Detection of melanoma from dermoscopic images of naevi acquired under uncontrolled conditions

Deep Learning Techniques for the Dermoscopic Differential Diagnosis of Benign/Malignant Melanocytic Skin Lesions: From the Past to the Present

There has been growing scientific interest in the research field of deep learning techniques applied to skin cancer diagnosis in the last decade. Though encouraging data have been globally reported, several discrepancies have been observed in terms of study methodology, result presentations and validation in clinical settings. The present review aimed to screen the scientific literature on the application of DL techniques to dermoscopic melanoma/nevi differential diagnosis and extrapolate those original studies adequately by reporting on a DL model, comparing them among clinicians and/or another DL architecture. The second aim was to examine those studies together according to a standard set of statistical measures, and the third was to provide dermatologists with a comprehensive explanation and definition of the most used artificial intelligence (AI) terms to better/further understand the scientific literature on this topic and, in parallel, to be updated on the newest applications in the medical dermatologic field, along with a historical perspective. After screening nearly 2000 records, a subset of 54 was selected. Comparing the 20 studies reporting on convolutional neural network (CNN)/deep convolutional neural network (DCNN) models, we have a scenario of highly performant DL algorithms, especially in terms of low false positive results, with average values of accuracy (83.99%), sensitivity (77.74%), and specificity (80.61%). Looking at the comparison with diagnoses by clinicians (13 studies), the main difference relies on the specificity values, with a +15.63% increase for the CNN/DCNN models (average specificity of 84.87%) compared to humans (average specificity of 64.24%) with a 14,85% gap in average accuracy; the sensitivity values were comparable (79.77% for DL and 79.78% for humans). To obtain higher diagnostic accuracy and feasibility in clinical practice, rather than in experimental retrospective settings, future DL models should be based on a large dataset integrating dermoscopic images with relevant clinical and anamnestic data that is prospectively tested and adequately compared with physicians.

A systematic review and meta-analysis of artificial intelligence versus clinicians for skin cancer diagnosis

Scientific research of artificial intelligence (AI) in dermatology has increased exponentially. The objective of this study was to perform a systematic review and meta-analysis to evaluate the performance of AI algorithms for skin cancer classification in comparison to clinicians with different levels of expertise. Based on PRISMA guidelines, 3 electronic databases (PubMed, Embase, and Cochrane Library) were screened for relevant articles up to August 2022. The quality of the studies was assessed using QUADAS-2. A meta-analysis of sensitivity and specificity was performed for the accuracy of AI and clinicians. Fifty-three studies were included in the systematic review, and 19 met the inclusion criteria for the meta-analysis. Considering all studies and all subgroups of clinicians, we found a sensitivity (Sn) and specificity (Sp) of 87.0% and 77.1% for AI algorithms, respectively, and a Sn of 79.78% and Sp of 73.6% for all clinicians (overall); differences were statistically significant for both Sn and Sp. The difference between AI performance (Sn 92.5%, Sp 66.5%) vs. generalists (Sn 64.6%, Sp 72.8%), was greater, when compared with expert clinicians. Performance between AI algorithms (Sn 86.3%, Sp 78.4%) vs expert dermatologists (Sn 84.2%, Sp 74.4%) was clinically comparable. Limitations of AI algorithms in clinical practice should be considered, and future studies should focus on real-world settings, and towards AI-assistance.

Artificial intelligence-assisted dermatology diagnosis: From unimodal to multimodal

Artificial Intelligence (AI) is progressively permeating medicine, notably in the realm of assisted diagnosis. However, the traditional unimodal AI models, reliant on large volumes of accurately labeled data and single data type usage, prove insufficient to assist dermatological diagnosis. Augmenting these models with text data from patient narratives, laboratory reports, and image data from skin lesions, dermoscopy, and pathologies could significantly enhance their diagnostic capacity. Large-scale pre-training multimodal models offer a promising solution, exploiting the burgeoning reservoir of clinical data and amalgamating various data types. This paper delves into unimodal models' methodologies, applications, and shortcomings while exploring how multimodal models can enhance accuracy and reliability. Furthermore, integrating cutting-edge technologies like federated learning and multi-party privacy computing with AI can substantially mitigate patient privacy concerns in dermatological datasets and further fosters a move towards high-precision self-diagnosis. Diagnostic systems underpinned by large-scale pre-training multimodal models can facilitate dermatology physicians in formulating effective diagnostic and treatment strategies and herald a transformative era in healthcare.

A Cloud Approach for Melanoma Detection based on Deep Learning Networks

In the era of digitized images, the goal is to be able to extract information from them and create new knowledge thanks to the use of Computer Vision techniques, Machine Learning and Deep Learning. This allows their use for early diagnosis and subsequent determination of the treatment of many pathologies. In the specific case treated here, deep neural networks are used in the dermatological field to distinguish between melanoma and non-melanoma images. In this work we have underlined two essential points of melanoma detection research. The first aspect taken into consideration is how even a simple modification of the parameters in the dataset determines a change of the accuracy of the classifiers, while working on the same original dataset. The second point is the need to have a system architecture that can be more flexible in updating the training datasets for the classification of this pathology. In this context, the proposed architecture reserves the goal of developing and implementing a hybrid architecture based on Cloud, Fog and Edge Computing in order to provide a Melanoma Detection service based on clinical and/or dermoscopic images. At the same time, this architecture must be able to interface with the amount of data to be analyzed by reducing the running time of the necessary computational operations. This has been highlighted with experiments carried out on a single machine and on different distribution systems, highlighting how a distributed approach guarantees the achievement of an output in a much more acceptable time without the need to fully rely on data scientists skills.

