Diagnostic image analysis of malignant melanoma in in vivo confocal laser-scanning microscopy: a preliminary study

Background/purpose: In this study we assessed the applicability of image analysis and a machine learning algorithm on diagnostic discrimination of benign and malignant melanocytic skin tumours in in vivo confocal laser-scanning microscopy (CLSM).
Methods: A total of 857 CLSM tumour images including 408 benign nevi and 449 melanoma images was evaluated. Image analysis was based on features of the wavelet transform. For classification purposes we used a classification tree software (CART). Moreover, automated image analysis results were compared with the prediction success of an independent human observer.
Results: CART analysis of the whole set of CLSM tumour images correctly classified 97.55% and 96.32% of melanoma and nevi images. In contrast, sensitivity and specificity of 85.52% and 80.15% could be reached by the human observer. When the image set was randomly divided into a learning (67% of the images) and a test set (33% of the images), overall 97.31% and 81.03% of the tumour images in the learning and test set could be classified correctly by the CART procedure.
Conclusion: Provided automated decisions can be used as a second opinion. This can be valuable in assisting diagnostic decisions in this new and exciting imaging technique.     

Correlation of image analysis features and visual morphology in melanocytic skin tumours using in vivo confocal laser scanning microscopy

Background/purpose: In vivo confocal laser scanning microscopy (CLSM) represents a novel imaging tool that allows the non-invasive examination of skin cancer morphology at a quasi histological resolution without biopsy. Previous studies dealt with the search for diagnostic, but subjective visual criteria. In this study we examined the correlation between objectively reproducible image-analysis features und visual morphology in melanocytic skin tumours using CLSM.
Methods: Eight hundred and fifty-seven CLSM tumour images including 408 benign nevi and 449 melanoma images were evaluated. Image analysis was based on features of the wavelet transform and classification tree analysis (CART) was used for classification purposes. In a second step, morphologic details of CLSM images, which have turned out to be of diagnostic significance by the classification algorithm were evaluated.
Results: CART analysis of the whole set of CLSM images correctly classified 97.55% of all melanoma images and 96.32% of all nevi images. Seven classification tree nodes seemed to indicate benign nevi, whereas six nodes were suggestive for melanoma morphology. The visual examination of selected nodes demonstrated that monomorphic melanocytic cells and melanocytic cell nests are characteristic for benign nevi whereas polymorphic melanocytic cells, disarray of melanocytic architecture and poorly defined or absent keratinocyte cell borders are characteristic for melanoma.
Conclusion: Well-known, but subjective CLSM criteria could be objectively reproduced by image analysis features and classification tree analysis. Moreover, features not accessible to the human eye seem to contribute to classification success.     

In vivo confocal laser scanning microscopy of melanocytic skin tumours: diagnostic applicability using unselected tumour images

BACKGROUND: In vivo confocal laser scanning microscopy (CLSM) represents a novel imaging tool that allows the noninvasive examination of skin cancer morphology in real time at a 'quasi-histopathological' resolution viewing microanatomical structures and individual cells. OBJECTIVES: To validate diagnostic confocal examination of melanocytic skin tumours using unselected tumour images. METHODS: In the present study, we used a total of 3709 unselected CLSM tumour images obtained from 20 malignant melanomas and 50 benign naevi. The entire set of images derived from each tumour was evaluated by independent observers. Classification tree analysis based on a subsample of 857 tumour images was performed to develop a diagnostic algorithm. RESULTS: Overall, sensitivity and specificity of 97.5% and 99% could be achieved by the independent observers (positive predictive value 97.5%, negative predictive value 99%). Classification tree analysis yielded a three-step algorithm based on only three morphological CLSM features, facilitating a correct classification in 92.4% of the benign naevus images and 97.6% of melanoma images. CONCLUSIONS: In vivo CLSM augurs a sea change in the way we will view skin tumour processes clinically at the bedside and merits application for use as a screening tool in skin oncology.     

