Figure - available from: Journal of Skin Cancer
This content is subject to copyright. Terms and conditions apply.
(a) In situ melanoma arising within a congenital lesion, 4 mm, abdomen, 34-year-old male. Peripheral pink with some eccentric pink centrally: in situ melanoma, 4 mm wide, right abdomen, arising within a superficial compound congenital nevus; pink rim sign is present (arrows) just external to irregular network lacking pink. Note negative network and peripheral globules that are fairly uniform and symmetric. (b) Pink throughout lesion: benign nevus, not biopsied, 5 mm nevus on upper arm in 35-year-old female; pink is present at rim but is most prominent centrally (arrows). Pink rim sign is absent. Note similar architecture and symmetrically arranged globules in both lesions.
Source publication
Background. In dermoscopic images, multiple shades of pink have been described in melanoma without specifying location of these areas within the lesion. Objective. The purpose of this study was to determine the statistics for the presence of centrally and peripherally located pink melanoma and benign melanocytic lesions. Methods. Three observers, u...
Citations
... ModelDerm misclassified a malignant lesion as benign, and examination of the corresponding counterfactuals revealed attributes such as darker pigmentation of the lesion and absence of erythema as influential for this decision (Fig. 4b). However, dermatologists would not typically associate darker pigmentation with decreased likelihood of melanoma, and the distribution of erythema does not match the 'pink rim' sometimes associated with melanoma 38 . ...
The inferences of most machine-learning models powering medical artificial intelligence are difficult to interpret. Here we report a general framework for model auditing that combines insights from medical experts with a highly expressive form of explainable artificial intelligence. Specifically, we leveraged the expertise of dermatologists for the clinical task of differentiating melanomas from melanoma ‘lookalikes’ on the basis of dermoscopic and clinical images of the skin, and the power of generative models to render ‘counterfactual’ images to understand the ‘reasoning’ processes of five medical-image classifiers. By altering image attributes to produce analogous images that elicit a different prediction by the classifiers, and by asking physicians to identify medically meaningful features in the images, the counterfactual images revealed that the classifiers rely both on features used by human dermatologists, such as lesional pigmentation patterns, and on undesirable features, such as background skin texture and colour balance. The framework can be applied to any specialized medical domain to make the powerful inference processes of machine-learning models medically understandable.
... This resembles the pink rim of the sun at sunset or the lunar eclipse. 23 19. Raindrops: Raindrops are described in a myriad of conditions, most seen in miliaria crystallina [ Figure 4] where the flaccid vesicles are seen on a nonerythematous base. ...
A aerial and Celestial based eponymous review in dermatology
... In melanoma suggests that the presence of a pink rim in the periphery of a melanocytic lesion on dermoscopy is an indicator of malignancy. 34 ...
Dermoscopy is a very useful technique devised for an earlier diagnosis of skin melanoma with a clinic-pathological correlation. Later it was found to be beneficial for the diagnosis of many other pigmented skin lesions, such as seborrheic keratosis, pigmented basal cell carcinoma, hemangioma, blue nevus, atypical nevus, and mole, which can often clinically simulate melanoma. Of late, its use in general clinical dermatology is growing with the recognition of new and specific patterns in conditions such as hair disorders, inflammatory disorders, and infections/infestation. It is still in the evolving phase and many new signs are described presently. Eponyms are used almost daily in dermatology practice. The eponyms in dermoscopy, trichoscopy, and onychoscopy are based on the imaginative capability of the authors and they have been very much successful in describing them. It becomes easier to memorize and identify the various appearances for early diagnosis and management. In this article we attempt to highlight the various dermoscopic signs described in dermatology.
... First, our results confirm results from a previous study that found that white or hypopigmented areas in the outer three deciles of the lesion were needed for the best discrimination of melanoma vs. benign [38]. Second, we have confirmed the clinical impression that colors within the lesion periphery, specifically the outer lesion deciles, are important in describing benign lesions, since a circumferential "hazy border", with coloration intermediate between that of skin and lesion, implies a benign lesion [32] and a "pink rim" is a sign of a melanoma [39]. Third, the patches or clusters of color that matter most are larger than a single pixel but still relatively small, in the single-digit range, as in the "peppering" of early malignancies [40,41]. ...
A fuzzy logic-based color histogram analysis technique is presented for discriminating benign skin lesions from malignant melanomas in dermoscopy images. The approach extends previous research for utilizing a fuzzy set for skin lesion color for a specified class of skin lesions, using alpha-cut and support set cardinality for quantifying a fuzzy ratio skin lesion color feature. Skin lesion discrimination results are reported for the fuzzy clustering ratio over different regions of the lesion over a data set of 517 dermoscopy images consisting of 175 invasive melanomas and 342 benign lesions. Experimental results show that the fuzzy clustering ratio applied over an eight-connected neighborhood on the outer 25% of the skin lesion with an alpha-cut of 0.08 can recognize 92.6% of melanomas with approximately 13.5% false positive lesions. These results show the critical importance of colors in the lesion periphery. Our fuzzy logic-based description of lesion colors offers relevance to clinical descriptions of malignant melanoma.
... 16 Furthermore, the existence of pink (poorly melanotic) regions is a recent novel tool for the dermoscopic identification of melanoma. 17 However, these features are also present in other malignancies; hence, disambiguation often requires histologic examination. ...
Solitary pink lesions can pose a particular challenge to dermatologists because they may be almost or completely featureless clinically and dermoscopically, previously requiring biopsy to exclude malignancy. However, these lesions usually are not particularly challenging histopathologically. Thus, the incorporation of in vivo reflectance confocal microscopy into the clinical practice, which allows for noninvasive examination of the skin at the cellular level revealing features previously seen only on histopathology, is particularly useful for this subset of clinically difficult lesions.
