Systematic review of dermoscopy and digital dermoscopy/ artificial intelligence for the diagnosis of melanoma

Diagnostic Accuracy of Artificial Intelligence-Based Automated Diabetic Retinopathy Screening in Real-World Settings: A Systematic Review and Meta-Analysis

PURPOSE

To evaluate the diagnostic accuracy of artificial intelligence (AI)-based automated diabetic retinopathy (DR) screening in real-world settings.

DESIGN

Systematic review and meta-analysis METHODS: We conducted a systematic review of relevant literature from January 2012 to August 2022 using databases including PubMed, Scopus and Web of Science. The quality of studies was evaluated using Quality Assessment for Diagnostic Accuracy Studies 2 (QUADAS-2) checklist. We calculated pooled accuracy, sensitivity, specificity, and diagnostic odds ratio (DOR) as summary measures. The study protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO - CRD42022367034).

RESULTS

We included 34 studies which utilized AI algorithms for diagnosing DR based on real-world fundus images. Quality assessment of these studies indicated a low risk of bias and low applicability concern. Among gradable images, the overall pooled accuracy, sensitivity, specificity, and DOR were 81%, 94% (95% CI: 92.0-96.0), 89% (95% CI: 85.0-92.0) and 128 (95% CI: 80-204) respectively. Sub-group analysis showed that, when acceptable quality imaging could be obtained, non-mydriatic fundus images had a better DOR of 143 (95% CI: 82-251) and studies using 2 field images had a better DOR of 161 (95% CI 74-347). Our meta-regression analysis revealed a statistically significant association between DOR and variables such as the income status, and the type of fundus camera.

CONCLUSION

Our findings indicate that AI algorithms have acceptable performance in screening for DR using fundus images compared to human graders. Implementing a fundus camera with AI-based software has the potential to assist ophthalmologists in reducing their workload and improving the accuracy of DR diagnosis.

Dermoscopy Differential Diagnosis Explorer (D3X) Ontology to Aggregate and Link Dermoscopic Patterns to Differential Diagnoses: Development and Usability Study

BACKGROUND

Dermoscopy is a growing field that uses microscopy to allow dermatologists and primary care physicians to identify skin lesions. For a given skin lesion, a wide variety of differential diagnoses exist, which may be challenging for inexperienced users to name and understand.

OBJECTIVE

In this study, we describe the creation of the dermoscopy differential diagnosis explorer (D3X), an ontology linking dermoscopic patterns to differential diagnoses.

METHODS

Existing ontologies that were incorporated into D3X include the elements of visuals ontology and dermoscopy elements of visuals ontology, which connect visual features to dermoscopic patterns. A list of differential diagnoses for each pattern was generated from the literature and in consultation with domain experts. Open-source images were incorporated from DermNet, Dermoscopedia, and open-access research papers.

RESULTS

D3X was encoded in the OWL 2 web ontology language and includes 3041 logical axioms, 1519 classes, 103 object properties, and 20 data properties. We compared D3X with publicly available ontologies in the dermatology domain using a semiotic theory-driven metric to measure the innate qualities of D3X with others. The results indicate that D3X is adequately comparable with other ontologies of the dermatology domain.

CONCLUSIONS

The D3X ontology is a resource that can link and integrate dermoscopic differential diagnoses and supplementary information with existing ontology-based resources. Future directions include developing a web application based on D3X for dermoscopy education and clinical practice.

Basic principles of artificial intelligence in dermatology explained using melanoma

The use of artificial intelligence (AI) continues to establish itself in the most diverse areas of medicine at an increasingly fast pace. Nevertheless, many healthcare professionals lack the basic technical understanding of how this technology works, which severely limits its application in clinical settings and research. Thus, we would like to discuss the functioning and classification of AI using melanoma as an example in this review to build an understanding of the technology behind AI. For this purpose, elaborate illustrations are used that quickly reveal the technology involved. Previous reviews tend to focus on the potential applications of AI, thereby missing the opportunity to develop a deeper understanding of the subject matter that is so important for clinical application. Malignant melanoma has become a significant burden for healthcare systems. If discovered early, a better prognosis can be expected, which is why skin cancer screening has become increasingly popular and is supported by health insurance. The number of experts remains finite, reducing their availability and leading to longer waiting times. Therefore, innovative ideas need to be implemented to provide the necessary care. Thus, machine learning offers the ability to recognize melanomas from images at a level comparable to experienced dermatologists under optimized conditions.

Grundprinzipien der künstlichen Intelligenz in der Dermatologie erklärt am Beispiel des Melanoms

Der Einsatz von künstlicher Intelligenz (KI) setzt sich in den verschiedensten Bereichen der Medizin immer schneller durch. Dennoch fehlt vielen medizinischen Kollegen das technische Grundverständnis für die Funktionsweise dieser Technologie, was ihre Anwendung in Klinik und Forschung stark einschränkt. Daher möchten wir in dieser Übersichtsarbeit die Funktionsweise und Klassifizierung der KI am Beispiel des Melanoms erörtern, um ein Verständnis für die Technologie hinter der KI zu schaffen. Dazu werden ausführliche Illustrationen verwendet, die die Technologie schnell erklären. Bisherige Übersichten konzentrieren sich eher auf die potenziellen Anwendungen der KI und verpassen die Gelegenheit, ein tieferes Verständnis für die Materie herauszuarbeiten, das für die klinische Anwendung so wichtig ist. Das maligne Melanom ist zu einer erheblichen Belastung für die Gesundheitssysteme geworden. Bei frühzeitiger Entdeckung ist eine bessere Prognose zu erwarten, weshalb das Hautkrebs‐Screening immer populärer und von den Krankenkassen unterstützt wird. Die Zahl der Fachärzte ist jedoch begrenzt, was ihre Verfügbarkeit einschränkt und zu längeren Wartezeiten führt. Daher müssen innovative Ideen umgesetzt werden, um die notwendige Versorgung zu gewährleisten. Das maschinelle Lernen bietet die Möglichkeit, Melanome auf Bildern zu erkennen, und zwar auf einem Niveau, das mit dem von erfahrenen Dermatologen – unter optimierten Bedingungen – vergleichbar ist.

The State of Artificial Intelligence in Skin Cancer Publications

BACKGROUND

Artificial intelligence (AI) in skin cancer is a promising research field to assist physicians and to provide support to patients remotely. Physicians' awareness to new developments in AI research is important to define the best practices and scope of integrating AI-enabled technologies within a clinical setting.

OBJECTIVES

To analyze the characteristics and trends of AI skin cancer publications from dermatology journals.

METHODS

AI skin cancer publications were retrieved in June 2022 from the Web of Science. Publications were screened by title, abstract, and keywords to assess eligibility. Publications were fully reviewed. Publications were divided between nonmelanoma skin cancer (NMSC), melanoma, and skin cancer studies. The primary measured outcome was the number of citations. The secondary measured outcomes were articles' general characteristics and features related to AI.

RESULTS

A total of 168 articles were included: 25 on NMSC, 77 on melanoma, and 66 on skin cancer. The most common types of skin cancers were melanoma (134, 79.8%), basal cell carcinoma (61, 36.3%), and squamous cell carcinoma (45, 26.9%). All articles were published between 2000 and 2022, with 49 (29.2%) of them being published in 2021. Original studies that developed or assessed an algorithm predominantly used supervised learning (66, 97.0%) and deep neural networks (42, 67.7%). The most used imaging modalities were standard dermoscopy (76, 45.2%) and clinical images (39, 23.2%).

CONCLUSIONS

Most publications focused on developing or assessing screening technologies with mainly deep neural network algorithms. This indicates the eminent need for dermatologists to label or annotate images used by novel AI systems.

The Use of Artificial Intelligence in the Liver Histopathology Field: A Systematic Review

Digital pathology (DP) has begun to play a key role in the evaluation of liver specimens. Recent studies have shown that a workflow that combines DP and artificial intelligence (AI) applied to histopathology has potential value in supporting the diagnosis, treatment evaluation, and prognosis prediction of liver diseases. Here, we provide a systematic review of the use of this workflow in the field of hepatology. Based on the PRISMA 2020 criteria, a search of the PubMed, SCOPUS, and Embase electronic databases was conducted, applying inclusion/exclusion filters. The articles were evaluated by two independent reviewers, who extracted the specifications and objectives of each study, the AI tools used, and the results obtained. From the 266 initial records identified, 25 eligible studies were selected, mainly conducted on human liver tissues. Most of the studies were performed using whole-slide imaging systems for imaging acquisition and applying different machine learning and deep learning methods for image pre-processing, segmentation, feature extractions, and classification. Of note, most of the studies selected demonstrated good performance as classifiers of liver histological images compared to pathologist annotations. Promising results to date bode well for the not-too-distant inclusion of these techniques in clinical practice.

The Reporting and Methodological Quality of Systematic Reviews Underpinning Clinical Practice Guidelines Focused on the Management of Cutaneous Melanoma: Cross-Sectional Analysis

BACKGROUND

Clinical practice guidelines (CPGs) inform evidence-based decision-making in the clinical setting; however, systematic reviews (SRs) that inform these CPGs may vary in terms of reporting and methodological quality, which affects confidence in summary effect estimates.

OBJECTIVE

Our objective was to appraise the methodological and reporting quality of the SRs used in CPGs for cutaneous melanoma and evaluate differences in these outcomes between Cochrane and non-Cochrane reviews.

METHODS

We conducted a cross-sectional analysis by searching PubMed for cutaneous melanoma guidelines published between January 1, 2015, and May 21, 2021. Next, we extracted SRs composing these guidelines and appraised their reporting and methodological rigor using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) and AMSTAR (A Measurement Tool to Assess Systematic Reviews) checklists. Lastly, we compared these outcomes between Cochrane and non-Cochrane SRs. All screening and data extraction occurred in a masked, duplicate fashion.

RESULTS

Of the SRs appraised, the mean completion rate was 66.5% (SD 12.29%) for the PRISMA checklist and 44.5% (SD 21.05%) for AMSTAR. The majority of SRs (19/50, 53%) were of critically low methodological quality, with no SRs being appraised as high quality. There was a statistically significant association (P<.001) between AMSTAR and PRISMA checklists. Cochrane SRs had higher PRISMA mean completion rates and higher methodological quality than non-Cochrane SRs.

CONCLUSIONS

SRs supporting CPGs focused on the management of cutaneous melanoma vary in reporting and methodological quality, with the majority of SRs being of low quality. Increasing adherence to PRISMA and AMSTAR checklists will likely increase the quality of SRs, thereby increasing the level of evidence supporting cutaneous melanoma CPGs.