Mobile teledermatology for skin tumour screening: diagnostic accuracy of clinical and dermoscopic image tele-evaluation using cellular phones

Background The ability to diagnose malignant skin tumours accurately and to distinguish them from benign lesions is vital in ensuring appropriate patient management. Little is known about the effects of mobile teledermatology services on diagnostic accuracy and their appropriateness for skin tumour surveillance. Objectives To evaluate the diagnostic accuracy of clinical and dermoscopic image tele‐evaluation for mobile skin tumour screening. Methods Over a 3‐month period up to three clinical and dermoscopic images were obtained of 113 skin tumours from 88 patients using a mobile phone camera. Dermoscopic images were taken with a dermatoscope applied to the camera lens. Clinical and dermoscopic images of each lesion together with clinical information were separately teletransmitted for decision‐making. Results were compared with those obtained by face‐to‐face examination and histopathology as the gold standard. Results A total of 322 clinical and 278 dermoscopic images were acquired; two (1%) clinical and 18 (6%) dermoscopic pictures were inadequate for decision‐making. After excluding inadequate images, the majority of which were dermoscopic pictures, only 104 of the 113 skin tumours from 80 of 88 patients could be tele‐evaluated. Among these 104 lesions, 25 (24%) benign nonmelanocytic, 15 (14%) benign melanocytic, 58 (56%) malignant nonmelanocytic and six (6%) malignant melanocytic lesions were identified. Clinical and dermoscopic tele‐evaluations demonstrated strong concordance with the gold standard (κ = 0·84 for each) and similar high sensitivity and specificity for all diagnostic categories. With regard to the detailed diagnoses, clinical image tele‐evaluation was superior to teledermoscopy resulting in 16 vs. 22 discordant cases. Conclusions Clinical image tele‐evaluation might be the method of choice for mobile tumour screening.     

Automated epiluminescence microscopy—tissue counter analysis using CART and 1-NN in the diagnosis of Melanoma

Background/purpose: In tissue counter analysis, digital images are overlayed with regularly distributed measuring masks (elements) of equal size and shape, and the digital contents (grey level, colour and texture parameters) of each element are used for statistical analysis. In this study we assessed the applicability of tissue counter analysis and machine learning algorithms on tumour segmentation and diagnostic discrimination of benign and malignant melanocytic skin lesions.
Methods: A total of 369 standardised dermatoscopic images (93 melanomas, 276 benign nevi) were evaluated. The Classification and Regression Tree (CART) analysis was performed in order to differentiate between melanocytic skin lesions and surrounding skin. Instance-based learning (1-NN) was tested for differentiating between benign and malignant tumour elements. For diagnostic assessment, only the percentage of elements suggestive for malignancy in each lesion was used.
Results: Evaluation of a total of 369 melanocytic skin lesions showed a suitable segmentation of the tumour portion in 97.6%. When instance-based learning was applied to an independent test set, a threshold value of 27.4% of elements suggestive for malignancy recognised 35 out of 35 melanomas and 100 out of 101 nevi (sensitivity 100%, specificity 99%, positive predictive value 97.2%, negative predictive value 100%).
Conclusion: Tissue counter analysis combined with machine learning algorithms turned out to be a useful method for diagnostic purposes in epiluminescence microscopy.     

High-resolution laser Doppler perfusion imaging aids in differentiating between benign and malignant melanocytic skin tumours

Malignant melanomas are characterized by heterogeneity and asymmetry as well as by a higher density of blood vessels than benign pigmented tumours. The aim of this study was to evaluate the benefit of high-resolution laser Doppler perfusion imaging (LDPI) in the differential diagnosis of pigmented skin tumours. One-hundred-and-eighty-nine patients were examined with the LDPI, 22 with malignant melanomas, 39 with clinically suspicious dysplastic melanocytic naevi and 27 with basal cell carcinomas. Following examination, the tumours were excised and examined histologically. A control group of 101 melanocytic naevi showed clinically and, with epiluminescence microscopy, definitely benign criteria. These naevi were not excised. In malignant melanomas there was a 3.6+/-1.5 times higher perfusion than in healthy skin. The corresponding figures for clinically suspicious melanocytic naevi and basal cell carcinomas were 2.2+/-1.1 and 2.0+/-0.7, respectively. The increase in flow in malignant melanomas was significantly higher than in clinically suspicious melanocytic naevi and basal cell carcinomas (p < 0.001). All malignant melanomas showed at least 1.8 times higher flow values than healthy skin. When this value is taken as the basis for the diagnosis "benign or malignant", the LDPI proved a sensitivity of 100% and a specificity of 85%. If only the distinction between malignant melanomas and clinically suspicious naevi is considered, the specificity is reduced to 48%. There was no correlation between tumour thickness and increase in the mean perfusion of malignant melanomas (r = 0.14; p = 0.5). High-resolution LDPI can be used as an additional automatic screening method.     

