Accuracy is the metric used most as a primary evaluation metric but does not allow for independently assessing FN and FP rates. This study found that higher accuracy scores are reported when fewer diagnostic classes are included. Interestingly and counterintuitively, our analysis also suggests that higher accuracy scores are reported when smaller sample sizes are included, which may be due to factors such as the type of images and the techniques used. Furthermore, only independent, external validation using a large, diverse, and unbiased database is fit to demonstrate the generality and reliability of any AI technology prior to clinical deployment.

Abbreviations

Multimedia appendix 1, multimedia appendix 2, multimedia appendix 3, multimedia appendix 4, multimedia appendix 5, multimedia appendix 6.

Conflicts of Interest: None declared.

Advances in Melanoma and Other Skin Cancers Research

Metastatic melanoma cells.

NCI-funded researchers are working to advance our understanding of how to treat melanoma and other skin cancers. Much progress has been made in treating people with melanoma that has spread in their bodies ( metastatic melanoma). Yet many people still don't benefit from the newest drugs, and others may relapse after initially successful treatment.

This page highlights some of the latest research in treatment for melanoma and other skin cancers, including clinical advances that may soon translate into improved care, NCI-supported programs that are fueling progress, and current research findings from recent studies.

Melanoma Treatment

Surgery remains the standard treatment for early-stage melanoma and may also be used as part of therapy for more advanced disease. However, researchers are now focusing on developing treatments that directly target specific mutations in melanoma cells or that harness the body’s immune system to attack melanoma.

Both of these approaches— targeted therapies and immunotherapies —have led to dramatic improvements in survival for patients with advanced melanoma over the last decade. Researchers are continuing to explore ways to make these treatments more effective for more patients.

Targeted therapies

Targeted therapies use drugs or other substances to attack specific types of cancer cells with less harm to normal cells. About half of people with melanoma that has metastasized  or can’t be removed with surgery ( unresectable melanoma) have mutations in the BRAF gene . These mutations result in abnormal B-Raf proteins that can cause uncontrolled growth of melanoma cells.

Drugs have been developed that block the effects of these altered B-Raf proteins. Other new drugs block proteins that work together with altered B-Raf proteins to promote cancer cell growth. These include proteins produced by the MEK genes. The combination of blocking both B-Raf and MEK has been found to be particularly successful in treating melanoma that has a mutation in the  BRAF gene. Three such combinations are approved for people with metastatic or unresectable melanoma that has mutations in the BRAF gene:

• dabrafenib (Tafinlar)   and t rametinib (Mekinist)
• encorafenib (Braftovi)  and b inimetinib (Mektovi) 
• vemurafenib (Zelboraf)  and c obimetinib (Cotellic) 

However, although these drug combinations may be effective initially, most people develop resistance to them within a year.  Researchers are studying how melanoma cells manage to grow in the presence of these targeted therapies, with the goal of finding ways to overcome resistance. Ideas being tested include new drug combinations and drugs that target the B-Raf pathway in different ways than existing drugs.

Immune checkpoint inhibitors

Immunotherapies are treatments that help the body’s immune system fight cancer more effectively. Melanoma tends to have a relatively high number of genetic mutations that can be recognized by the immune system compared with other cancer types. This makes it more likely that melanoma will respond to immunotherapy.

One type of immunotherapy, called immune checkpoint inhibition , has shown impressive results in some people with advanced melanoma. Four immune checkpoint inhibitors are now approved for the treatment of melanoma that can’t be removed with surgery or that has metastasized:

• i pilimumab (Yervoy)
• pembrolizumab (Keytruda)
• nivolumab (Opdivo)
• atezolizumab (Tecentriq) , when used in combination with two targeted drugs

The combination of ipilimumab and nivolumab is also approved for some patients with metastatic or unresectable melanoma. In the study that led to its approval, more than half of the people who received the combination were alive 5 years after treatment . Another clinical trial showed that this combination can also shrink melanoma that has spread to the brain in some patients.

The combination of nivolumab with a new type of immune checkpoint inhibitor called relatlimab also improved the amount of time people with advanced melanoma lived without their cancer getting worse . This combination received FDA approval in 2022, under the name Opdualag , for people aged 12 or older with untreated melanoma that can't be removed surgically or has spread within the body.

Scientists are looking for ways for more people to have success with these drugs.  Unfortunately, even when used in combination, immune checkpoint inhibitors don't work for all patients with metastatic or unresectable melanoma. However, patients whose tumors do shrink or disappear often have responses that last for years. Researchers are now testing ways to increase the number of people with melanoma who benefit from this type of treatment, such as these below.