Towards the automatic detection of skin lesion shape asymmetry, color variegation and diameter in dermoscopic images

Asymmetry, color variegation and diameter are considered strong indicators of malignant melanoma. The subjectivity inherent in the first two features and the fact that 10% of melanomas tend to be missed in the early diagnosis due to having a diameter less than 6mm, deem it necessary to develop an objective computer vision system to evaluate these criteria and aid in the early detection of melanoma which could eventually lead to a higher 5-year survival rate. This paper proposes an approach for evaluating the three criteria objectively, whereby we develop a measure to find asymmetry with the aid of a decision tree which we train on the extracted asymmetry measures and then use to predict the asymmetry of new skin lesion images. A range of colors that demonstrate the suspicious colors for the color variegation feature have been derived, and Feret's diameter has been utilized to find the diameter of the skin lesion. The decision tree is 80% accurate in determining the asymmetry of skin lesions, and the number of suspicious colors and diameter values are objectively identified.

Characterizing the role of dermatologists in developing artificial intelligence for assessment of skin cancer: A systematic review

BACKGROUND

The use of artificial intelligence (AI) for skin cancer assessment has been an emerging topic in dermatology. Leadership of dermatologists is necessary in defining how these technologies fit into clinical practice.

OBJECTIVE

To characterize the evolution of AI in skin cancer assessment and characterize the involvement of dermatologists in developing these technologies.

METHODS

An electronic literature search was performed using PubMed by searching machine learning or artificial intelligence combined with skin cancer or melanoma. Articles were included if they used AI for screening and diagnosis of skin cancer using data sets consisting of dermoscopic images or photographs of gross lesions.

RESULTS

Fifty-one articles were included, and 41% of these had dermatologists included as authors. Articles that included dermatologists described algorithms built with more images versus articles that did not include dermatologists (mean, 12,111 vs 660 images, respectively). In terms of underlying technology, AI used for skin cancer assessment has followed trends in the field of image recognition.

LIMITATIONS

This review focused on models described in the medical literature and did not account for those described elsewhere.

CONCLUSIONS

Greater involvement of dermatologists is needed in thinking through issues in data collection, data set biases, and applications of technology. Dermatologists can provide access to large, diverse data sets that are increasingly important for building these models.

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.

Performance of a dermoscopy-based computer vision system for the diagnosis of pigmented skin lesions compared with visual evaluation by experienced dermatologists

BACKGROUND

It is often difficult to differentiate early melanomas from benign melanocytic nevi even by expert dermatologists, and the task is even more challenging for primary care physicians untrained in dermatology and dermoscopy. A computer system can provide an objective and quantitative evaluation of skin lesions, reducing subjectivity in the diagnosis.

OBJECTIVE

Our objective is to make a low-cost computer aided diagnostic tool applicable in primary care based on a consumer grade camera with attached dermatoscope, and compare its performance to that of experienced dermatologists.

METHODS AND MATERIALS

We propose several new image-derived features computed from automatically segmented dermoscopic pictures. These are related to the asymmetry, color, border, geometry, and texture of skin lesions. The diagnostic accuracy of the system is compared with that of three dermatologists.

RESULTS

With a data set of 206 skin lesions, 169 benign and 37 melanomas, the classifier was able to provide competitive sensitivity (86%) and specificity (52%) scores compared with the sensitivity (85%) and specificity (48%) of the most accurate dermatologist using only dermoscopic images.

CONCLUSION

We show that simple statistical classifiers can be trained to provide a recommendation on whether a pigmented skin lesion requires biopsy to exclude skin cancer with a performance that is comparable to and exceeds that of experienced dermatologists.

Quantitative color assessment of dermoscopy images using perceptible color regions

BACKGROUND

Dermoscopy is a non-invasive in vivo skin imaging technique that assists dermatologists in diagnosing melanoma. However, the use of dermoscopy for diagnosis requires extensive training since this approach often provides extremely complex and subjective information. The presence of an imperceptible color difference in dermoscopy images is one of the serious problems associated with the use of this technique. This imperceptible color difference leads to inaccurate lesion extraction at the borders and hinders the assessment of lesion features. Therefore, objective and quantitative assessment based on perceptible color differences is important for the diagnosis of melanoma using dermoscopy.

METHODS

In this study, we developed a method for assessing colors in a lesion. Twenty-seven perceptible color regions based on the multi-thresholding method in each color channel were constructed, and dominant color region (DCR), bluish dominant region (BDR), and the number of colors were assessed as three diagnostic parameters from these perceptible color regions on 150 dermoscopy images.

RESULTS/CONCLUSION

Diagnostic accuracy was calculated by combination of three diagnostic parameters derived from DCR, BDR, and the number of colors. Diagnostic accuracy with 73.33% sensitivity and 90.67% specificity was obtained in case of positive features in more than two parameters.

Automatic segmentation of dermoscopic images by iterative classification

Accurate detection of the borders of skin lesions is a vital first step for computer aided diagnostic systems. This paper presents a novel automatic approach to segmentation of skin lesions that is particularly suitable for analysis of dermoscopic images. Assumptions about the image acquisition, in particular, the approximate location and color, are used to derive an automatic rule to select small seed regions, likely to correspond to samples of skin and the lesion of interest. The seed regions are used as initial training samples, and the lesion segmentation problem is treated as binary classification problem. An iterative hybrid classification strategy, based on a weighted combination of estimated posteriors of a linear and quadratic classifier, is used to update both the automatically selected training samples and the segmentation, increasing reliability and final accuracy, especially for those challenging images, where the contrast between the background skin and lesion is low.