A systematic review of clinical health conditions predicted by machine learning diagnostic and prognostic models trained or validated using real-world primary health care data

With the advances in technology and data science, machine learning (ML) is being rapidly adopted by the health care sector. However, there is a lack of literature addressing the health conditions targeted by the ML prediction models within primary health care (PHC) to date. To fill this gap in knowledge, we conducted a systematic review following the PRISMA guidelines to identify health conditions targeted by ML in PHC. We searched the Cochrane Library, Web of Science, PubMed, Elsevier, BioRxiv, Association of Computing Machinery (ACM), and IEEE Xplore databases for studies published from January 1990 to January 2022. We included primary studies addressing ML diagnostic or prognostic predictive models that were supplied completely or partially by real-world PHC data. Studies selection, data extraction, and risk of bias assessment using the prediction model study risk of bias assessment tool were performed by two investigators. Health conditions were categorized according to international classification of diseases (ICD-10). Extracted data were analyzed quantitatively. We identified 106 studies investigating 42 health conditions. These studies included 207 ML prediction models supplied by the PHC data of 24.2 million participants from 19 countries. We found that 92.4% of the studies were retrospective and 77.3% of the studies reported diagnostic predictive ML models. A majority (76.4%) of all the studies were for models' development without conducting external validation. Risk of bias assessment revealed that 90.8% of the studies were of high or unclear risk of bias. The most frequently reported health conditions were diabetes mellitus (19.8%) and Alzheimer's disease (11.3%). Our study provides a summary on the presently available ML prediction models within PHC. We draw the attention of digital health policy makers, ML models developer, and health care professionals for more future interdisciplinary research collaboration in this regard.

Differentiating between early melanomas and melanocytic nevi: A state-of-the-art review

Clinicians and dermatologists are challenged by accurate diagnosis of melanocytic lesions, due to melanoma's resemblance to benign skin conditions. Several methodologies have been proposed to diagnose melanoma, and to differentiate between a cancerous and a benign skin condition. First, the ABCD rule and Menzies method use skin lesion characteristics to interpret the condition. The 7-point checklist, 3-point checklist, and CASH algorithm are score-based methods. Each of these methods attributes a score point to the features found on the skin lesion. Furthermore, reflectance confocal microscopy (RCM), an integrated clinical and dermoscopic risk scoring system (iDscore), and a deep convoluted neural network (DCNN) also aids in diagnosis. RCM optically sections live tissues to reveal morphological and cellular structures. The skin lesion's clinical parameters determine iDscore's score point system. The DCNN model is based on a detailed learning algorithm. Therefore, we discuss the conventional and new methodologies for the identification of skin diseases. Moreover, our review attempts to provide clinicians with a comprehensible summary of the wide range of techniques that can help differentiate between early melanomas and melanocytic nevi.

The accuracy of artificial intelligence used for non-melanoma skin cancer diagnoses: a meta-analysis

BACKGROUND

With rising incidence of skin cancer and relatively increased mortality rates, an improved diagnosis of such a potentially fatal disease is of vital importance. Although frequently curable, it nevertheless places a considerable burden upon healthcare systems. Among the various types of skin cancers, non-melanoma skin cancer is most prevalent. Despite such prevalence and its associated cost, scant proof concerning the diagnostic accuracy via Artificial Intelligence (AI) for non-melanoma skin cancer exists. This study meta-analyzes the diagnostic test accuracy of AI used to diagnose non-melanoma forms of skin cancer, and it identifies potential covariates that account for heterogeneity between extant studies.

METHODS

Various electronic databases (Scopus, PubMed, ScienceDirect, SpringerLink, and Dimensions) were examined to discern eligible studies beginning from March 2022. Those AI studies predictive of non-melanoma skin cancer were included. Summary estimates of sensitivity, specificity, and area under receiver operating characteristic curves were used to evaluate diagnostic accuracy. The revised Quality Assessment of Diagnostic Studies served to assess any risk of bias.

RESULTS

A literature search produced 39 eligible articles for meta-analysis. The summary sensitivity, specificity, and area under receiver operating characteristic curve of AI for diagnosing non-melanoma skin cancer was 0.78, 0.98, & 0.97, respectively. Skin cancer typology, data sources, cross validation, ensemble models, types of techniques, pre-trained models, and image augmentation became significant covariates accounting for heterogeneity in terms of both sensitivity and/or specificity.

CONCLUSIONS

Meta-analysis results revealed that AI is predictive of non-melanoma with an acceptable performance, but sensitivity may become improved. Further, ensemble models and pre-trained models are employable to improve true positive rating.

Melanin Stacking Differences in Pigmented and Non-Pigmented Melanomas: Quantitative Differentiation between Pigmented and Non-Pigmented Melanomas Based on Light-Scattering Properties

Cutaneous melanoma is a cancer with metastatic potential characterized by varying amounts of pigment-producing melanocytes, and it is one of the most aggressive and fatal forms of skin malignancy, with several hundreds of thousands of cases each year. Early detection and therapy can lead to decreased morbidity and decreased cost of therapy. In the clinic, this often translates to annual skin screenings, especially for high-risk patients, and generous use of the ABCDE (asymmetry, border irregularity, color, diameter, evolving) criteria. We have used a new technique termed vibrational optical coherence tomography (VOCT) to non-invasively differentiate between pigmented and non-pigmented melanomas in a pilot study. The VOCT results reported in this study indicate that both pigmented and non-pigmented melanomas have similar characteristics, including new 80, 130, and 250 Hz peaks. Pigmented melanomas have larger 80 Hz peaks and smaller 250 Hz peaks than non-pigmented cancers. The 80 and 250 Hz peaks can be used to quantitative characterize differences between different melanomas. In addition, infrared light penetration depths indicated that melanin in pigmented melanomas has higher packing densities than in non-pigmented lesions. Using machine learning techniques, the sensitivity and specificity of differentiating skin cancers from normal skin are shown to range from about 78% to over 90% in this pilot study. It is proposed that using AI on both lesion histopathology and mechanovibrational peak heights may provide even higher specificity and sensitivity for differentiating the metastatic potential of different melanocytic lesions.

Smartphone-based single snapshot spatial frequency domain imaging

We report a handheld, smartphone-based spatial frequency domain imaging device. We first examined the linear dynamic range of the smartphone camera sensor. We then calculated optical properties for a series of liquid phantoms with varying concentrations of nigrosin ink and Intralipid, demonstrating separation of absorption and scattering. The device was then tested on a human wrist, where optical properties and hemoglobin-based chromophores were calculated. Finally, we performed an arterial occlusion on a human hand and captured hemodynamics using our device. We hope to lay the foundation for an accessible SFDI device with mass-market appeal designed for dermatological and cosmetic applications.

Artificial intelligence for the automated single-shot assessment of psoriasis severity

BACKGROUND

PASI score is globally used to assess disease activity of psoriasis. However, it is relatively complicated and time-consuming, and the score will vary due to the inconsistent subjectivity between dermatologists. Therefore, an AI system capable of assessing psoriasis severity will be useful.

OBJECTIVES

To propose a simplified PASI system (Single-Shot PASI) and associated AI models capable of assessing psoriasis severity.

METHODS

705 psoriasis images of the trunk's front and back were used in our research. Considering the relatively small number of images, we used data augmentation techniques to expand the data. A psoriasis expert's scores were used as teacher data. Various convolutional neural network models and hyperparameters were adjusted using a five-fold cross validation. From these adjustments, we discovered that fine-tuning Imagenet2012-pretrained InceptionV3 whose last linear layer was replaced by two-layer perceptron (30 hidden units and five output units) exhibited the best performance.

RESULTS

To validate our deep learning system, 10 images were selected as test sets and were excluded from the training sets. The AI assessment of Single-Shot PASI was almost consistent with the clinical severity. We examined whether AI assistance would affect human scoring. 13 dermatologists and nine medical students were invited as evaluators. Mean absolute differences from AI scores and standard deviation among evaluators reduced with AI assistance. In addition, the evaluator's scores got close to the teacher's score with AI's assistance.

CONCLUSIONS

We proposed a Single-Shot PASI system and developed associated AI system capable of assessing psoriasis severity simply by uploading a single clinical image. An easy-to-use scoring system and our freely available AI software would help dermatologists and patients with psoriasis.

Predictive ability of current machine learning algorithms for type 2 diabetes mellitus: A meta-analysis

AIMS/INTRODUCTION

Recently, an increasing number of cohort studies have suggested using machine learning (ML) to predict type 2 diabetes mellitus. However, its predictive ability remains inconclusive. This meta-analysis evaluated the current ability of ML algorithms for predicting incident type 2 diabetes mellitus.

MATERIALS AND METHODS

We systematically searched longitudinal studies published from 1 January 1950 to 17 May 2020 using MEDLINE and EMBASE. Included studies had to compare ML's classification with the actual incidence of type 2 diabetes mellitus, and present data on the number of true positives, false positives, true negatives and false negatives. The dataset for these four values was pooled with a hierarchical summary receiver operating characteristic and a bivariate random effects model.

RESULTS

There were 12 eligible studies. The pooled sensitivity, specificity, positive likelihood ratio and negative likelihood ratio were 0.81 (95% confidence interval [CI] 0.67-0.90), 0.82 [95% CI 0.74-0.88], 4.55 [95% CI 3.07-6.75] and 0.23 [95% CI 0.13-0.42], respectively. The area under the summarized receiver operating characteristic curve was 0.88 (95% CI 0.85-0.91).

CONCLUSIONS

Current ML algorithms have sufficient ability to help clinicians determine whether individuals will develop type 2 diabetes mellitus in the future. However, persons should be cautious before changing their attitude toward future diabetes risk after learning the result of the diabetes prediction test using ML algorithms.

Quality assessment standards in artificial intelligence diagnostic accuracy systematic reviews: a meta-research study

Artificial intelligence (AI) centred diagnostic systems are increasingly recognised as robust solutions in healthcare delivery pathways. In turn, there has been a concurrent rise in secondary research studies regarding these technologies in order to influence key clinical and policymaking decisions. It is therefore essential that these studies accurately appraise methodological quality and risk of bias within shortlisted trials and reports. In order to assess whether this critical step is performed, we undertook a meta-research study evaluating adherence to the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool within AI diagnostic accuracy systematic reviews. A literature search was conducted on all studies published from 2000 to December 2020. Of 50 included reviews, 36 performed the quality assessment, of which 27 utilised the QUADAS-2 tool. Bias was reported across all four domains of QUADAS-2. Two hundred forty-three of 423 studies (57.5%) across all systematic reviews utilising QUADAS-2 reported a high or unclear risk of bias in the patient selection domain, 110 (26%) reported a high or unclear risk of bias in the index test domain, 121 (28.6%) in the reference standard domain and 157 (37.1%) in the flow and timing domain. This study demonstrates the incomplete uptake of quality assessment tools in reviews of AI-based diagnostic accuracy studies and highlights inconsistent reporting across all domains of quality assessment. Poor standards of reporting act as barriers to clinical implementation. The creation of an AI-specific extension for quality assessment tools of diagnostic accuracy AI studies may facilitate the safe translation of AI tools into clinical practice.