• Combining immune checkpoint inhibitors with immunostimulant s. Immunostimulants produce a type of chemical alarm in the body that tells the immune system that a threat exists. In a small clinical trial that combined pembrolizumab with an immunostimulant, tumors shrank in almost 80% of people who received the two treatments together . Larger trials of this and other combinations of immunotherapy drugs are underway.
• Testing new and existing immune checkpoint inhibitors in combination with targeted therapies and other types of drugs.
• Changing people’s gut microbes before treatment with an immune checkpoint inhibitor. For example, a study led in part by NCI researchers found that changing some people’s gut microbes could make their melanoma more likely to shrink during treatment with an immune checkpoint inhibitor.

Learning what treatments to give first

Melanoma researchers are also looking to understand how best to use existing therapies. One pressing question had been whether it is better for people who have advanced melanoma with mutations in the BRAF gene to receive targeted drugs or immune checkpoint inhibitors first. 

An NCI-sponsored trial, DREAMseq, has helped answer this question. Patients with advanced melanoma were randomly assigned to receive either a combination of targeted drugs or a combination of immune checkpoint inhibitors. When their cancer recurred, they received the other combination. The study found that more people who received the immune checkpoint inhibitor combination first were still alive 2 years later than people who received the combination of the targeted drugs first.

Researchers are also searching for biomarkers in melanoma that can predict which tumors might respond to other immunotherapies or drug combinations.

Harnessing the body’s immune cells

Adoptive cell therapy. Another type of immunotherapy, called adoptive cell therapy (ACT), is also being tested in patients with metastatic melanoma. In ACT, T cells (a type of immune cell) are given to a patient to help the body fight diseases, such as cancer.

In a small, early-phase clinical trial of ACT, about half of patients with metastatic melanoma saw their tumors shrink, and a quarter remained in remission for as long as the study tracked them—in some cases for up to a decade. But the procedure is complicated and expensive, and half of people do not benefit from the treatment at all or experience dangerous or even fatal side effects. Researchers are looking for ways to make ACT work for more patients such as the examples below.

• One idea being tested is the use of immune cells that have been collected from patients, altered to make them better at killing cancer cells, and then infused back into patients. Such therapies include CAR T cells, a type of treatment where a patient's T cells are changed in the lab so they will attack cancer cells . Researchers are also testing other ways to boost the ability of T cells to kill tumor cells.
• Another idea is to find common proteins that are present in many people's tumors. This could allow for the creation of “off-the-shelf” T-cell therapies that don’t have to be made on a custom basis for each patient.

Immunotherapy following surgery for advanced melanoma

Adjuvant therapy is additional cancer treatment given after primary surgical treatment. Nivolumab, ipilimumab, and pembrolizumab have all been approved as adjuvant therapies for melanoma that has spread to nearby lymph nodes but can be removed with surgery. In clinical trials, all three immune checkpoint inhibitors reduced the risk of recurrence for some patients when given after surgery, although many patients experienced serious side effects.

Another study tested pembrolizumab in patients with early-stage melanoma that has not spread to the lymph nodes but had a high risk of doing so. It found that giving pembrolizumab after surgery reduced the chance of the cancer coming back or spreading elsewhere in the body . However, the treatment can cause significant side effects. More studies are needed to understand how to identify the people with this type of high-risk, early-stage melanoma who would benefit the most from such treatment. 

Researchers are also exploring whether immune checkpoint inhibitors might be more effective if given before surgery. One NCI-sponsored trial compared the outcomes of patients with melanoma at high risk of recurring who receive pembrolizumab both before and after surgery with those in patients who receive the drug only after surgery. That trial found that people who received the drug both before and after surgery had a substantially lower risk of their cancer coming back than those who only received adjuvant treatment.

Rare Melanoma Types

Some rare types of melanoma have lagged behind melanoma of the skin in terms of advances in treatment. These include intraocular (uveal) melanoma , which starts in the eye; desmoplastic melanoma , a rare form of melanoma of the skin; and mucosal melanoma, which begins in the mucosal membranes , such as the linings of the nose and mouth.

However, recent small clinical trials suggest that these types of melanoma may also respond to immunotherapies. One NCI-sponsored trial  tested pembrolizumab in people with desmoplastic melanoma . Initial results from this trial showed that the drug shrinks both tumors that can be removed surgically and those that cannot. The trial participants are still being tracked to see if pembrolizumab improves how long they live overall.

Immune checkpoint inhibitors have been less effective in intraocular melanoma than in other types of melanoma. However, a different type of immunotherapy called a bispecific fusion protein has shown promise for treating this rare cancer. These drugs bind to melanoma cells and the body’s own immune cells at the same time, to bring them together. This allows the immune cells to kill the melanoma cells. In a clinical trial, one such drug, called tebentafusp, became the first drug to show an improvement in overall survival for patients with metastatic intraocular melanoma.