In vivo reflectance confocal microscopy image interpretation for the dermatopathologist

Reflectance confocal microscopy (RCM) is a technology utilized for bedside diagnosis of cutaneous pathology by non‐invasive, in vivo, cellular‐level imaging. With the recent establishment of reimbursement codes by the US Centers for Medicaid and Medicare Services, RCM is now likely to be employed by clinical dermatologists and impact decision making on skin cancer management. Dermatopathologists, therefore, would benefit from learning how to interpret RCM images and how RCM findings correlate with histopathological criteria of diagnosis. This review briefly explains the principles behind RCM image acquisition, describes the key RCM features of normal skin, and delineates the RCM characteristics of frequently observed benign and malignant neoplasms.     

Digital image analysis for diagnosis of cutaneous melanoma. Development of a highly effective computer algorithm based on analysis of 837 melanocytic lesions

BACKGROUND: Digital image analysis has been introduced into the diagnosis of skin lesions based on dermoscopic pictures. OBJECTIVES: To develop a computer algorithm for the diagnosis of melanocytic lesions and to compare its diagnostic accuracy with the results of established dermoscopic classification rules. METHODS: In the Department of Dermatology, University of Tuebingen, Germany, 837 melanocytic skin lesions were prospectively imaged by a dermoscopy video system in consecutive patients. Of these lesions, 269 were excised and examined by histopathology: 84 were classified as cutaneous melanomas and 185 as benign melanocytic naevi. The remaining 568 lesions were diagnosed by dermoscopy as benign. Digital image analysis was performed in all 837 benign and malignant melanocytic lesions using 64 different analytical parameters. RESULTS: For lesions imaged completely (diameter < or = 12 mm), three analytical parameters were found to distinguish clearly between benign and malignant lesions, while in incompletely imaged lesions six parameters enabled differentiation. Based on the respective parameters and logistic regression analysis, a diagnostic computer algorithm for melanocytic lesions was developed. Its diagnostic accuracy was 82% for completely imaged and 84% for partially imaged lesions. All 837 melanocytic lesions were classified by established dermoscopic algorithms and the diagnostic accuracy was found to be in the same range (ABCD rule 78%, Menzies' score 83%, seven-point checklist 88%, and seven features for melanoma 81%). CONCLUSIONS: A diagnostic algorithm for digital image analysis of melanocytic lesions can achieve the same range of diagnostic accuracy as the application of dermoscopic classification rules by experts. The present diagnostic algorithm, however, still requires a medical expert who is qualified to recognize cutaneous lesions as being of melanocytic origin.     

Blood flow compared in benign melanocytic naevi, malignant melanomas and basal cell carcinomas

Histologically, increasing malignancy is reflected by increased tumour vascularization. We therefore assessed whether laser Doppler perfusion imaging might allow differentiation between different types of skin tumour. The perfusion patterns of 27 melanocytic naevi, 16 basal cell carcinomas and six malignant melanomas were examined with a laser Doppler perfusion imager before excision for precise diagnosis. Malignant melanomas were significantly more perfused than basal cell carcinomas and tended also to be more so than melanocytic naevi. Further, for the first time, the perfusion pattern of the entire tumour area was visualized two-dimensionally; it was higher in the centre of the tumours than at the periphery for malignant melanomas and melanocytic naevi but was the same throughout for basal cell carcinomas. Laser Doppler perfusion imaging can thus provide information to assist in the differential diagnosis of skin tumours. In addition, with increases in resolution of the technique such data may well become steadily more useful in the future.     

Quantitative surface topography as a tool in the differential diagnosis between melanoma and naevus

Background/aims: The mainstays of the clinical diagnosis of melanoma are asymmetry, border irregularity, color variegation, and a diameter >6 mm, and any major progress in diagnostic accuracy will probably be related to the development of additional criteria. Such independent criteria might arise from the study of the geometry of the tumour surface, because this quality has been substantially disregarded until now. Our work is aimed at utilizing the surface topography for the differential diagnosis between malignant melanoma and naevocytic naevus. Methods: Dynamically focusing optical profilometry was used to measure the surfaces of negative replicas of melanocytic skin tumours and of the surrounding normal skin. 21 silicone imprints of superficial spreading melanomas and 25 imprints of naevocytic naevi were examined. Results: Melanomas and naevi differed with respect to a variety of statistical surface parameters, and a linear discriminant analysis correctly allocated 19 out of 21 melanomas (90%) and 21 out of 25 naevi (84%). To get an unbiased estimate of the errors to be expected with this classification rule, we calculated bootstrap-corrections to the apparent errors. Estimated probabilities of correct allocation were 84.1% for melanomas and 77.1% for naevi. Conclusion: Our findings suggest that simple statistical parameters of surface topography can differentiate effectively between malignant melanomas and naevocytic naevi.     