Diagnosing onychomycosis: A step forward?

BACKGROUND AND AIMS

There are a number of available methods for diagnosing onychomycosis, but more emerge as technology advances. This review briefly discusses the common diagnostic methods, the use of artificial intelligence (AI) as a diagnostic tool in dermatology as a whole, and then examines research on the use of AI for diagnosing onychomycosis. The studies discussed implemented convolutional neural networks (CNNs) to examine datasets of images of entire nails or histological images and then used the information learned from those datasets to make a diagnostic decision of onychomycosis or not.

RESULTS

Results: It was found that, on average, AI were able to diagnose onychomycosis from the images provided at an equivalent level as human dermatologists. However, there are a number of clear limitations for using AI in this manner. The AI models implemented relied solely on images and therefore were limited by image quality. As only images were examined, other clinical data were not taken into consideration, which could be important to the diagnostic outcome.

CONCLUSION

Conclusion: In conclusion, although AI can be a very helpful tool in the diagnostic process by increasing efficiency and reducing costs, it still requires the precision and expertise of professional dermatologists to be used optimally.

Dermatoscopy

The dermatoscope has gained tremendous popularity among dermatologists as an adjunctive tool to better visualize subsurface structures and identify patterns that may improve the diagnosis of a wide range of skin diseases. Initially, the pigmented lesion experts who were the early adopters promoted the use of the dermatoscope to increase diagnostic accuracy of early melanomas and decrease the harvesting of benign lesions. With current near universal adoption of the diagnostic technique by dermatologists, the dermatoscope is now employed to help identify a wide variety of inflammatory, infectious, and vascular conditions of the skin, hair, and nails, resulting in the emergence of several branches of dermatoscopy-inflammoscopy, trichoscopy, onychoscopy, and entodermoscopy. The future of dermatoscopy will involve incorporation of artificial intelligence that will make the assessment process increasingly objective, more accurate, and universally available. Despite the wide acceptance and adoption of dermatoscopy, the overall impact of its widespread use still remains unclear, whether it has decreased biopsy rates of benign lesions, reduced health care costs, or improved patient outcomes.

Artificial intelligence in dermatology for the clinician

As medicine enters the era of artificial intelligence (AI)-augmented practice, dermatology is beginning to witness the integration of AI into the daily practice, particularly in the areas of diagnosis, prognosis, and treatment of skin diseases. Many of the current electronic medical records that dermatologists have incorporated provide guidance in billing, a form of AI at work. The recent advances in visual recognition AI make application and integration of the technology particularly suited for perceptual specialties such as radiology and dermatology. In dermatology, AI is poised to improve the efficiency and accuracy of traditional diagnostic approaches, including visual examination, skin biopsy, and histopathologic examination. This review highlights the current progress of AI in dermatology and provides a basic overview of the technology.

Artificial intelligence in dermatology

Dermatology and medicine are producing data at an increasing rate that are progressively difficult to sort and manage. Artificial intelligence (AI) and machine learning are examples of tools that may have the capability to produce significant and meaningful results from these data. Currently, AI and machine learning have a variety of applications in medicine including, but not limited to, diagnostics, patient management, preventive medicine, and genomic analysis. Although the role of AI in dermatology is greater than ever, its use is still extremely limited. As AI is continually developed and implemented, it is essential that stakeholders understand AI terminology, applications, limitations, and projected uses in dermatology. With the continued development of AI technology, however, its implementation may afford greater dermatologist efficiency, greater increased patient access to dermatologic care, and improved patient outcomes.

Acral melanoma detection using dermoscopic images and convolutional neural networks

Acral melanoma (AM) is a rare and lethal type of skin cancer. It can be diagnosed by expert dermatologists, using dermoscopic imaging. It is challenging for dermatologists to diagnose melanoma because of the very minor differences between melanoma and non-melanoma cancers. Most of the research on skin cancer diagnosis is related to the binary classification of lesions into melanoma and non-melanoma. However, to date, limited research has been conducted on the classification of melanoma subtypes. The current study investigated the effectiveness of dermoscopy and deep learning in classifying melanoma subtypes, such as, AM. In this study, we present a novel deep learning model, developed to classify skin cancer. We utilized a dermoscopic image dataset from the Yonsei University Health System South Korea for the classification of skin lesions. Various image processing and data augmentation techniques have been applied to develop a robust automated system for AM detection. Our custom-built model is a seven-layered deep convolutional network that was trained from scratch. Additionally, transfer learning was utilized to compare the performance of our model, where AlexNet and ResNet-18 were modified, fine-tuned, and trained on the same dataset. We achieved improved results from our proposed model with an accuracy of more than 90 % for AM and benign nevus, respectively. Additionally, using the transfer learning approach, we achieved an average accuracy of nearly 97 %, which is comparable to that of state-of-the-art methods. From our analysis and results, we found that our model performed well and was able to effectively classify skin cancer. Our results show that the proposed system can be used by dermatologists in the clinical decision-making process for the early diagnosis of AM.

The Classification of Six Common Skin Diseases Based on Xiangya-Derm: Development of a Chinese Database for Artificial Intelligence

Background

Skin and subcutaneous disease is the fourth-leading cause of the nonfatal disease burden worldwide and constitutes one of the most common burdens in primary care. However, there is a severe lack of dermatologists, particularly in rural Chinese areas. Furthermore, although artificial intelligence (AI) tools can assist in diagnosing skin disorders from images, the database for the Chinese population is limited.

Objective

This study aims to establish a database for AI based on the Chinese population and presents an initial study on six common skin diseases.

Methods

Each image was captured with either a digital camera or a smartphone, verified by at least three experienced dermatologists and corresponding pathology information, and finally added to the Xiangya-Derm database. Based on this database, we conducted AI-assisted classification research on six common skin diseases and then proposed a network called Xy-SkinNet. Xy-SkinNet applies a two-step strategy to identify skin diseases. First, given an input image, we segmented the regions of the skin lesion. Second, we introduced an information fusion block to combine the output of all segmented regions. We compared the performance with 31 dermatologists of varied experiences.

Results

Xiangya-Derm, as a new database that consists of over 150,000 clinical images of 571 different skin diseases in the Chinese population, is the largest and most diverse dermatological data set of the Chinese population. The AI-based six-category classification achieved a top 3 accuracy of 84.77%, which exceeded the average accuracy of dermatologists (78.15%).

Conclusions

Xiangya-Derm, the largest database for the Chinese population, was created. The classification of six common skin conditions was conducted based on Xiangya-Derm to lay a foundation for product research.

Machine Learning and Deep Learning Methods for Skin Lesion Classification and Diagnosis: A Systematic Review

Computer-aided systems for skin lesion diagnosis is a growing area of research. Recently, researchers have shown an increasing interest in developing computer-aided diagnosis systems. This paper aims to review, synthesize and evaluate the quality of evidence for the diagnostic accuracy of computer-aided systems. This study discusses the papers published in the last five years in ScienceDirect, IEEE, and SpringerLink databases. It includes 53 articles using traditional machine learning methods and 49 articles using deep learning methods. The studies are compared based on their contributions, the methods used and the achieved results. The work identified the main challenges of evaluating skin lesion segmentation and classification methods such as small datasets, ad hoc image selection and racial bias.

Role of Artificial Intelligence Applications in Real-Life Clinical Practice: Systematic Review

BACKGROUND

Artificial intelligence (AI) applications are growing at an unprecedented pace in health care, including disease diagnosis, triage or screening, risk analysis, surgical operations, and so forth. Despite a great deal of research in the development and validation of health care AI, only few applications have been actually implemented at the frontlines of clinical practice.

OBJECTIVE

The objective of this study was to systematically review AI applications that have been implemented in real-life clinical practice.

METHODS

We conducted a literature search in PubMed, Embase, Cochrane Central, and CINAHL to identify relevant articles published between January 2010 and May 2020. We also hand searched premier computer science journals and conferences as well as registered clinical trials. Studies were included if they reported AI applications that had been implemented in real-world clinical settings.

RESULTS

We identified 51 relevant studies that reported the implementation and evaluation of AI applications in clinical practice, of which 13 adopted a randomized controlled trial design and eight adopted an experimental design. The AI applications targeted various clinical tasks, such as screening or triage (n=16), disease diagnosis (n=16), risk analysis (n=14), and treatment (n=7). The most commonly addressed diseases and conditions were sepsis (n=6), breast cancer (n=5), diabetic retinopathy (n=4), and polyp and adenoma (n=4). Regarding the evaluation outcomes, we found that 26 studies examined the performance of AI applications in clinical settings, 33 studies examined the effect of AI applications on clinician outcomes, 14 studies examined the effect on patient outcomes, and one study examined the economic impact associated with AI implementation.

CONCLUSIONS

This review indicates that research on the clinical implementation of AI applications is still at an early stage despite the great potential. More research needs to assess the benefits and challenges associated with clinical AI applications through a more rigorous methodology.

Cutaneous Malignant Melanoma: A Review of Early Diagnosis and Management

Cutaneous melanoma (CM) is a malignant tumor formed from pigment-producing cells called melanocytes. It is one of the most aggressive and fatal forms of skin malignancy. In the last decades, CM's incidence has gradually risen, with 351,880 new cases in 2015. Since the 1960s, its incidence has increased steadily, in 2019, with approximately 96,000 new cases. A greater understanding of early diagnosis and management of CM is urgently needed because of the high mortality rates due to metastatic melanoma. Timely detection of melanoma is crucial for successful treatment, but diagnosis with histopathology may also pose a significant challenge to this objective. Early diagnosis and management are essential and contribute to better survival rates of the patient. To better control this malignancy, such information is expected to be particularly useful in the early detection of possible metastatic lesions and the development of new therapeutic approaches. This article reviews the available information on the early diagnosis and management of CM and discusses such information's potential in facilitating the future prospective.

A Point-of-Care, Real-Time Artificial Intelligence System to Support Clinician Diagnosis of a Wide Range of Skin Diseases

Dermatological diagnosis remains challenging for nonspecialists because the morphologies of primary skin lesions widely vary from patient to patient. Although previous studies have used artificial intelligence (AI) to classify lesions as benign or malignant, there have not been extensive studies examining the use of AI on identifying and categorizing a primary skin lesion's morphology. In this study, we evaluate the performance of a standalone AI tool to correctly categorize a skin lesion's morphology from a test bank of images. To provide a marker of performance, we evaluate the accuracy of primary care physicians to categorize skin lesion morphology in the same test bank of images without any aids and then with the aid of a simple visual guide. The AI system achieved an accuracy of 68% in determining the single most likely morphology from the test image bank. When the AI's top prediction was broadened to its top three most likely predictions, accuracy improved to 80%. In comparison, the diagnostic accuracy of primary care physicians was 36% without any aids and 68% with the visual guide (P < 0.001). The AI was subsequently tested on an additional set of 222 heterogeneous images of varying Fitzpatrick skin types and achieved an overall accuracy of 70% in the Fitzpatrick I-III skin type group and 68% in the Fitzpatrick IV-VI skin type group (P = 0.79). An AI is a powerful tool to assist physicians in the diagnosis of skin lesions while still requiring the user to critically consider other possible diagnoses.