Merkel Cell Carcinoma

Another rare type of skin cancer, called Merkel cell carcinoma (MCC), has been shown to be the most sensitive of any tumor type to treatment with a single immune checkpoint inhibitor. In 2017, an immunotherapy called avelumab (Bavencio) received the first-ever FDA approval for a drug to treat MCC. In addition, more than half of patients with MCC in a small clinical trial had their tumors shrink or disappear during treatment with pembrolizumab, which received FDA approval for the treatment of MCC in 2018.

In 2023, a third immunotherapy drug called retifanlimab (Zynyz) received FDA approval for the treatment of MCC that has recurred or spread elsewhere in the body. Other immunotherapy drugs are currently being tested in this rare cancer type.

Treatment for Advanced Basal Cell Carcinoma and Squamous Cell Carcinoma

Basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) of the skin are the most common cancers in the United States. They rarely spread to other organs and are seldom fatal. However, every year many people are diagnosed with advanced BCC or SCC.

Recent breakthroughs in targeted therapies and immunotherapies have changed the way people with advanced BCC and SCC are treated. Ongoing research seeks to build on these breakthroughs such as:

• The targeted drugs sonidegib (Odomzo) and vismodegib (Erivedge)  which can control tumors for a long time in some people with BCC. However, resistance often develops. In addition, side effects can cause some patients who need to take the drugs for a long time to stop taking them. Researchers are now looking for ways to change when and how much of these drugs are given, both to delay the development of resistance and to make them easier to tolerate.
• cemiplimab (Libtayo)  for some people with metastatic or locally advanced SCC  that can't be removed with surgery. Cemiplimab is also being tested as a treatment given before surgery for some people whose cancer can be removed.
• pembrolizumab for some people with recurrent or metastatic SCC
• cemiplimab for some people with advanced BCC whose tumors have become resistant to targeted therapy

New clinical trials are now testing other immunotherapy drugs and combinations in SCC and BCC. 

For people whose BCC or SCC has not spread, surgery remains the mainstay of treatment. But less-intensive versions of radiation therapy have been developed for people who can’t tolerate surgery for larger tumors, such as the frail elderly.

NCI-Supported Research Programs

Many NCI-funded researchers at the NIH campus, and across the United States and world, are seeking ways to address melanoma and other skin cancers more effectively. Some research is basic, exploring questions as diverse as the biological underpinnings of cancer and the social factors that affect cancer risk. And some is more clinical, seeking to translate this basic information into improving patient outcomes. The programs listed below are a small sampling of NCI’s research efforts for melanoma and other skin cancers.

Scientists in the  Division of Cancer Epidemiology and Genetics (DCEG) study families in which multiple members have developed certain cancers. In collaboration with the Melanoma Genetics Consortium (GenoMEL), DCEG researchers are searching for new genes in both melanoma-prone families and through a genome-wide association study to find genes that may increase the risk of melanoma.

The  Skin Specialized Programs of Research Excellence (Skin SPOREs)  are designed to quickly move basic scientific findings into clinical settings. The Skin SPORE program’s main focus is on melanoma research activities, but it also includes projects in other skin cancer types, such as cutaneous T-cell lymphoma.

NCI's National Clinical Trials Network (NCTN) is a collection of organizations and clinicians that coordinates and supports cancer clinical trials at more than 3,000 sites across the United States and Canada. NCTN currently has a variety of trials testing treatments for skin cancer .

The Division of Cancer Control and Population Sciences (DCCPS) oversees the Cancer Trends Progress Report, an online report that tracks the nation's progress against cancer from prevention through end of life. Topics in the report that inform melanoma and skin cancer research are sun-protective behavior , indoor and outdoor tanning , and sunburn . The division’s Health Behaviors Research Branch (HBRB) supports research in the area of sun protection and reducing indoor tanning practices, through both measurement and intervention studies.

Clinical Trials

NCI funds and oversees both early- and late-stage clinical trials to develop new treatments and improve patient care. Trials are available for melanoma prevention and treatment and non-melanoma skin cancer prevention and treatment .

Melanoma and Other Skin Cancers Research Results

The following are some of our latest news articles about research on melanoma and other skin cancers:

• First Cancer TIL Therapy Gets FDA Approval for Advanced Melanoma
• Rare Melanoma Very Likely to Respond to Treatment with Pembrolizumab
• Immunotherapy before Surgery Appears Effective for Some with Melanoma
• Androgen Receptor May Explain Sex Differences in Melanoma Treatment Response
• Study Adds to Debate about Screening for Melanoma
• Opdualag Becomes First FDA-Approved Immunotherapy to Target LAG-3

View the full list of Melanoma and Other Skin Cancers Research Results and Study Updates .