A systematic heuristic approach for feature selection for melanoma discrimination using clinical images

BACKGROUND: Numerous features are derived from the asymmetry, border irregularity, color variegation, and diameter of the skin lesion in dermatology for diagnosing malignant melanoma. Feature selection for the development of automated skin lesion discrimination systems is an important consideration. METHODS: In this research, a systematic heuristic approach is investigated for feature selection and lesion classification. The approach integrates statistical-, correlation-, histogram-, and expert system-based components. Using statistical and correlation measures, interrelationships among features are determined. Expert system analysis is performed to identify redundant features. The feature selection process is applied to 19 shape and color features for a clinical image data set containing 355 malignant melanomas, 125 basal cell carcinomas, 177 dysplastic nevi, 199 nevocellular nevi, 139 seborrheic keratoses, and 45 vascular lesions. RESULTS: Experimental results show reduced lesion classification error rates based on condensing the shape and color feature set from 19 features to 13 features using the feature selection process. Specifically, average test lesion classification error rates for discriminating malignant melanoma from non-melanoma lesions were reduced from 26.6% for 19 features to 23.2% for 13 features over five randomly generated training and test sets. CONCLUSIONS: The experimental results show that the systematic heuristic approach for feature reduction can be successfully applied to achieve improved lesion discrimination. The feature reduction technique facilitates the elimination of redundant information that may inhibit lesion classification performance. The clinical application of this result is that automated skin lesion classification algorithm development can be fostered with systematic feature selection techniques.     

Assessment of clinical parameters associated with mutational status in metastatic malignant melanoma: a single‐centre investigation of 141 patients

Background Inhibitors of the mutated, constitutively activated BRAF protein have shown efficacy in the treatment of metastatic melanoma in clinical trials. Mutation analysis especially of the BRAF, NRAS and KIT genes is essential to identify patients suitable for targeted therapies and has been introduced into routine patient care. Objectives To correlate mutational status with clinical parameters including age, skin type, number of melanocytic naevi, primary tumour location, chronic sun damage and exposure to ultraviolet (UV) irradiation. The overall aim was to define subgroups with an increased or decreased likelihood of gene mutations. Additionally, the impact of activating BRAF mutations on clinical course was investigated. Methods In a single‐centre, retrospective approach, mutation analysis was performed on patients with metastatic malignant melanoma. Clinical parameters were correlated with molecular findings. The total sun‐burden score was assessed using a validated standardized questionnaire. Results The analysis included 141 patients with metastatic melanoma. Forty‐four per cent of patients had activating BRAF mutations and were significantly younger than patients with wild‐type BRAF or with NRAS mutations. KIT mutations were detected in only 3% of the patients. BRAF‐mutated melanomas developed preferentially in intermittently sun‐exposed areas of the body, and patients had significantly more melanocytic naevi. Once patients had progressed into stage IV disease, survival times were identical for those with BRAF‐mutated and BRAF wild‐type tumours. Conclusions Mutations of the BRAF gene are correlated with younger age, a higher number of melanocytic naevi and a tumour location in intermittently UV‐exposed skin. Signs of chronic photodamage are not indicative of mutational status. Patients with metastatic melanoma with BRAF mutations showed a nonsignificant tendency to progress later to stage IV disease, but once metastases were present the prognosis was identical to that with BRAF wild‐type tumours.     

Brooke–Spiegler syndrome: focus on reflectance confocal microscopy findings of trichoepithelioma and flat cylindroma

Brooke–Spiegler syndrome (BSS) is a rare, autosomal dominant disorder characterized by multiple adnexal tumours, especially trichoepitheliomas, cylindromas and occasionally spiradenomas. These lesions usually begin to appear in the second or third decade of life. Malignant transformation of pre‐existing tumours may occur. In vivo reflectance‐mode confocal microscopy (RCM) is a noninvasive method that can be used to visualize the epidermis and the upper dermis at almost histological resolution. It has been used to evaluate several skin conditions, especially malignant lesions, and has been reported to be useful for differentiating between malignant and benign skin tumours. Only a few studies have reported on the use of confocal microscopy to characterize the features of benign adnexal neoplasms. We present a patient with BSS who presented to our clinic with multiple skin tumours. The possible utility of RCM for identifying adnexal neoplasms is emphasized in this report, which also describes the observed microscopic features.     