Artificial Intelligence in Health Care: Bibliometric Analysis

Background

As a critical driving power to promote health care, the health care–related artificial intelligence (AI) literature is growing rapidly.

Objective

The purpose of this analysis is to provide a dynamic and longitudinal bibliometric analysis of health care–related AI publications.

Methods

The Web of Science (Clarivate PLC) was searched to retrieve all existing and highly cited AI-related health care research papers published in English up to December 2019. Based on bibliometric indicators, a search strategy was developed to screen the title for eligibility, using the abstract and full text where needed. The growth rate of publications, characteristics of research activities, publication patterns, and research hotspot tendencies were computed using the HistCite software.

Results

The search identified 5235 hits, of which 1473 publications were included in the analyses. Publication output increased an average of 17.02% per year since 1995, but the growth rate of research papers significantly increased to 45.15% from 2014 to 2019. The major health problems studied in AI research are cancer, depression, Alzheimer disease, heart failure, and diabetes. Artificial neural networks, support vector machines, and convolutional neural networks have the highest impact on health care. Nucleosides, convolutional neural networks, and tumor markers have remained research hotspots through 2019.

Conclusions

This analysis provides a comprehensive overview of the AI-related research conducted in the field of health care, which helps researchers, policy makers, and practitioners better understand the development of health care–related AI research and possible practice implications. Future AI research should be dedicated to filling in the gaps between AI health care research and clinical applications.

The use of noninvasive imaging techniques in the diagnosis of melanoma: a prospective diagnostic accuracy study

BACKGROUND

Early detection of melanoma is crucial to improving the detection of thin curable melanomas. Noninvasive, computer-assisted methods have been developed to use at the bedside to aid in diagnoses but have not been compared directly in a clinical setting.

OBJECTIVE

We conducted a prospective diagnostic accuracy study comparing a dermatologist's clinical examination at the bedside, teledermatology, and noninvasive imaging techniques (FotoFinder, MelaFind, and Verisante Aura).

METHODS

A total of 184 patients were recruited prospectively from an outpatient dermatology clinic, with lesions imaged, assessed, and excised. Skin specimens were assessed by 2 blinded pathologists, providing the gold standard comparison.

RESULTS

Fifty-nine lesions from 56 patients had a histopathologic diagnosis of melanoma, whereas 150 lesions from 128 patients were diagnosed as benign. Sensitivities and specificities were, respectively, MelaFind (82.5%, 52.4%), Verisante Aura (21.4%, 86.2%), and FotoFinder Moleanalyzer Pro (88.1%, 78.8%). The sensitivity and specificity of the teledermoscopist (84.5% and 82.6%, respectively) and local dermatologist (96.6% and 32.2%, respectively) were also compared.

LIMITATIONS

There are inherent limitations in using pathology as the gold standard to compare sensitivities and specificities.

CONCLUSION

This study demonstrates that the highest sensitivity and specificity of the instruments were established with the FotoFinder Moleanalyzer Pro, which could be a valuable tool to assist with, but not replace, clinical decision making.

Stigma, biomarkers, and algorithmic bias: recommendations for precision behavioral health with artificial intelligence

Effective implementation of artificial intelligence in behavioral healthcare delivery depends on overcoming challenges that are pronounced in this domain. Self and social stigma contribute to under-reported symptoms, and under-coding worsens ascertainment. Health disparities contribute to algorithmic bias. Lack of reliable biological and clinical markers hinders model development, and model explainability challenges impede trust among users. In this perspective, we describe these challenges and discuss design and implementation recommendations to overcome them in intelligent systems for behavioral and mental health.

Detection of Malignant Melanoma Using Artificial Intelligence: An Observational Study of Diagnostic Accuracy

Background

Malignant melanoma can most successfully be cured when diagnosed at an early stage in the natural history. However, there is controversy over screening programs and many advocate screening only for high-risk individuals.

Objectives

This study aimed to evaluate the accuracy of an artificial intelligence neural network (Deep Ensemble for Recognition of Melanoma [DERM]) to identify malignant melanoma from dermoscopic images of pigmented skin lesions and to show how this compared to doctors' performance assessed by meta-analysis.

Methods

DERM was trained and tested using 7,102 dermoscopic images of both histologically confirmed melanoma (24%) and benign pigmented lesions (76%). A meta-analysis was conducted of studies examining the accuracy of naked-eye examination, with or without dermoscopy, by specialist and general physicians whose clinical diagnosis was compared to histopathology. The meta-analysis was based on evaluation of 32,226 pigmented lesions including 3,277 histopathology-confirmed malignant melanoma cases. The receiver operating characteristic (ROC) curve was used to examine and compare the diagnostic accuracy.

Results

DERM achieved a ROC area under the curve (AUC) of 0.93 (95% confidence interval: 0.92-0.94), and sensitivity and specificity of 85.0% and 85.3%, respectively. Avoidance of false-negative results is essential, so different decision thresholds were examined. At 95% sensitivity DERM achieved a specificity of 64.1% and at 95% specificity the sensitivity was 67%. The meta-analysis showed primary care physicians (10 studies) achieve an AUC of 0.83 (95% confidence interval: 0.79-0.86), with sensitivity and specificity of 79.9% and 70.9%; and dermatologists (92 studies) 0.91 (0.88-0.93), 87.5%, and 81.4%, respectively.

Conclusions

DERM has the potential to be used as a decision support tool in primary care, by providing dermatologist-grade recommendation on the likelihood of malignant melanoma.

The Possibility of Deep Learning-Based, Computer-Aided Skin Tumor Classifiers

The incidence of skin tumors has steadily increased. Although most are benign and do not affect survival, some of the more malignant skin tumors present a lethal threat if a delay in diagnosis permits them to become advanced. Ideally, an inspection by an expert dermatologist would accurately detect malignant skin tumors in the early stage; however, it is not practical for every single patient to receive intensive screening by dermatologists. To overcome this issue, many studies are ongoing to develop dermatologist-level, computer-aided diagnostics. Whereas, many systems that can classify dermoscopic images at this dermatologist-equivalent level have been reported, a much fewer number of systems that can classify conventional clinical images have been reported thus far. Recently, the introduction of deep-learning technology, a method that automatically extracts a set of representative features for further classification has dramatically improved classification efficacy. This new technology has the potential to improve the computer classification accuracy of conventional clinical images to the level of skilled dermatologists. In this review, this new technology and present development of computer-aided skin tumor classifiers will be summarized.

Artificial Intelligence Versus Clinicians in Disease Diagnosis: Systematic Review

Background

Artificial intelligence (AI) has been extensively used in a range of medical fields to promote therapeutic development. The development of diverse AI techniques has also contributed to early detections, disease diagnoses, and referral management. However, concerns about the value of advanced AI in disease diagnosis have been raised by health care professionals, medical service providers, and health policy decision makers.

Objective

This review aimed to systematically examine the literature, in particular, focusing on the performance comparison between advanced AI and human clinicians to provide an up-to-date summary regarding the extent of the application of AI to disease diagnoses. By doing so, this review discussed the relationship between the current advanced AI development and clinicians with respect to disease diagnosis and thus therapeutic development in the long run.

Methods

We systematically searched articles published between January 2000 and March 2019 following the Preferred Reporting Items for Systematic reviews and Meta-Analysis in the following databases: Scopus, PubMed, CINAHL, Web of Science, and the Cochrane Library. According to the preset inclusion and exclusion criteria, only articles comparing the medical performance between advanced AI and human experts were considered.

Results

A total of 9 articles were identified. A convolutional neural network was the commonly applied advanced AI technology. Owing to the variation in medical fields, there is a distinction between individual studies in terms of classification, labeling, training process, dataset size, and algorithm validation of AI. Performance indices reported in articles included diagnostic accuracy, weighted errors, false-positive rate, sensitivity, specificity, and the area under the receiver operating characteristic curve. The results showed that the performance of AI was at par with that of clinicians and exceeded that of clinicians with less experience.

Conclusions

Current AI development has a diagnostic performance that is comparable with medical experts, especially in image recognition-related fields. Further studies can be extended to other types of medical imaging such as magnetic resonance imaging and other medical practices unrelated to images. With the continued development of AI-assisted technologies, the clinical implications underpinned by clinicians’ experience and guided by patient-centered health care principle should be constantly considered in future AI-related and other technology-based medical research.

The use of in vivo reflectance confocal microscopy for the diagnosis of melanoma

INTRODUCTION

The use of reflectance confocal microscopy (RCM) for imaging the skin non-invasively raised constantly during the last decade. One of the main field of application is skin cancer diagnosis, and in particular melanoma diagnosis. Several studies have investigated the diagnostic accuracy of RCM as compared to dermoscopic examination, and its value in enhancing early diagnosis of dermoscopic difficult melanomas. Areas covered: The purpose of this paper was to review the principles behind RCM image acquisition as well as to describe and discuss key RCM features of melanoma. Moreover, we conducted a literature search in order to highlight the current available evidence about RCM sensitivity and specificity in the diagnosis of melanoma. Expert commentary: During the last decade, we assisted at the increasing interest in non invasive imaging tools for the diagnosis of skin cancer. RCM is one of the most studied of a series of diagnostic methods that are emerging in the field of melanoma imaging. Most probably in the future, RCM will be more frequently available in tertiary referral centres, thus the knowledge of the pros and contra of the tool and its clinical applicability is of upmost importance in order to allow correct referrals with the final aim of improving diagnostic accuracy.

Automated detection of nonmelanoma skin cancer using digital images: a systematic review

Background

Computer-aided diagnosis of skin lesions is a growing area of research, but its application to nonmelanoma skin cancer (NMSC) is relatively under-studied. The purpose of this review is to synthesize the research that has been conducted on automated detection of NMSC using digital images and to assess the quality of evidence for the diagnostic accuracy of these technologies.

Methods

Eight databases (PubMed, Google Scholar, Embase, IEEE Xplore, Web of Science, SpringerLink, ScienceDirect, and the ACM Digital Library) were searched to identify diagnostic studies of NMSC using image-based machine learning models. Two reviewers independently screened eligible articles. The level of evidence of each study was evaluated using a five tier rating system, and the applicability and risk of bias of each study was assessed using the Quality Assessment of Diagnostic Accuracy Studies tool.