Skin Cancer: Description, Causes, and Treatment Research Paper

Skin cancer is one of the most common types of cancer; the three most common types of skin cancer are basal cell carcinoma, squamous cell carcinoma, and melanoma. Skin cancer incidences gradually increased in the last decades, presenting a significant threat to the population’s well-being (Cameron et al., 2019). Skin cancer is characterized by an uncontrollable growth of skin cells, during which they could spread to other human body parts and cause damage. According to Cameron et al. (2019), a higher percentage of risk of developing skin cancer (20% to 30%) is associated with the white population. The review conducted by Kim et al. (2019) suggests that skin cancer prognosis could be connected with light eye color and freckles combined with red or blonde hair color. Family history of skin cancer and continuous exposure to direct sunlight, photosensitizing drugs, or carcinogenic chemicals also contribute to skin cancer development. Millions of people are diagnosed with nonmelanoma skin cancer in a span of one year, and mortality rates are estimated at 0.12 per 100,000 cases (Kim et al., 2019). In general, the main risk of developing skin cancer is UV radiation and exposure to sunlight.

Dermatologists or physician assistants could diagnose skin cancer through biopsy, which allows fast and accurate results. Depending on the size and shape of the tumor, the diagnosis could be performed either through a punch biopsy or a shave biopsy, designed for larger areas of skin. Moreover, shave biopsy allows a more accurate result due to the decreased chance of sampling error (Cameron et al., 2019). Non-invasive options for skin cancer diagnosis include optical methods, such as coherence tomography and reflectance confocal microscopy (Cameron et al., 2019). Both methods operate based on infrared light projection and could also be used to provide an accurate result.

Treatment of skin cancer is primarily focused on the local excision of tumors. However, the excision does not guarantee the full elimination of disease as recurrences could occur significantly later after the initial treatment. According to Kim et al. (2019), recurrence rates or surgical excision are between 3 to 12 percent of cases, and they mostly take place more than five years post-treatment. Therefore, besides the initial treatment, the necessary measures also include follow-up checkups.

Currently, there are many available options for skin cancer treatment. Surgical excision is recommended for tumors located in neck and trunk areas. Incomplete excisions in surgical treatment could result in recurrence in approximately 38 percent of cases (Kim et al., 2019). Mohs surgery is recommended for the treatment of high-risk tumors and recurrent skin cancer. On the other hand, for low-risk tumors, treatment measures could be faster and more cost-effective, with methods such as electrodesiccation and curettage (Kim et al., 2019). One of the treatment procedures developed recently for low-risk skin cancer tumors is cryosurgery, which focuses on freezing the surrounding margin of the tumor. Overall, the choice of treatment is based on the size of the tumor, its location area, and available equipment.

As skin cancer is associated with UV radiation, it is recommended that the population, especially those with a higher risk of developing skin cancer, take preventative measures. The preventive methods include reducing time spent in direct sunlight, wearing protective clothing and equipment, and using sunscreen products that reduce the harm from UV radiation. Lastly, it is necessary to educate the population on the importance of regular self-skin examination and prompt turn to professionals for a diagnosis to prevent adverse outcomes.

Cameron, M. C., Lee, E., Hibler, B., Giordano, C. N., Barker, C. A., Mori, S., Cordova, M., Nehal, K. S., & Rossi, A. M. (2019). Basal cell carcinoma: Contemporary approaches to diagnosis, treatment, and prevention. Journal of the American Academy of Dermatology, 80 (2), 321-339. Web.

Kim, D. P., Kus, K. J. B., & Ruiz, E. (2019). Basal cell carcinoma review. Hematology/Oncology Clinics of North America, 33 (1), 13–24. Web.

• Chicago (A-D)
• Chicago (N-B)

IvyPanda. (2022, December 21). Skin Cancer: Description, Causes, and Treatment. https://ivypanda.com/essays/skin-cancer-description-causes-and-treatment/

"Skin Cancer: Description, Causes, and Treatment." IvyPanda , 21 Dec. 2022, ivypanda.com/essays/skin-cancer-description-causes-and-treatment/.

IvyPanda . (2022) 'Skin Cancer: Description, Causes, and Treatment'. 21 December.

IvyPanda . 2022. "Skin Cancer: Description, Causes, and Treatment." December 21, 2022. https://ivypanda.com/essays/skin-cancer-description-causes-and-treatment/.

1. IvyPanda . "Skin Cancer: Description, Causes, and Treatment." December 21, 2022. https://ivypanda.com/essays/skin-cancer-description-causes-and-treatment/.

Bibliography

IvyPanda . "Skin Cancer: Description, Causes, and Treatment." December 21, 2022. https://ivypanda.com/essays/skin-cancer-description-causes-and-treatment/.