... The presence of pink in the periphery or rim of a dermoscopic melanocytic lesion image provides a suspicion of malignancy and is referred to a "pink rim" sign, which is a clue to the dermoscopic diagnosis of invasive melanoma. [6] Pore sign Epidermal (epidermoid) cyst can usually be diagnosed clinically as a roughly circular subcutaneous nodule of variable size, of firm consistency, covered by smooth, and normal-colored skin with a central black punctum. The diagnosis of epidermal cyst can easily be made by identifying the punctum, the pore of the follicle from which the cyst has been derived. ...
Authors have attempted to discuss recently described "signs" in the field of dermatology.
... Pink is also present in benign scalp nevi, often throughout the entire lesion [12]. Pink throughout the entire lesion was found in the majority of both melanoma and benign lesions [13]; while pink at the rim of the lesion (pink rim sign) was described as a useful feature in detecting melanoma [13]. ...
... Pink is also present in benign scalp nevi, often throughout the entire lesion [12]. Pink throughout the entire lesion was found in the majority of both melanoma and benign lesions [13]; while pink at the rim of the lesion (pink rim sign) was described as a useful feature in detecting melanoma [13]. ...
... In a previous study [13], we analyzed 1290 lesions, 296 melanomas (170 in situ lesions; 126 invasive lesions), and 994 benign lesions, with regard to three general locations of pink: peripheral only (pink in the outer part of the lesion but not in the central part), central only (pink in the central part of the lesion but not in the outer part), and pink throughout the lesion (present in both the outer and the central part of the lesion) as shown in Figure 1. Data presented in Figure 1 show that, although pink present throughout the lesion is the most common pattern, pink found in the periphery of the lesion is more discriminatory for melanoma, yielding an odds ratio of 2.51 [13]. ...
Early detection of malignant melanoma is an important public health challenge. In the USA, dermatologists are seeing more melanomas at an early stage, before classic melanoma features have become apparent. Pink color is a feature of these early melanomas. If rapid and accurate automatic detection of pink color in these melanomas could be accomplished, there could be significant public health benefits.
Detection of three shades of pink (light pink, dark pink, and orange pink) was accomplished using color analysis techniques in five color planes (red, green, blue, hue, and saturation). Color shade analysis was performed using a logistic regression model trained with an image set of 60 dermoscopic images of melanoma that contained pink areas. Detected pink shade areas were further analyzed with regard to the location within the lesion, average color parameters over the detected areas, and histogram texture features.
Logistic regression analysis of a separate set of 128 melanomas and 128 benign images resulted in up to 87.9% accuracy in discriminating melanoma from benign lesions measured using area under the receiver operating characteristic curve. The accuracy in this model decreased when parameters for individual shades, texture, or shade location within the lesion were omitted.
Texture, color, and lesion location analysis applied to multiple shades of pink can assist in melanoma detection. When any of these three details: color location, shade analysis, or texture analysis were omitted from the model, accuracy in separating melanoma from benign lesions was lowered. Separation of colors into shades and further details that enhance the characterization of these color shades are needed for optimal discrimination of melanoma from benign lesions.
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.
Amelanotic/hypomelanotic melanoma (AHM) may be difficult to diagnose because of a lack of pigmentation. To evaluate whether dermoscopy can be useful for the diagnosis of early AHM, 133 digital dermoscopic images of lesions histopathologically diagnosed as amelanotic/hypomelanotic superficial spreading melanoma with ≤1 mm thickness (AHSSMs) ( n = 27), amelanotic/hypomelanotic non‐melanocytic lesions (AHNMLs) (e.g., seborrhoeic keratosis and basal cell carcinoma) ( n = 79), and amelanotic/hypomelanotic benign melanocytic lesions (AHBMLs) (e.g., compound and dermal nevi) ( n = 27), were dermoscopically assessed by three blinded dermatologists. Using multivariate analysis, we found a significantly increased risk of diagnosing AHSSM versus AHNML and AHBML when the lesion was characterized by the presence of more than one shade of pink (odds ratio [OR] 37.11), irregular dots/globules (OR 23.73), asymmetric pigmentation (OR 8.85), and structureless pattern (OR 7.33). In conclusion, dermoscopy may improve early AHM detection, discriminating AHSSM from AHNMLs.
Despite the proliferation and clinical deployment of artificial intelligence (AI)-based medical software devices, most remain black boxes that are uninterpretable to key stakeholders including patients, physicians, and even the developers of the devices. Here, we present a general model auditing framework that combines insights from medical experts with a highly expressive form of explainable AI that leverages generative models, to understand the reasoning processes of AI devices. We then apply this framework to generate the first thorough, medically interpretable picture of the reasoning processes of machine-learning–based medical image AI. In our synergistic framework, a generative model first renders "counterfactual" medical images, which in essence visually represent the reasoning process of a medical AI device, and then physicians translate these counterfactual images to medically meaningful features. As our use case, we audit five high-profile AI devices in dermatology, an area of particular interest since dermatology AI devices are beginning to achieve deployment globally. We reveal how dermatology AI devices rely both on features used by human dermatologists, such as lesional pigmentation patterns, as well as multiple, previously unreported, potentially undesirable features, such as background skin texture and image color balance. Our study also sets a precedent for the rigorous application of explainable AI to understand AI in any specialized domain and provides a means for practitioners, clinicians, and regulators to uncloak AI's powerful but previously enigmatic reasoning processes in a medically understandable way.
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.