Results

Thirty-nine studies were reviewed. Twenty-four models were designed to detect basal cell carcinoma, two were designed to detect squamous cell carcinoma, and thirteen were designed to detect both. All studies were conducted in silico. The overall diagnostic accuracy of the classifiers, defined as concordance with histopathologic diagnosis, was high, with reported accuracies ranging from 72 to 100% and areas under the receiver operating characteristic curve ranging from 0.832 to 1. Most studies had substantial methodological limitations, but several were robustly designed and presented a high level of evidence.

Conclusion

Most studies of image-based NMSC classifiers report performance greater than or equal to the reported diagnostic accuracy of the average dermatologist, but relatively few studies have presented a high level of evidence. Clinical studies are needed to assess whether these technologies can feasibly be implemented as a real-time aid for clinical diagnosis of NMSC.

Electronic supplementary material

The online version of this article (10.1186/s12880-019-0307-7) contains supplementary material, which is available to authorized users.

Optical coherence tomography for diagnosing skin cancer in adults

BACKGROUND

Early accurate detection of all skin cancer types is essential to guide appropriate management and to improve morbidity and survival. Melanoma and squamous cell carcinoma (SCC) are high-risk skin cancers, which have the potential to metastasise and ultimately lead to death, whereas basal cell carcinoma (BCC) is usually localised, with potential to infiltrate and damage surrounding tissue. Anxiety around missing early cases needs to be balanced against inappropriate referral and unnecessary excision of benign lesions. Optical coherence tomography (OCT) is a microscopic imaging technique, which magnifies the surface of a skin lesion using near-infrared light. Used in conjunction with clinical or dermoscopic examination of suspected skin cancer, or both, OCT may offer additional diagnostic information compared to other technologies.

OBJECTIVES

To determine the diagnostic accuracy of OCT for the detection of cutaneous invasive melanoma and atypical intraepidermal melanocytic variants, basal cell carcinoma (BCC), or cutaneous squamous cell carcinoma (cSCC) in adults.

SEARCH METHODS

We undertook a comprehensive search of the following databases from inception up to August 2016: Cochrane Central Register of Controlled Trials; MEDLINE; Embase; CINAHL; CPCI; Zetoc; Science Citation Index; US National Institutes of Health Ongoing Trials Register; NIHR Clinical Research Network Portfolio Database; and the World Health Organization International Clinical Trials Registry Platform. We studied reference lists and published systematic review articles.

SELECTION CRITERIA

We included studies of any design evaluating OCT in adults with lesions suspicious for invasive melanoma and atypical intraepidermal melanocytic variants, BCC or cSCC, compared with a reference standard of histological confirmation or clinical follow-up.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted data using a standardised data extraction and quality assessment form (based on QUADAS-2). Our unit of analysis was lesions. Where possible, we estimated summary sensitivities and specificities using the bivariate hierarchical model.

MAIN RESULTS

We included five studies with 529 cutaneous lesions (282 malignant lesions) providing nine datasets for OCT, two for visual inspection alone, and two for visual inspection plus dermoscopy. Studies were of moderate to unclear quality, using data-driven thresholds for test positivity and giving poor accounts of reference standard interpretation and blinding. Studies may not have been representative of populations eligible for OCT in practice, for example due to high disease prevalence in study populations, and may not have reflected how OCT is used in practice, for example by using previously acquired OCT images.It was not possible to make summary statements regarding accuracy of detection of melanoma or of cSCC because of the paucity of studies, small sample sizes, and for melanoma differences in the OCT technologies used (high-definition versus conventional resolution OCT), and differences in the degree of testing performed prior to OCT (i.e. visual inspection alone or visual inspection plus dermoscopy).Pooled data from two studies using conventional swept-source OCT alongside visual inspection and dermoscopy for the detection of BCC estimated the sensitivity of OCT as 95% (95% confidence interval (CI) 91% to 97%) and specificity of 77% (95% CI 69% to 83%).When applied to a hypothetical population of 1000 lesions at the mean observed BCC prevalence of 60%, OCT would miss 31 BCCs (91 fewer than would be missed by visual inspection alone and 53 fewer than would be missed by visual inspection plus dermoscopy), and OCT would lead to 93 false-positive results for BCC (a reduction in unnecessary excisions of 159 compared to using visual inspection alone and of 87 compared to visual inspection plus dermoscopy).

AUTHORS' CONCLUSIONS

Insufficient data are available on the use of OCT for the detection of melanoma or cSCC. Initial data suggest conventional OCT may have a role for the diagnosis of BCC in clinically challenging lesions, with our meta-analysis showing a higher sensitivity and higher specificity when compared to visual inspection plus dermoscopy. However, the small number of studies and varying methodological quality means implications to guide practice cannot currently be drawn.Appropriately designed prospective comparative studies are required, given the paucity of data comparing OCT with dermoscopy and other similar diagnostic aids such as reflectance confocal microscopy.

Reflectance confocal microscopy for diagnosing cutaneous melanoma in adults

BACKGROUND

Melanoma has one of the fastest rising incidence rates of any cancer. It accounts for a small percentage of skin cancer cases but is responsible for the majority of skin cancer deaths. Early detection and treatment is key to improving survival; however, anxiety around missing early cases needs to be balanced against appropriate levels of referral and excision of benign lesions. Used in conjunction with clinical or dermoscopic suspicion of malignancy, or both, reflectance confocal microscopy (RCM) may reduce unnecessary excisions without missing melanoma cases.

OBJECTIVES

To determine the diagnostic accuracy of reflectance confocal microscopy for the detection of cutaneous invasive melanoma and atypical intraepidermal melanocytic variants in adults with any lesion suspicious for melanoma and lesions that are difficult to diagnose, and to compare its accuracy with that of dermoscopy.

SEARCH METHODS

We undertook a comprehensive search of the following databases from inception up to August 2016: Cochrane Central Register of Controlled Trials; MEDLINE; Embase; and seven other databases. We studied reference lists and published systematic review articles.

SELECTION CRITERIA

Studies of any design that evaluated RCM alone, or RCM in comparison to dermoscopy, in adults with lesions suspicious for melanoma or atypical intraepidermal melanocytic variants, compared with a reference standard of either histological confirmation or clinical follow-up.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted all data using a standardised data extraction and quality assessment form (based on QUADAS-2). We contacted authors of included studies where information related to the target condition or diagnostic threshold were missing. We estimated summary sensitivities and specificities per algorithm and threshold using the bivariate hierarchical model. To compare RCM with dermoscopy, we grouped studies by population (defined by difficulty of lesion diagnosis) and combined data using hierarchical summary receiver operating characteristic (SROC) methods. Analysis of studies allowing direct comparison between tests was undertaken. To facilitate interpretation of results, we computed values of specificity at the point on the SROC curve with 90% sensitivity as this value lies within the estimates for the majority of analyses. We investigated the impact of using a purposely developed RCM algorithm and in-person test interpretation.

MAIN RESULTS

The search identified 18 publications reporting on 19 study cohorts with 2838 lesions (including 658 with melanoma), which provided 67 datasets for RCM and seven for dermoscopy. Studies were generally at high or unclear risk of bias across almost all domains and of high or unclear concern regarding applicability of the evidence. Selective participant recruitment, lack of blinding of the reference test to the RCM result, and differential verification were particularly problematic. Studies may not be representative of populations eligible for RCM, and test interpretation was often undertaken remotely from the patient and blinded to clinical information.Meta-analysis found RCM to be more accurate than dermoscopy in studies of participants with any lesion suspicious for melanoma and in participants with lesions that were more difficult to diagnose (equivocal lesion populations). Assuming a fixed sensitivity of 90% for both tests, specificities were 82% for RCM and 42% for dermoscopy for any lesion suspicious for melanoma (9 RCM datasets; 1452 lesions and 370 melanomas). For a hypothetical population of 1000 lesions at the median observed melanoma prevalence of 30%, this equated to a reduction in unnecessary excisions with RCM of 280 compared to dermoscopy, with 30 melanomas missed by both tests. For studies in equivocal lesions, specificities of 86% would be observed for RCM and 49% for dermoscopy (7 RCM datasets; 1177 lesions and 180 melanomas). At the median observed melanoma prevalence of 20%, this reduced unnecessary excisions by 296 with RCM compared with dermoscopy, with 20 melanomas missed by both tests. Across all populations, algorithms and thresholds assessed, the sensitivity and specificity of the Pellacani RCM score at a threshold of three or greater were estimated at 92% (95% confidence interval (CI) 87 to 95) for RCM and 72% (95% CI 62 to 81) for dermoscopy.

AUTHORS' CONCLUSIONS

RCM may have a potential role in clinical practice, particularly for the assessment of lesions that are difficult to diagnose using visual inspection and dermoscopy alone, where the evidence suggests that RCM may be both more sensitive and specific in comparison to dermoscopy. Given the paucity of data to allow comparison with dermoscopy, the results presented require further confirmation in prospective studies comparing RCM with dermoscopy in a real-world setting in a representative population.

Exfoliative cytology for diagnosing basal cell carcinoma and other skin cancers in adults

BACKGROUND

Early accurate detection of all skin cancer types is essential to guide appropriate management, reduce morbidity and improve survival. Basal cell carcinoma (BCC) is usually localised to the skin but has potential to infiltrate and damage surrounding tissue, while cutaneous squamous cell carcinoma (cSCC) and melanoma have a much higher potential to metastasise and ultimately lead to death. Exfoliative cytology is a non-invasive test that uses the Tzanck smear technique to identify disease by examining the structure of cells obtained from scraped samples. This simple procedure is a less invasive diagnostic test than a skin biopsy, and for BCC it has the potential to provide an immediate diagnosis that avoids an additional clinic visit to receive skin biopsy results. This may benefit patients scheduled for either Mohs micrographic surgery or non-surgical treatments such as radiotherapy. A cytology scrape can never give the same information as a skin biopsy, however, so it is important to better understand in which skin cancer situations it may be helpful.

OBJECTIVES

To determine the diagnostic accuracy of exfoliative cytology for detecting basal cell carcinoma (BCC) in adults, and to compare its accuracy with that of standard diagnostic practice (visual inspection with or without dermoscopy). Secondary objectives were: to determine the diagnostic accuracy of exfoliative cytology for detecting cSCC, invasive melanoma and atypical intraepidermal melanocytic variants, and any other skin cancer; and for each of these secondary conditions to compare the accuracy of exfoliative cytology with visual inspection with or without dermoscopy in direct test comparisons; and to determine the effect of observer experience.