• Breast Cells: Histopathology
• Carcinoma in Situ Management Options
• Malignant Melanoma: Diagnosis and Treatment
• Medical Anthropology. Female Genital Mutilation
• Tetanus Infection, Disease and Treatment
• Basal Cell Carcinoma - Symptoms and Causes
• Histopathology: Epidermal Changes
• Hepatocellular Carcinoma Diagnosis and Treatment
• Humanities and Ethical Issues: Reflection
• Hepatocellular Carcinoma: Case Study Assignment
• Colon Cancer: Symptoms, Genes, and Immunosuppression
• Acute Lymphoblastic Leukemia: Causes, Origin, and Gene Mutation
• Researching the Colon Cancer and Nursing
• Researching of Cause and Effects of Melanoma
• Cancer Terminology and Characteristics

We use cookies to enhance our website for you. Proceed if you agree to this policy or learn more about it.

• Essay Database >
• Essay Examples >
• Essays Topics >
• Essay on Melanoma

Example Of Skin Cancer Research Paper

Type of paper: Research Paper

Topic: Melanoma , Skin Cancer , Nation , Cancer , Smoking , Treatment , Skin , Carcinoma

Words: 1000

Published: 01/27/2020

ORDER PAPER LIKE THIS

Skin cancers are cancers that affect the skin. These include melanoma, basal and squamous cell cancers. The symptoms include skin color change and appearance of skin ulcers, as well as, alterations in the moles that are on the skin. Smoking and ultraviolet radiations are some of the leading causes of skin cancer. The condition may be treated through radiations, chemotherapy among other means.

Introduction

Skin cancers refer to the types of cancer that affect the skin. These cancers include the basal cell cancer, squamous cell cancer and melanoma. The basal cell cancer is the type of cancer that arises from the middle layer, has more likelihood of spreading, and may lead to death if not treated. The melanoma type of skin cancer originates in the cells that are involved with the produce the skin pigment known as malanocytes. Lack of treatment may cause the cancer to be fatal. Skin cancer has become among the common cancer types in US (National Cancer Institute). The basal cell carcinoma mainly appears on the skin parts that are usually exposed to the sun with the face being the most affected part. The basal cell carcinoma metastasizes on rare occasions and do cause death on rare occasions. The cancer is easily treated using surgical operations or by the use of radiations. The squamous cell carcinoma, on the other hand, is not as common as the basal cell carcinoma, but they metastasize in a more frequent way than the basal cell carcinoma. Other than the squamous cell carcinoma affecting the ear, lips, and in patients who are immunosuppressed, the other forms of squamous cell carcinoma metastasize at a low rate. The melanoma is the only form of skin cancer that is least frequent. The melanomas metastasize in a more frequent way and have a potential to cause death when they spread. The malignant melanoma appears asymmetrical with irregular borders. It has varying colorations and has a diameter of more than 6 mm (Swetter). There are different signs and symptoms that appear during the incidents of skin cancer (National Cancer Institute). The symptoms include skin color change and appearance of skin ulcers, as well as, alterations in the moles that are on the skin. The basal cell carcinoma is presented as a smooth, pearly bump that is raised and appears on the portions of the skin that have a direct exposure to the sun. These parts include parts of the head, shoulder, and neck. In some cases, the small blood vessels appear even in the absence of the tumor. The center of the tumor may experience bleeding and clustering, and in most cases, it is confused for a sore that is not healing (Rockoff). Squamous cell carcinoma appears as a red, thickened patch and appeared after the skin has been exposed to the sun. They are red in color with some being firm hard nodules with a dome like shape. There may be bleeding and ulceration and may develop to be a large mass if not treated. On the other side, the symptoms of melanoma include being brown to black while others are red, pink or fleshy in color. Some of the warming signs of associated with melanoma are changes in shape, size as well as, the mole elevation (Rockoff). The leading of skin cancer is the ultraviolet radiations that originate from the sun. Other factors that may result to skin cancer include smoking, infection by the HPV genetic syndromes as well as wounds that are not healing. Artificial UV radiations, ionizing radiations, aging, and lack of dark color on the skin are other causes for skin cancer (Saladi and Persaud). Treatment of skin cancer is usually dependent on cancer type, age of the patient, cancer location, as well as, whether the incident of cancer is the first one, or it is a recurrence. Some of the methods employed in the treatment of skin cancers are surgical excision, radiation, cryosurgery, curettage and desiccation, and Mohs micrographic surgery (National Cancer Institute). In the case of low-risk disease, subjecting the cells to an external beam of radiation, or freezing the cells may offer adequate cure for controlling the disease. Some of the drugs used in chemotherapy management of skin cancer management include 5-fluorouracil and imiquimod. Generally, most of the non-melanoma cases respond well to radiations than melanoma cancers. Most of the non-melanoma skin cancer are curable and can hit a success rate of 100% if the incident is presented to a physician before they can spread (Rockoff). Preventive measures such are use of Sunscreen have been shown to be effective in preventing both melanoma and squamous cell carcinoma. However, there is minimal evidence that sunscreen is effective in basal cell melanoma prevention. Other protective measures include the use of protective clothing, sunglasses as well as reducing the periods that one is exposed to the sun. There is also need to have regular surveillance especially for those people who are at a higher risk of skin cancer like those with light skin color. The surveillance may be dome through self-examination or by undergoing a physical examination in a clinic. Those people who have ever had skin cancer incidents should seek regular medical examinations to help in early detection of a new form of cancer or reoccurrence of a previous incident (Rockoff).