Moesin and CD44 expression in cutaneous melanocytic tumours

The ERM (ezrin, radixin and moesin) family members, located just beneath the plasma membranes, are thought to be involved in the association of actin filaments with the plasma membrane. One of the family members, moesin, is reported to bind to CD44. Splice variants of CD44 are thought to be associated with tumour progression or differentiation. Our aim was to investigate immunohistochemically the expression of moesin together with CD44 on paraffin tissue sections of a series of melanocytic tumours. The material included 12 ordinary melanocytic naevi, six Spitz naevi, eight dysplastic naevi, six blue naevi, seven malignant melanomas in situ , 15 primary malignant melanomas, five metastatic melanomas to the skin and five lymph node metastases. In the normal skin and the melanocytic tumours the expression of moesin was largely similar to that of CD44 standard. Strong moesin staining was observed in benign melanocytic lesions and melanomas in situ . However, the expression was decreased in advanced malignant melanomas. The moesin labelling in melanoma cells was downregulated with the depth of dermal invasion. The immunoreactivity was also diminished in the skin metastases and the lymph node metastases of melanoma. These results suggest that in melanocytic tumours, the alternation in the expression of moesin may be involved in the progression of malignancy.     

Digital computer analysis of dermatoscopical images of 260 melanocytic skin lesions; perimeter/area ratio for the differentiation between malignant melanomas and melanocytic nevi

BACKGROUND: Digital computer analysis of dermatoscopical images has been reported to facilitate the differential diagnosis of pigmented skin lesions in recent years. OBJECTIVE: The aim of our study was to perform digital computer analysis of a set of different melanocytic lesions and compare the objective results. METHODS: The set of 260 melanocytic lesions (150 excised difficult cases (46 melanomas, 47 atypical nevi, 57 common nevi and 110 unexcised common nevi) was automatically analysed by the digital dermatoscopical system microDERM. We searched for differences in asymmetry, size, compactness and colour distribution. Perimeter/area ratio was calculated. RESULTS: The perimeter/area ratio was detected as the most important criterion for differentiation between malignant and benign melanocytic lesions (sensitivity 91.3% and specificity 90.7% for malignant melanomas vs. all benign nevi; sensitivity 91.3% and specificity 80.8% for melanomas vs. clinically atypical nevi). Differences in size of the lesion, shape and asymmetry of colour were found and statistically verified. Using step-wise logistic regression the formula for calculation of probability of malignant nature of every analysed lesion was constructed. CONCLUSION: The perimeter/area ratio is a simple parameter for the differential diagnosis of melanocytic skin lesions.     

In vivo confocal laser scanning microscopy in the diagnosis of melanocytic skin tumours

Summary Melanoma of the skin represents one of the greatest challenges in early or preventive detection. Whereas surgical excision in early stages of melanoma development is almost always curative, delayed recognition puts the patient at risk for destructive growth and death from disease once the tumour has progressed to competence for metastasis. The worldwide introduction of dermoscopy has led to improved diagnostic accuracy for melanocytic skin tumours. Whereas dermoscopy has probably reached the method's inherent potential diagnostic accuracy because of the lack of cellular level evaluation, further improvements could be expected by in vivo confocal laser scanning microscopy. In vivo confocal microscopy represents a novel imaging tool that allows the noninvasive examination of skin cancer morphology in real time at a 'quasihistopathological' resolution viewing microanatomical structures and individual cells. Numerous morphological confocal features of melanocytic skin tumours have been described and histopathological correlates of confocal structures have been previously elucidated. Recently, several studies have evaluated the diagnostic accuracy of in vivo confocal microscopy for melanocytic skin tumours, investigating approximately 50 000 tumour images. Remarkably, sensitivity superior to the diagnostic accuracy achieved with dermoscopy could be reached by this imaging modality. These studies represent a significant contribution to the body of research necessary for the evaluation and implementation of in vivo confocal microscopy in clinical practice to avoid many currently unnecessary biopsies. In vivo confocal microscopy probably augurs a sea change in the way we evaluate melanocytic skin tumours in the future and will ultimately move the art of histological diagnosis closer to the bedside.     