SEARCH METHODS

We undertook a comprehensive search of the following databases from inception up to August 2016: Cochrane Central Register of Controlled Trials; MEDLINE; Embase; CINAHL; CPCI; Zetoc; Science Citation Index; US National Institutes of Health Ongoing Trials Register; NIHR Clinical Research Network Portfolio Database; and the World Health Organization International Clinical Trials Registry Platform. We also studied the reference lists of published systematic review articles.

SELECTION CRITERIA

Studies evaluating exfoliative cytology in adults with lesions suspicious for BCC, cSCC or melanoma, compared with a reference standard of histological confirmation.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted all data using a standardised data extraction and quality assessment form (based on QUADAS-2). Where possible we estimated summary sensitivities and specificities using the bivariate hierarchical model.

MAIN RESULTS

We synthesised the results of nine studies contributing a total of 1655 lesions to our analysis, including 1120 BCCs (14 datasets), 41 cSCCs (amongst 401 lesions in 2 datasets), and 10 melanomas (amongst 200 lesions in 1 dataset). Three of these datasets (one each for BCC, melanoma and any malignant condition) were derived from one study that also performed a direct comparison with dermoscopy. Studies were of moderate to poor quality, providing inadequate descriptions of participant selection, thresholds used to make cytological and histological diagnoses, and blinding. Reporting of participants' prior referral pathways was particularly poor, as were descriptions of the cytodiagnostic criteria used to make diagnoses. No studies evaluated the use of exfoliative cytology as a primary diagnostic test for detecting BCC or other skin cancers in lesions suspicious for skin cancer. Pooled data from seven studies using standard cytomorphological criteria (but various stain methods) to detect BCC in participants with a high clinical suspicion of BCC estimated the sensitivity and specificity of exfoliative cytology as 97.5% (95% CI 94.5% to 98.9%) and 90.1% (95% CI 81.1% to 95.1%). respectively. When applied to a hypothetical population of 1000 clinically suspected BCC lesions with a median observed BCC prevalence of 86%, exfoliative cytology would miss 21 BCCs and would lead to 14 false positive diagnoses of BCC. No false positive cases were histologically confirmed to be melanoma. Insufficient data are available to make summary statements regarding the accuracy of exfoliative cytology to detect melanoma or cSCC, or its accuracy compared to dermoscopy.

AUTHORS' CONCLUSIONS

The utility of exfoliative cytology for the primary diagnosis of skin cancer is unknown, as all included studies focused on the use of this technique for confirming strongly suspected clinical diagnoses. For the confirmation of BCC in lesions with a high clinical suspicion, there is evidence of high sensitivity and specificity. Since decisions to treat low-risk BCCs are unlikely in practice to require diagnostic confirmation given that clinical suspicion is already high, exfoliative cytology might be most useful for cases of BCC where the treatments being contemplated require a tissue diagnosis (e.g. radiotherapy). The small number of included studies, poor reporting and varying methodological quality prevent us from drawing strong conclusions to guide clinical practice. Despite insufficient data on the use of cytology for cSCC or melanoma, it is unlikely that cytology would be useful in these scenarios since preservation of the architecture of the whole lesion that would be available from a biopsy provides crucial diagnostic information. Given the paucity of good quality data, appropriately designed prospective comparative studies may be required to evaluate both the diagnostic value of exfoliative cytology by comparison to dermoscopy, and its confirmatory value in adequately reported populations with a high probability of BCC scheduled for further treatment requiring a tissue diagnosis.

Computer-assisted diagnosis techniques (dermoscopy and spectroscopy-based) for diagnosing skin cancer in adults

BACKGROUND

Early accurate detection of all skin cancer types is essential to guide appropriate management and to improve morbidity and survival. Melanoma and cutaneous squamous cell carcinoma (cSCC) are high-risk skin cancers which have the potential to metastasise and ultimately lead to death, whereas basal cell carcinoma (BCC) is usually localised with potential to infiltrate and damage surrounding tissue. Anxiety around missing early curable cases needs to be balanced against inappropriate referral and unnecessary excision of benign lesions. Computer-assisted diagnosis (CAD) systems use artificial intelligence to analyse lesion data and arrive at a diagnosis of skin cancer. When used in unreferred settings ('primary care'), CAD may assist general practitioners (GPs) or other clinicians to more appropriately triage high-risk lesions to secondary care. Used alongside clinical and dermoscopic suspicion of malignancy, CAD may reduce unnecessary excisions without missing melanoma cases.

OBJECTIVES

To determine the accuracy of CAD systems for diagnosing cutaneous invasive melanoma and atypical intraepidermal melanocytic variants, BCC or cSCC in adults, and to compare its accuracy with that of dermoscopy.

SEARCH METHODS

We undertook a comprehensive search of the following databases from inception up to August 2016: Cochrane Central Register of Controlled Trials (CENTRAL); MEDLINE; Embase; CINAHL; CPCI; Zetoc; Science Citation Index; US National Institutes of Health Ongoing Trials Register; NIHR Clinical Research Network Portfolio Database; and the World Health Organization International Clinical Trials Registry Platform. We studied reference lists and published systematic review articles.

SELECTION CRITERIA

Studies of any design that evaluated CAD alone, or in comparison with dermoscopy, in adults with lesions suspicious for melanoma or BCC or cSCC, and compared with a reference standard of either histological confirmation or clinical follow-up.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted all data using a standardised data extraction and quality assessment form (based on QUADAS-2). We contacted authors of included studies where information related to the target condition or diagnostic threshold were missing. We estimated summary sensitivities and specificities separately by type of CAD system, using the bivariate hierarchical model. We compared CAD with dermoscopy using (a) all available CAD data (indirect comparisons), and (b) studies providing paired data for both tests (direct comparisons). We tested the contribution of human decision-making to the accuracy of CAD diagnoses in a sensitivity analysis by removing studies that gave CAD results to clinicians to guide diagnostic decision-making.

MAIN RESULTS

We included 42 studies, 24 evaluating digital dermoscopy-based CAD systems (Derm-CAD) in 23 study cohorts with 9602 lesions (1220 melanomas, at least 83 BCCs, 9 cSCCs), providing 32 datasets for Derm-CAD and seven for dermoscopy. Eighteen studies evaluated spectroscopy-based CAD (Spectro-CAD) in 16 study cohorts with 6336 lesions (934 melanomas, 163 BCC, 49 cSCCs), providing 32 datasets for Spectro-CAD and six for dermoscopy. These consisted of 15 studies using multispectral imaging (MSI), two studies using electrical impedance spectroscopy (EIS) and one study using diffuse-reflectance spectroscopy. Studies were incompletely reported and at unclear to high risk of bias across all domains. Included studies inadequately address the review question, due to an abundance of low-quality studies, poor reporting, and recruitment of highly selected groups of participants.Across all CAD systems, we found considerable variation in the hardware and software technologies used, the types of classification algorithm employed, methods used to train the algorithms, and which lesion morphological features were extracted and analysed across all CAD systems, and even between studies evaluating CAD systems. Meta-analysis found CAD systems had high sensitivity for correct identification of cutaneous invasive melanoma and atypical intraepidermal melanocytic variants in highly selected populations, but with low and very variable specificity, particularly for Spectro-CAD systems. Pooled data from 22 studies estimated the sensitivity of Derm-CAD for the detection of melanoma as 90.1% (95% confidence interval (CI) 84.0% to 94.0%) and specificity as 74.3% (95% CI 63.6% to 82.7%). Pooled data from eight studies estimated the sensitivity of multispectral imaging CAD (MSI-CAD) as 92.9% (95% CI 83.7% to 97.1%) and specificity as 43.6% (95% CI 24.8% to 64.5%). When applied to a hypothetical population of 1000 lesions at the mean observed melanoma prevalence of 20%, Derm-CAD would miss 20 melanomas and would lead to 206 false-positive results for melanoma. MSI-CAD would miss 14 melanomas and would lead to 451 false diagnoses for melanoma. Preliminary findings suggest CAD systems are at least as sensitive as assessment of dermoscopic images for the diagnosis of invasive melanoma and atypical intraepidermal melanocytic variants. We are unable to make summary statements about the use of CAD in unreferred populations, or its accuracy in detecting keratinocyte cancers, or its use in any setting as a diagnostic aid, because of the paucity of studies.

AUTHORS' CONCLUSIONS

In highly selected patient populations all CAD types demonstrate high sensitivity, and could prove useful as a back-up for specialist diagnosis to assist in minimising the risk of missing melanomas. However, the evidence base is currently too poor to understand whether CAD system outputs translate to different clinical decision-making in practice. Insufficient data are available on the use of CAD in community settings, or for the detection of keratinocyte cancers. The evidence base for individual systems is too limited to draw conclusions on which might be preferred for practice. Prospective comparative studies are required that evaluate the use of already evaluated CAD systems as diagnostic aids, by comparison to face-to-face dermoscopy, and in participant populations that are representative of those in which the test would be used in practice.

Reflectance confocal microscopy for diagnosing keratinocyte skin cancers in adults

BACKGROUND

Early accurate detection of all skin cancer types is important to guide appropriate management and improve morbidity and survival. Basal cell carcinoma (BCC) is usually a localised skin cancer but with potential to infiltrate and damage surrounding tissue, whereas cutaneous squamous cell carcinoma (cSCC) and melanoma are higher risk skin cancers with the potential to metastasise and ultimately lead to death. When used in conjunction with clinical or dermoscopic suspicion of malignancy, or both, reflectance confocal microscopy (RCM) may help to identify cancers eligible for non-surgical treatment without the need for a diagnostic biopsy, particularly in people with suspected BCC. Any potential benefit must be balanced against the risk of any misdiagnoses.

OBJECTIVES

To determine the diagnostic accuracy of RCM for the detection of BCC, cSCC, or any skin cancer in adults with any suspicious lesion and lesions that are difficult to diagnose (equivocal); and to compare its accuracy with that of usual practice (visual inspection or dermoscopy, or both).

SEARCH METHODS

We undertook a comprehensive search of the following databases from inception to August 2016: Cochrane Central Register of Controlled Trials; MEDLINE; Embase; CINAHL; CPCI; Zetoc; Science Citation Index; US National Institutes of Health Ongoing Trials Register; NIHR Clinical Research Network Portfolio Database; and the World Health Organization International Clinical Trials Registry Platform. We studied reference lists and published systematic review articles.

SELECTION CRITERIA

Studies of any design that evaluated the accuracy of RCM alone, or RCM in comparison to visual inspection or dermoscopy, or both, in adults with lesions suspicious for skin cancer compared with a reference standard of either histological confirmation or clinical follow-up, or both.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted data using a standardised data extraction and quality assessment form (based on QUADAS-2). We contacted authors of included studies where information related to the target condition or diagnostic threshold were missing. We estimated summary sensitivities and specificities using the bivariate hierarchical model. For computation of likely numbers of true-positive, false-positive, false-negative, and true-negative findings in the 'Summary of findings' tables, we applied summary sensitivity and specificity estimates to lower quartile, median and upper quartiles of the prevalence observed in the study groups. We also investigated the impact of observer experience.