Three types of skin cancer have been identified, melanoma, basal and squamous cell cancers. Smoking and ultraviolet radiations are some of the leading causes of skin cancer. The conditions usually managed through a number of ways including radiations and chemotherapy.

Works Cited

National Cancer Institute. Skin Cancer. 2013. 16 March 2013. . Rockoff, Alan. Skin Cancer. 2013. 16 March 2013. . Saladi, R. N and A. N. Persaud. "The causes of skin cancer: a comprehensive review." Drugs of Today 41.5 (2005): 37–53. Swetter, Susan M. Cutaneous Melanoma. 2012. 16 March 2013. .

Cite this page

Share with friends using:

Removal Request

Finished papers: 913

This paper is created by writer with

ID 281775813

If you want your paper to be:

Well-researched, fact-checked, and accurate

Original, fresh, based on current data

Eloquently written and immaculately formatted

275 words = 1 page double-spaced

Get your papers done by pros!

Other Pages

Ice research papers, pressurization essays, quicksand essays, daimler essays, hansom essays, banjo essays, clef essays, airman essays, grados essays, composing essays, the french revolution essays, peruse essays, ordinary men essays, prater essays, cree essays, confrontation essays, adversaries essays, tomahawk essays, mastering essays, progressive party essays, new freedom essays, new nationalism essays, whitening essays, pigment essays, chlorophyll essays, medical facilities essays, peaks essays, haute couture essays, example of social marketing essay, choice of the letter formats course work, example of google analytics 20 essay, results report examples, free article review on determination of eligibility special education, the no child left behind act has increased academic standards in school research paper, example of essay on algebra projects, breaking the da vinci code essay sample, cloning research paper sample, free evaluation of visual material essay example, good example of the super anti hero representing the zeitgeist of western culture thesis proposal, sample creative writing on understanding terrorism, good literature review about media and the national development, free top reasons for companies going international essay example, an evaluation of the advantages and disadvantages of using case study comparative and historical approaches as research methods case study example.

Password recovery email has been sent to [email protected]

Use your new password to log in

You are not register!

By clicking Register, you agree to our Terms of Service and that you have read our Privacy Policy .

Now you can download documents directly to your device!

Check your email! An email with your password has already been sent to you! Now you can download documents directly to your device.

or Use the QR code to Save this Paper to Your Phone

The sample is NOT original!

Short on a deadline?

Don't waste time. Get help with 11% off using code - GETWOWED

No, thanks! I'm fine with missing my deadline

Sun Exposure And Skin Cancer Research Paper

View sample Sun Exposure And Skin Cancer Research Paper. Browse other  research paper examples and check the list of research paper topics for more inspiration. If you need a religion research paper written according to all the academic standards, you can always turn to our experienced writers for help. This is how your paper can get an A! Feel free to contact our custom writing service for professional assistance. We offer high-quality assignments for reasonable rates.

Sun exposure can be defined as the exposure of human beings to natural and artificial ultraviolet radiation (UVR). UVR is a component of natural sunlight and is produced artificially (e.g., by sunbeds) for cosmetic (e.g., to get tanned skin) or medical reasons (e.g., therapy of psoriasis). Sun exposure is the most important environmental risk factor for all forms of skin cancer (Marks 1995). Sun protection behavior, on the other hand, is the most effective form of skin cancer prevention. This research paper provides an overview of the incidence and causation of skin cancer, of the psychological conditions for sun exposure and sun protection behavior, and of skin cancer prevention programs.

Academic Writing, Editing, Proofreading, And Problem Solving Services

Get 10% off with 24start discount code, 1. skin cancer: incidence and causation.

The two major classes of skin cancer are malignant melanoma (MM) and nonmelanoma skin cancers (NSC; subtypes: basal cell carcinoma, squamous cell carcinoma). In many countries, skin cancer is the cancer with the most rapidly increasing incidence rate. In the United States, the incidence rates of skin cancer are 12.0 per 100,000 Caucasians for MM (mortality rates: men 3.3 per 100,000 people/year, women 1.7) and 236.6 per 100,000 Caucasians for NSC (mortality rates: men 0.7, women 0.3) (Marks 1995). Incidence rates for people with darker skin are much lower.

The development of skin cancer depends on constitutional and environmental factors (for an overview, see Marks 1995). The major constitutional factors are skin color, skin reaction to strong sunlight, and, additionally for MM, large numbers of moles, freckles, and family history of MM. The major environmental risk factor for skin cancer is sun exposure. Cumulative sun exposure is related to the development of NSC. For MM, the role of sun exposure is less clear. Increased risk of MM seems to be related to severe sunburns, in particular in childhood, and intermittent recreational sun exposure.