First experiences using reflectance confocal microscopy on equivocal skin lesions in Queensland

Background/Objectives: Reflectance confocal microscopy (RCM) is a non‐invasive method of imaging human skin in vivo. The purpose of this study was to observe the experience of using RCM on equivocal skin lesions in a tertiary clinical setting in Queensland. Methods: Fifty equivocal lesions on 42 patients were imaged using a reflectance confocal microscope immediately prior to being excised. The images were then analysed blind to the histopathological diagnosis. The experience and problems encountered when using RCM on skin lesions for the first time was also observed. Results: On RCM analysis 12/13 melanomas (92.3% sensitivity, 75% specificity), 19/22 benign naevi (86% sensitivity, 95% specificity), 6/9 basal cell carcinomas (66.7% sensitivity, 100% specificity)and 6/6 squamous cell carcinomas and its precursors (100% sensitivity, 75% specificity) were diagnosed correctly when using histology as the gold standard. We identified three common problems that affected image quality: object artefacts; positioning artefacts; and movement artefacts. Conclusions: Using simple techniques we found that common RCM features were readily identifiable and common artefacts could be minimized, making RCM a useful tool to aid the diagnosis of equivocal skin lesions in a clinical setting.     

In vivo confocal microscopy for diagnosis of melanoma and Basal cell carcinoma using a two-step method: analysis of 710 consecutive clinically equivocal cases

We describe two algorithms to diagnose basal cell carcinomas (BCCs) and melanomas (MMs) using in vivo reflectance confocal microscopy (RCM). A total of 710 consecutive cutaneous lesions excised to exclude malignancy (216 MMs, 266 nevi, 119 BCCs, 67 pigmented facial macules, and 42 other skin tumors) were imaged by RCM. RCM features were correlated with pathology diagnosis to develop diagnostic algorithms. The diagnostic accuracy of the BCC algorithm defined on multivariate analysis of the training set (50%) and tested on the remaining cases was 100% sensitivity, 88.5% specificity. Positive features were polarized elongated features, telangiectasia and convoluted vessels, basaloid nodules, and epidermal shadowing corresponding to horizontal clefting. Negative features were non-visible papillae, disarrangement of the epidermal layer, and cerebriform nests. Multivariate discriminant analysis on the training set (excluding the BCCs) identified seven independently significant features for MM diagnosis. The diagnostic accuracy of the MM algorithm on the test set was 87.6% sensitivity, 70.8% specificity. The four invasive MMs that were misdiagnosed by RCM were all of nevoid subtype. RCM is a highly accurate non-invasive technique for BCC diagnosis. Good diagnostic accuracy was achieved also for MM diagnosis, although rare variants of melanocytic tumors may limit the strict application of the algorithm.     

A meta‐analysis of reflectance confocal microscopy for the diagnosis of malignant skin tumours

Early diagnosis is extremely important for treatment and prognosis of skin cancer. Reflectance confocal microscopy (RCM) is a recently developed technique used to diagnose skin cancer. This meta‐analysis was carried out to assess the accuracy of RCM for the diagnosis of malignant skin tumours. We conducted a systematic literature search of EMBASE, PubMed, the Cochrane Library and Web of Science database for relevant articles in English published up to 24 December 2015. The quality of the included studies was assessed using the QUADAS‐2 tool. Statistical analyses were conducted using the software Meta‐Disc version 1.4 and STATA version 12.0. A total of 21 studies involving 3108 patients with a total of 3602 lesions were included in the per‐lesion analysis. The corresponding pooled results for sensitivity and specificity were 93.6% (95% CI: 0.92–0.95) and 82.7% (95% CI: 0.81–0.84) respectively. Positive likelihood ratio and negative likelihood ratio were 5.84 (95% CI: 4.27–7.98) and 0.08 (95% CI: 0.07–0.10) respectively. Subgroup analysis showed that RCM had a sensitivity of 92.7% (95% CI: 0.90–0.95) and a specificity of 78.3% (95% CI: 0.76–0.81) for detecting melanoma. The pooled sensitivity and specificity of RCM for detecting basal cell carcinoma were 91.7% (95% CI: 0.87–0.95) and 91.3% (95% CI: 0.94–0.96) respectively. RCM is a valid method of identifying malignant skin tumours accurately.