MAIN RESULTS

The review included 10 studies reporting on 11 study cohorts. All 11 cohorts reported data for the detection of BCC, including 2037 lesions (464 with BCC); and four cohorts reported data for the detection of cSCC, including 834 lesions (71 with cSCC). Only one study also reported data for the detection of BCC or cSCC using dermoscopy, limiting comparisons between RCM and dermoscopy. Studies were at high or unclear risk of bias across almost all methodological quality domains, and were of high or unclear concern regarding applicability of the evidence. Selective participant recruitment, unclear blinding of the reference test, and exclusions due to image quality or technical difficulties were observed. It was unclear whether studies were representative of populations eligible for testing with RCM, and test interpretation was often undertaken using images, remotely from the participant and the interpreter blinded to clinical information that would normally be available in practice.Meta-analysis found RCM to be more sensitive but less specific for the detection of BCC in studies of participants with equivocal lesions (sensitivity 94%, 95% confidence interval (CI) 79% to 98%; specificity 85%, 95% CI 72% to 92%; 3 studies) compared to studies that included any suspicious lesion (sensitivity 76%, 95% CI 45% to 92%; specificity 95%, 95% CI 66% to 99%; 4 studies), although CIs were wide. At the median prevalence of disease of 12.5% observed in studies including any suspicious lesion, applying these results to a hypothetical population of 1000 lesions results in 30 BCCs missed with 44 false-positive results (lesions misdiagnosed as BCCs). At the median prevalence of disease of 15% observed in studies of equivocal lesions, nine BCCs would be missed with 128 false-positive results in a population of 1000 lesions. Across both sets of studies, up to 15% of these false-positive lesions were observed to be melanomas mistaken for BCCs. There was some suggestion of higher sensitivities in studies with more experienced observers. Summary sensitivity and specificity could not be estimated for the detection of cSCC due to paucity of data.

AUTHORS' CONCLUSIONS

There is insufficient evidence for the use of RCM for the diagnosis of BCC or cSCC in either population group. A possible role for RCM in clinical practice is as a tool to avoid diagnostic biopsies in lesions with a relatively high clinical suspicion of BCC. The potential for, and consequences of, misclassification of other skin cancers such as melanoma as BCCs requires further research. Importantly, data are lacking that compare RCM to standard clinical practice (with or without dermoscopy).

High-frequency ultrasound for diagnosing skin cancer in adults

BACKGROUND

Early, accurate detection of all skin cancer types is essential to guide appropriate management and to improve morbidity and survival. Melanoma and squamous cell carcinoma (SCC) are high-risk skin cancers with the potential to metastasise and ultimately lead to death, whereas basal cell carcinoma (BCC) is usually localised, with potential to infiltrate and damage surrounding tissue. Anxiety around missing early curable cases needs to be balanced against inappropriate referral and unnecessary excision of benign lesions. Ultrasound is a non-invasive imaging technique that relies on the measurement of sound wave reflections from the tissues of the body. At lower frequencies, the deeper structures of the body such as the internal organs can be visualised, while high-frequency ultrasound (HFUS) with transducer frequencies of 20 MHz or more has a much lower depth of tissue penetration but produces a higher resolution image of tissues and structures closer to the skin surface. Used in conjunction with clinical and/or dermoscopic examination of suspected skin cancer, HFUS may offer additional diagnostic information compared to other technologies.

OBJECTIVES

To assess the diagnostic accuracy of HFUS to assist in the diagnosis of a) cutaneous invasive melanoma and atypical intraepidermal melanocytic variants, b) cutaneous squamous cell carcinoma (cSCC), and c) basal cell carcinoma (BCC) in adults.

SEARCH METHODS

We undertook a comprehensive search of the following databases from inception up to August 2016: Cochrane Central Register of Controlled Trials; MEDLINE; Embase; CINAHL; CPCI; Zetoc; Science Citation Index; US National Institutes of Health Ongoing Trials Register; NIHR Clinical Research Network Portfolio Database; and the World Health Organization International Clinical Trials Registry Platform. We studied reference lists as well as published systematic review articles.

SELECTION CRITERIA

Studies evaluating HFUS (20 MHz or more) in adults with lesions suspicious for melanoma, cSCC or BCC versus a reference standard of histological confirmation or clinical follow-up.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted all data using a standardised data extraction and quality assessment form (based on QUADAS-2). Due to scarcity of data and the poor quality of studies, we did not undertake a meta-analysis for this review. For illustrative purposes, we plot estimates of sensitivity and specificity on coupled forest plots.

MAIN RESULTS

We included six studies, providing 29 datasets: 20 for diagnosis of melanoma (1125 lesions and 242 melanomas) and 9 for diagnosis of BCC (993 lesions and 119 BCCs). We did not identify any data relating to the diagnosis of cSCC.Studies were generally poorly reported, limiting judgements of methodological quality. Half the studies did not set out to establish test accuracy, and all should be considered preliminary evaluations of the potential usefulness of HFUS. There were particularly high concerns for applicability of findings due to selective study populations and data-driven thresholds for test positivity. Studies reporting qualitative assessments of HFUS images excluded up to 22% of lesions (including some melanomas) due to lack of visualisation in the test.Derived sensitivities for qualitative HFUS characteristics were at least 83% (95% CI 75% to 90%) for the detection of melanoma; the combination of three features (lesions appearing hypoechoic, homogenous and well defined) demonstrating 100% sensitivity in two studies (lower limits of the 95% CIs were 94% and 82%), with variable corresponding specificities of 33% (95% CI 20% to 48%) and 73% (95% CI 57% to 85%), respectively. Quantitative measurement of HFUS outputs in two studies enabled decision thresholds to be set to achieve 100% sensitivity; specificities were 93% (95% CI 77% to 99%) and 65% (95% CI 51% to 76%). It was not possible to make summary statements regarding HFUS accuracy for the diagnosis of BCC due to highly variable sensitivities and specificities.

AUTHORS' CONCLUSIONS

Insufficient data are available on the potential value of HFUS in the diagnosis of melanoma or BCC. Given the between-study heterogeneity, unclear to low methodological quality and limited volume of evidence, we cannot draw any implications for practice. The main value of the preliminary studies included may be in providing guidance on the possible components of new diagnostic rules for diagnosis of melanoma or BCC using HFUS that will require future evaluation. A prospective evaluation of HFUS added to visual inspection and dermoscopy alone in a standard healthcare setting, with a clearly defined and representative population of participants, would be required for a full and proper evaluation of accuracy.

Visual inspection and dermoscopy, alone or in combination, for diagnosing keratinocyte skin cancers in adults

BACKGROUND

Early accurate detection of all skin cancer types is important to guide appropriate management, to reduce morbidity and to improve survival. Basal cell carcinoma (BCC) is almost always a localised skin cancer with potential to infiltrate and damage surrounding tissue, whereas a minority of cutaneous squamous cell carcinomas (cSCCs) and invasive melanomas are higher-risk skin cancers with the potential to metastasise and cause death. Dermoscopy has become an important tool to assist specialist clinicians in the diagnosis of melanoma, and is increasingly used in primary-care settings. Dermoscopy is a precision-built handheld illuminated magnifier that allows more detailed examination of the skin down to the level of the superficial dermis. Establishing the value of dermoscopy over and above visual inspection for the diagnosis of BCC or cSCC in primary- and secondary-care settings is critical to understanding its potential contribution to appropriate skin cancer triage, including referral of higher-risk cancers to secondary care, the identification of low-risk skin cancers that might be treated in primary care and to provide reassurance to those with benign skin lesions who can be safely discharged.

OBJECTIVES

To determine the diagnostic accuracy of visual inspection and dermoscopy, alone or in combination, for the detection of (a) BCC and (b) cSCC, in adults. We separated studies according to whether the diagnosis was recorded face-to-face (in person) or based on remote (image-based) assessment.

SEARCH METHODS

We undertook a comprehensive search of the following databases from inception up to August 2016: Cochrane Central Register of Controlled Trials; MEDLINE; Embase; CINAHL; CPCI; Zetoc; Science Citation Index; US National Institutes of Health Ongoing Trials Register; NIHR Clinical Research Network Portfolio Database; and the World Health Organization International Clinical Trials Registry Platform. We studied reference lists and published systematic review articles.

SELECTION CRITERIA

Studies of any design that evaluated visual inspection or dermoscopy or both in adults with lesions suspicious for skin cancer, compared with a reference standard of either histological confirmation or clinical follow-up.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted all data using a standardised data extraction and quality assessment form (based on QUADAS-2). We contacted authors of included studies where information related to the target condition or diagnostic thresholds were missing. We estimated accuracy using hierarchical summary ROC methods. We undertook analysis of studies allowing direct comparison between tests. To facilitate interpretation of results, we computed values of sensitivity at the point on the SROC curve with 80% fixed specificity and values of specificity with 80% fixed sensitivity. We investigated the impact of in-person test interpretation; use of a purposely-developed algorithm to assist diagnosis; and observer expertise.

MAIN RESULTS

We included 24 publications reporting on 24 study cohorts, providing 27 visual inspection datasets (8805 lesions; 2579 malignancies) and 33 dermoscopy datasets (6855 lesions; 1444 malignancies). The risk of bias was mainly low for the index test (for dermoscopy evaluations) and reference standard domains, particularly for in-person evaluations, and high or unclear for participant selection, application of the index test for visual inspection and for participant flow and timing. We scored concerns about the applicability of study findings as of 'high' or 'unclear' concern for almost all studies across all domains assessed. Selective participant recruitment, lack of reproducibility of diagnostic thresholds and lack of detail on observer expertise were particularly problematic.The detection of BCC was reported in 28 datasets; 15 on an in-person basis and 13 image-based. Analysis of studies by prior testing of participants and according to observer expertise was not possible due to lack of data. Studies were primarily conducted in participants referred for specialist assessment of lesions with available histological classification. We found no clear differences in accuracy between dermoscopy studies undertaken in person and those which evaluated images. The lack of effect observed may be due to other sources of heterogeneity, including variations in the types of skin lesion studied, in dermatoscopes used, or in the use of algorithms and varying thresholds for deciding on a positive test result.Meta-analysis found in-person evaluations of dermoscopy (7 evaluations; 4683 lesions and 363 BCCs) to be more accurate than visual inspection alone for the detection of BCC (8 evaluations; 7017 lesions and 1586 BCCs), with a relative diagnostic odds ratio (RDOR) of 8.2 (95% confidence interval (CI) 3.5 to 19.3; P < 0.001). This corresponds to predicted differences in sensitivity of 14% (93% versus 79%) at a fixed specificity of 80% and predicted differences in specificity of 22% (99% versus 77%) at a fixed sensitivity of 80%. We observed very similar results for the image-based evaluations.When applied to a hypothetical population of 1000 lesions, of which 170 are BCC (based on median BCC prevalence across studies), an increased sensitivity of 14% from dermoscopy would lead to 24 fewer BCCs missed, assuming 166 false positive results from both tests. A 22% increase in specificity from dermoscopy with sensitivity fixed at 80% would result in 183 fewer unnecessary excisions, assuming 34 BCCs missed for both tests. There was not enough evidence to assess the use of algorithms or structured checklists for either visual inspection or dermoscopy.Insufficient data were available to draw conclusions on the accuracy of either test for the detection of cSCCs.