2. Predictors Of Sun Exposure

Sun exposure not only accompanies many working and recreational outdoor activities, it is also seeked intentionally (for an overview of studies on psychosocial factors of sun exposure, see Arthey and Clarke 1995, Eid and Schwenkmezger 1997). The major motivational predictor for unprotected sun exposure is the desire for a suntan. Level of suntan is related to the perception of other people as being healthy and attractive in a curvilinear manner, with a medium tan considered the healthiest and most attractive skin color. Therefore, getting a suntan helps people to make a good impression on others. The belief that a suntan improves one’s physical appearance is the strongest predictor of suntanning. Voluntary sun exposure is related to thinking more about one’s physical appearance, to a higher fear of negative evaluations by others, and to a higher body self-consciousness and body self-esteem. In addition to these variables of personal appearance, suntanning is related to a positive attitude toward risk taking in general and to having friends who sunbathe and use sunbeds.

3. Predictors Of Sun Protection Behavior

The harmful effects of UVR on the skin can be prevented by sun protection behavior. The four most important protection behaviors are: (a) Avoiding direct sunlight, particularly during the peak hours of daylight (10 am to 4 pm), (b) seeking shade, (c) wearing a hat, (d) wearing skin-covering clothing, and (e) using sunscreen with a sun protection factor of at least 15. Even in high-risk countries such as Australia, the prevalence rates of these protective behaviors are rather low indicating that about half of the population are inadequately protected on sunny days (e.g., Lower et al. 1998). Voluntary sun exposure and the use of sun protection are two largely independent and only weakly correlated behaviors. Moreover, the predictors for sun protection behavior differ from the predictors for sun exposure. In contrast to sun exposure, appearance-oriented variables are less important for predicting sun protection behavior. The major predictors for sun protection behavior are perceived threat of skin cancer, the benefits and barriers of different sun protection behaviors, social factors, and knowledge about skin cancer (Eid and Schwenkmezger 1997, Hill et al. 1984).

Perceived threat of skin cancer depends on the perceived severity of and the vulnerability to skin cancer. People who rate skin cancer as a severe disease and who think that they are more vulnerable to skin cancer are more likely to use sun protection. Furthermore, individuals with a sensitive and fair skin, both which are more prone to skin cancer, use sun protection more often. This risk perception is strongly related to gender. Women have stronger vulnerability and severity beliefs. Moreover, women use sunscreen more often and apply sunscreen with a higher sun protection factor than men.

The actual display of specific sun protection behaviors depends on (a) the belief that these behaviors are effective in preventing skin damage, sunburns, and skin cancer, and (b) the specific barriers of each behavior (Hill et al. 1984). Specific barriers to the application of sunscreen is the belief that sunscreen is greasy and sticky and the fact that sunscreen has to be applied repeatedly. An additional barrier for men is the belief that sunscreen makes men appear seemingly sissy and looks unattractive. Barriers against wearing hats are the beliefs that hats are inconvenient to wear as well as causes problems with hairstyle. Additionally, for men, wearing a hat is associated negatively with the beliefs that wearing a hat causes baldness and a sweaty head, is inconvenient on windy days, and gets in the way when playing sports. The major barriers against wearing shirts are the discomfort from heat and the feeling of being overdressed. Therefore, a goal of skin cancer prevention programs must be the removal of these specific barriers against sun protection behavior. Because these barriers partly depend on temporary fashion influences, the present-day barriers of specific sun protection behaviors have to be examined before a prevention campaign is planned.

Social factors also have an important influence on sun protection. Use of sun protection is correlated with peers’ sun protection behavior, parental influence, and parental sun protection behavior. Furthermore, people who know a person suffering from skin cancer use sun protection more often. Finally, knowledge about the risks of sun exposure has been proven a significant predictor for sun protection in many studies. In particular, people with more knowledge about skin cancer use sunscreen with a higher sun protection factor.

4. Skin Cancer Prevention

With respect to the target groups considered, sun exposure and sun protection modification programs can be grouped into three classes: (a) programs using mass media, (b) community intervention programs, and (c) educational programs (for an overview of these programs, see Eid and Schwenkmezger 1997, Loescher et al. 1995, Morris and Elwood 1996, Rossi et al. 1995).

4.1 Mass Media Programs

Typical mass media programs use informational pamphlets, comic books, newspaper reports, TV and radio advertisements, or videotapes to carry their messages. These mass media campaigns are either often not evaluated, or the evaluations are published in reports that are not accessible to the public, or the evaluations suffer from methodological problems (e.g., missing control group). However, there are some studies evaluating single videotapes or informational pamphlets by an experimental or control-group design (for an overview see Eid and Schwenkmezger 1997, Morris and Elwood 1996). According to these studies, videotapes and informational pamphlets can increase knowledge about, perceived severity of, and vulnerability to skin cancer, as well as increase individual sun protection intentions. Studies focusing on long-term effects on behavior, however, are missing. Furthermore, these studies show that an emotional prevention video imparting knowledge by employing a person with skin cancer might be more appropriate than a more unemotional presentation of facts. Moreover, educational texts focusing on the negative effects of sun exposure for physical appearance (wrinkles, skin aging) might be more appropriate than texts on the negative health-related consequences. This, however, might only be true for people with a low appearance orientation, whereas appearance-based messages can result in boomerang effects for high appearance oriented people. Finally, the manner in which a health message is framed is important. Detweiler et al. (1999) demonstrated that messages on sun protection behavior are more effective when they highlight potential positive effects of the displayed sun protection behavior (gain-framed messages) than when they focus on the negative effects of the omitted behavior (loss-framed messages).

4.2 Community Intervention Programs

Mass media are often a component of community intervention programs, but community intervention programs can implement several other methods as well. Community-based programs can make use of local peculiarities (e.g., the daily UVR rate in a community) and can focus on local places of risk behavior, for example, swimming pools and beaches. Based on principles of learning theory, Lombard et al. (1991) designed an intervention program with pool lifeguards as models for sun protection behaviors and daily feedback posters of the sun protection behavior displayed the day before. This program was effective with respect to two sun protection behaviors (staying in the shade, wearing shirts) but not other ones. Rossi et al. (1995) describe several new techniques such as special mirrors (sun scanner) and photographs (sun damage instant photography) that have been used in intervention programs on beaches to show the effects of sun damage and photoaging in a more dramatic fashion.

4.3 Educational Programs

Because of the cumulative risk of sun exposure and the risks of sunburn in childhood, intervention programs are particularly important for children and adolescents. Educational programs have been developed for students of different ages from preschool to high school. In general, these programs are effective in enhancing knowledge about skin cancer and the awareness of the risks of unprotected sun exposure. Changing attitudes toward sun protection, however, have only been found after participation in programs consisting of more than one session. With respect to changes in behavioral intentions the results are inconsistent. Significant changes in intentions have been found in only a few studies and not consistently across the intentions to different behaviors. Changes in observed behavior, however, have not been analyzed. The failure of educational programs in changing behavioral intentions might be due to the fact that these programs have focused more on motivational than on behavioral factors. Multimethod educational programs including components on the implementation of behavioral routines that are very important for developing health behavior in childhood are desiderata for future research.

5. Future Directions

Sun exposure and skin cancer prevention are relatively new fields of research. Future research will profit from the use of multimethod assessment strategies of behavior that does not only assess self-reported behavioral intentions but also uses observational and physical measures of behavior (e.g., spectrophotometers for measuring the melanin skin content). Furthermore, research on new components of intervention programs is needed. For example, future programs could consist of behavior change methods in addition to attitude change methods as well as strategies for weakening the association between suntan and attractivity.

Bibliography:

• Arthey S, Clarke V A 1995 Suntanning and sun protection: A review of the psychological literature. Social Science Medicine 40: 265–74
• Detweiler J B, Bedell B T, Salovey P, Pronin E, Rothman A 1999 Message framing and sunscreen use: Gain-framed messages motivate beach-goers. Health Psychology 18: 189–96
• Eid M, Schwenkmezger P 1997 Sonnenschutzverhalten (Sun protection behavior). In: Schwarzer R (ed.) Gesundheitspsychologie (Health Psychology). Hogrefe, Gottingen
• Hill D, Rassaby J, Gardner G 1984 Determinants of intentions to take precautions against skin cancer. Community Health Studies 8: 33–44
• Loescher L J, Buller M K, Buller D B, Emerson J, Taylor A M 1995 Public education projects in skin cancer. Cancer Supplement 75: 651–56
• Lombard D, Neubauer T E, Canfield D, Winett R 1991 Behavioral community intervention to reduce the risk of skin cancer. Journal Applied Behavioral Analysis 4: 677–86
• Lower T, Grigis A, Sanson-Fisher R 1998 The prevalence and predictors of solar protection use among adolescents. Preventative Medicine 27: 391–9
• Marks R 1995 An overview of skin cancers: Incidence and causation. Cancer Supplement 75: 607–12
• Morris J, Elwood M 1996 Sun exposure modification programmes and their evaluation: A review of the literature. Health Promotion International 11: 321–32
• Rossi J S, Blais L M, Redding C A, Weinstock M A 1995 Preventing skin cancer through behavior change. Implications for interventions. Dermatol. Clin. 13: 613–22

ORDER HIGH QUALITY CUSTOM PAPER