AUTHORS' CONCLUSIONS

Dermoscopy may be a valuable tool for the diagnosis of BCC as an adjunct to visual inspection of a suspicious skin lesion following a thorough history-taking including assessment of risk factors for keratinocyte cancer. The evidence primarily comes from secondary-care (referred) populations and populations with pigmented lesions or mixed lesion types. There is no clear evidence supporting the use of currently-available formal algorithms to assist dermoscopy diagnosis.

Dermoscopy, with and without visual inspection, for diagnosing melanoma in adults

BACKGROUND

Melanoma has one of the fastest rising incidence rates of any cancer. It accounts for a small percentage of skin cancer cases but is responsible for the majority of skin cancer deaths. Although history-taking and visual inspection of a suspicious lesion by a clinician are usually the first in a series of 'tests' to diagnose skin cancer, dermoscopy has become an important tool to assist diagnosis by specialist clinicians and is increasingly used in primary care settings. Dermoscopy is a magnification technique using visible light that allows more detailed examination of the skin compared to examination by the naked eye alone. Establishing the additive value of dermoscopy over and above visual inspection alone across a range of observers and settings is critical to understanding its contribution for the diagnosis of melanoma and to future understanding of the potential role of the growing number of other high-resolution image analysis techniques.

OBJECTIVES

To determine the diagnostic accuracy of dermoscopy alone, or when added to visual inspection of a skin lesion, for the detection of cutaneous invasive melanoma and atypical intraepidermal melanocytic variants in adults. We separated studies according to whether the diagnosis was recorded face-to-face (in-person), or based on remote (image-based), assessment.

SEARCH METHODS

We undertook a comprehensive search of the following databases from inception up to August 2016: CENTRAL; MEDLINE; Embase; CINAHL; CPCI; Zetoc; Science Citation Index; US National Institutes of Health Ongoing Trials Register; NIHR Clinical Research Network Portfolio Database; and the World Health Organization International Clinical Trials Registry Platform. We studied reference lists and published systematic review articles.

SELECTION CRITERIA

Studies of any design that evaluated dermoscopy in adults with lesions suspicious for melanoma, compared with a reference standard of either histological confirmation or clinical follow-up. Data on the accuracy of visual inspection, to allow comparisons of tests, was included only if reported in the included studies of dermoscopy.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted all data using a standardised data extraction and quality assessment form (based on QUADAS-2). We contacted authors of included studies where information related to the target condition or diagnostic threshold were missing. We estimated accuracy using hierarchical summary receiver operating characteristic (SROC),methods. Analysis of studies allowing direct comparison between tests was undertaken. To facilitate interpretation of results, we computed values of sensitivity at the point on the SROC curve with 80% fixed specificity and values of specificity with 80% fixed sensitivity. We investigated the impact of in-person test interpretation; use of a purposely developed algorithm to assist diagnosis; observer expertise; and dermoscopy training.

MAIN RESULTS

We included a total of 104 study publications reporting on 103 study cohorts with 42,788 lesions (including 5700 cases), providing 354 datasets for dermoscopy. The risk of bias was mainly low for the index test and reference standard domains and mainly high or unclear for participant selection and participant flow. Concerns regarding the applicability of study findings were largely scored as 'high' concern in three of four domains assessed. Selective participant recruitment, lack of reproducibility of diagnostic thresholds and lack of detail on observer expertise were particularly problematic.The accuracy of dermoscopy for the detection of invasive melanoma or atypical intraepidermal melanocytic variants was reported in 86 datasets; 26 for evaluations conducted in person (dermoscopy added to visual inspection), and 60 for image-based evaluations (diagnosis based on interpretation of dermoscopic images). Analyses of studies by prior testing revealed no obvious effect on accuracy; analyses were hampered by the lack of studies in primary care, lack of relevant information and the restricted inclusion of lesions selected for biopsy or excision. Accuracy was higher for in-person diagnosis compared to image-based evaluations (relative diagnostic odds ratio (RDOR) 4.6, 95% confidence interval (CI) 2.4 to 9.0; P < 0.001).We compared accuracy for (a), in-person evaluations of dermoscopy (26 evaluations; 23,169 lesions and 1664 melanomas),versus visual inspection alone (13 evaluations; 6740 lesions and 459 melanomas), and for (b), image-based evaluations of dermoscopy (60 evaluations; 13,475 lesions and 2851 melanomas),versus image-based visual inspection (11 evaluations; 1740 lesions and 305 melanomas). For both comparisons, meta-analysis found dermoscopy to be more accurate than visual inspection alone, with RDORs of (a), 4.7 (95% CI 3.0 to 7.5; P < 0.001), and (b), 5.6 (95% CI 3.7 to 8.5; P < 0.001). For a), the predicted difference in sensitivity at a fixed specificity of 80% was 16% (95% CI 8% to 23%; 92% for dermoscopy + visual inspection versus 76% for visual inspection), and predicted difference in specificity at a fixed sensitivity of 80% was 20% (95% CI 7% to 33%; 95% for dermoscopy + visual inspection versus 75% for visual inspection). For b) the predicted differences in sensitivity was 34% (95% CI 24% to 46%; 81% for dermoscopy versus 47% for visual inspection), at a fixed specificity of 80%, and predicted difference in specificity was 40% (95% CI 27% to 57%; 82% for dermoscopy versus 42% for visual inspection), at a fixed sensitivity of 80%.Using the median prevalence of disease in each set of studies ((a), 12% for in-person and (b), 24% for image-based), for a hypothetical population of 1000 lesions, an increase in sensitivity of (a), 16% (in-person), and (b), 34% (image-based), from using dermoscopy at a fixed specificity of 80% equates to a reduction in the number of melanomas missed of (a), 19 and (b), 81 with (a), 176 and (b), 152 false positive results. An increase in specificity of (a), 20% (in-person), and (b), 40% (image-based), at a fixed sensitivity of 80% equates to a reduction in the number of unnecessary excisions from using dermoscopy of (a), 176 and (b), 304 with (a), 24 and (b), 48 melanomas missed.The use of a named or published algorithm to assist dermoscopy interpretation (as opposed to no reported algorithm or reported use of pattern analysis), had no significant impact on accuracy either for in-person (RDOR 1.4, 95% CI 0.34 to 5.6; P = 0.17), or image-based (RDOR 1.4, 95% CI 0.60 to 3.3; P = 0.22), evaluations. This result was supported by subgroup analysis according to algorithm used. We observed higher accuracy for observers reported as having high experience and for those classed as 'expert consultants' in comparison to those considered to have less experience in dermoscopy, particularly for image-based evaluations. Evidence for the effect of dermoscopy training on test accuracy was very limited but suggested associated improvements in sensitivity.

AUTHORS' CONCLUSIONS

Despite the observed limitations in the evidence base, dermoscopy is a valuable tool to support the visual inspection of a suspicious skin lesion for the detection of melanoma and atypical intraepidermal melanocytic variants, particularly in referred populations and in the hands of experienced users. Data to support its use in primary care are limited, however, it may assist in triaging suspicious lesions for urgent referral when employed by suitably trained clinicians. Formal algorithms may be of most use for dermoscopy training purposes and for less expert observers, however reliable data comparing approaches using dermoscopy in person are lacking.

Guidelines of care for the management of primary cutaneous melanoma

The incidence of primary cutaneous melanoma continues to increase each year. Melanoma accounts for the majority of skin cancer-related deaths, but treatment is usually curative following early detection of disease. In this American Academy of Dermatology clinical practice guideline, updated treatment recommendations are provided for patients with primary cutaneous melanoma (American Joint Committee on Cancer stages 0-IIC and pathologic stage III by virtue of a positive sentinel lymph node biopsy). Biopsy techniques for a lesion that is clinically suggestive of melanoma are reviewed, as are recommendations for the histopathologic interpretation of cutaneous melanoma. The use of laboratory, molecular, and imaging tests is examined in the initial work-up of patients with newly diagnosed melanoma and for follow-up of asymptomatic patients. With regard to treatment of primary cutaneous melanoma, recommendations for surgical margins and the concepts of staged excision (including Mohs micrographic surgery) and nonsurgical treatments for melanoma in situ, lentigo maligna type (including topical imiquimod and radiation therapy), are updated. The role of sentinel lymph node biopsy as a staging technique for cutaneous melanoma is described, with recommendations for its use in clinical practice. Finally, current data regarding pregnancy and melanoma, genetic testing for familial melanoma, and management of dermatologic toxicities related to novel targeted agents and immunotherapies for patients with advanced disease are summarized.

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

OPINION STATEMENT

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

Automated decision support in melanocytic lesion management

An automated melanocytic lesion image-analysis algorithm is described that aims to reproduce the decision-making of a dermatologist. The utility of the algorithm lies in its ability to identify lesions requiring excision from lesions not requiring excision. Using only wavelet coefficients as features, and testing three different machine learning algorithms, a cohort of 250 images of pigmented lesions is classified based on expert dermatologists’ recommendations of either excision (165 images) or no excision (85 images). It is shown that the best algorithm utilises the Shannon4 wavelet coupled to the support vector machine, where the latter is used as the classifier. In this case the algorithm, utilising only 22 othogonal features, achieves a 10-fold cross validation sensitivity and specificity of 0.96 and 0.87, resulting in a diagnostic-odds ratio of 261. The advantages of this method over diagnostic algorithms–which make a melanoma/no melanoma decision–are twofold: first, by reproducing the decision-making of a dermatologist, the average number of lesions excised per melanoma among practioners in general can be reduced without compromising the detection of melanoma; and second, the intractable problem of clinically differentiating between many atypical dysplastic naevi and melanoma is avoided. Since many atypical naevi that require excision on clinical grounds will not be melanoma, the algorithm–in contrast to diagnostic algorithms–can aim for perfect specificities without clinical concerns, thus lowering the excision rate of non-melanoma. Finally, the algorithm has been implemented as a smart phone application to investigate its utility in clinical practice and to streamline the assimilation of hitherto unseen tested images into the training set.