Malignant Pigmented Skin Lesions Research Articles

Discrimination of cancerous from benign pigmented skin lesions based on multispectral autofluorescence lifetime imaging dermoscopy and machine learning.

.SignificanceAccurate early diagnosis of malignant skin lesions is critical in providing adequate and timely treatment; unfortunately, initial clinical evaluation of similar-looking benign and malignant skin lesions can result in missed diagnosis of malignant lesions and unnecessary biopsy of benign ones.AimTo develop and validate a label-free and objective image-guided strategy for the clinical evaluation of suspicious pigmented skin lesions based on multispectral autofluorescence lifetime imaging (maFLIM) dermoscopy.ApproachWe tested the hypothesis that maFLIM-derived autofluorescence global features can be used in machine-learning (ML) models to discriminate malignant from benign pigmented skin lesions. Clinical widefield maFLIM dermoscopy imaging of 41 benign and 19 malignant pigmented skin lesions from 30 patients were acquired prior to tissue biopsy sampling. Three different pools of global image-level maFLIM features were extracted: multispectral intensity, time-domain biexponential, and frequency-domain phasor features. The classification potential of each feature pool to discriminate benign versus malignant pigmented skin lesions was evaluated by training quadratic discriminant analysis (QDA) classification models and applying a leave-one-patient-out cross-validation strategy.ResultsClassification performance estimates obtained after unbiased feature selection were as follows: 68% sensitivity and 80% specificity with the phasor feature pool, 84% sensitivity, and 71% specificity with the biexponential feature pool, and 84% sensitivity and 32% specificity with the intensity feature pool. Ensemble combinations of QDA models trained with phasor and biexponential features yielded sensitivity of 84% and specificity of 90%, outperforming all other models considered.ConclusionsSimple classification ML models based on time-resolved (biexponential and phasor) autofluorescence global features extracted from maFLIM dermoscopy images have the potential to provide objective discrimination of malignant from benign pigmented lesions. ML-assisted maFLIM dermoscopy could potentially assist with the clinical evaluation of suspicious lesions and the identification of those patients benefiting the most from biopsy examination.

Non-Invasive Skin Cancer Diagnosis Using Hyperspectral Imaging for In-Situ Clinical Support.

Skin cancer is one of the most common forms of cancer worldwide and its early detection its key to achieve an effective treatment of the lesion. Commonly, skin cancer diagnosis is based on dermatologist expertise and pathological assessment of biopsies. Although there are diagnosis aid systems based on morphological processing algorithms using conventional imaging, currently, these systems have reached their limit and are not able to outperform dermatologists. In this sense, hyperspectral (HS) imaging (HSI) arises as a new non-invasive technology able to facilitate the detection and classification of pigmented skin lesions (PSLs), employing the spectral properties of the captured sample within and beyond the human eye capabilities. This paper presents a research carried out to develop a dermatological acquisition system based on HSI, employing 125 spectral bands captured between 450 and 950 nm. A database composed of 76 HS PSL images from 61 patients was obtained and labeled and classified into benign and malignant classes. A processing framework is proposed for the automatic identification and classification of the PSL based on a combination of unsupervised and supervised algorithms. Sensitivity and specificity results of 87.5% and 100%, respectively, were obtained in the discrimination of malignant and benign PSLs. This preliminary study demonstrates, as a proof-of-concept, the potential of HSI technology to assist dermatologists in the discrimination of benign and malignant PSLs during clinical routine practice using a real-time and non-invasive hand-held device.

Analysis of Collective Human Intelligence for Diagnosis of Pigmented Skin Lesions Harnessed by Gamification Via a Web-Based Training Platform: Simulation Reader Study.

BackgroundThe diagnosis of pigmented skin lesion is error prone and requires domain-specific expertise, which is not readily available in many parts of the world. Collective intelligence could potentially decrease the error rates of nonexperts.ObjectiveThe aim of this study was to evaluate the feasibility and impact of collective intelligence for the detection of skin cancer.MethodsWe created a gamified study platform on a stack of established Web technologies and presented 4216 dermatoscopic images of the most common benign and malignant pigmented skin lesions to 1245 human raters with different levels of experience. Raters were recruited via scientific meetings, mailing lists, and social media posts. Education was self-declared, and domain-specific experience was tested by screening tests. In the target test, the readers had to assign 30 dermatoscopic images to 1 of the 7 disease categories. The readers could repeat the test with different lesions at their own discretion. Collective human intelligence was achieved by sampling answers from multiple readers. The disease category with most votes was regarded as the collective vote per image.ResultsWe collected 111,019 single ratings, with a mean of 25.2 (SD 18.5) ratings per image. As single raters, nonexperts achieved a lower mean accuracy (58.6%) than experts (68.4%; mean difference=−9.4%; 95% CI −10.74% to −8.1%; P<.001). Collectives of nonexperts achieved higher accuracies than single raters, and the improvement increased with the size of the collective. A collective of 4 nonexperts surpassed single nonexperts in accuracy by 6.3% (95% CI 6.1% to 6.6%; P<.001). The accuracy of a collective of 8 nonexperts was 9.7% higher (95% CI 9.5% to 10.29%; P<.001) than that of single nonexperts, an improvement similar to single experts (P=.73). The sensitivity for malignant images increased for nonexperts (66.3% to 77.6%) and experts (64.6% to 79.4%) for answers given faster than the intrarater mean.ConclusionsA high number of raters can be attracted by elements of gamification and Web-based marketing via mailing lists and social media. Nonexperts increase their accuracy to expert level when acting as a collective, and faster answers correspond to higher accuracy. This information could be useful in a teledermatology setting.

Pigmented skin lesions displaying regression features: Dermoscopy and reflectance confocal microscopy criteria for diagnosis.

Melanomas and nevi displaying regression features can be difficult to differentiate. To describe reflectance confocal microscopy features in benign and malignant pigmented skin lesions characterized by regression features in dermoscopy. Observational retrospective study. Inclusion criteria were presence of dermoscopic features of regression; availability of clinical, dermoscopic and RCM imaging; definite histopathologic diagnosis. The study sample comprised 217 lesions; 108 (49.8%) melanomas and 109 were benign lesions, of which 102 (47.0%) nevi and 7 (3.2%) lichen planus-like keratosis (lplk). Patients with melanoma were significantly older than those with benign lesions (61.9±15.4 vs 46.1±14.8; P<0.001) and a higher proportion of melanomas displayed dermoscopic regression structures in more than 50% of lesion surface (n=83/108; 76.9%; P<0.001). On RCM examination, pagetoid cells were significantly more reported in melanoma group, than in benign lesions (86.1% vs 59.6%; P<0.001) and were more frequently widespread distributed (65.6% vs 20.0%; P<0.001) and both dendritic and roundish (36.6% vs 15.4%; P<0.001) in shape. Aspecific architecture at the dermo-epidermal junction (DEJ) was more commonly seen among melanomas than benign lesions (23.1% vs 11.9%; P=0.002) with higher presence of dendritic and both dendritic and roundish atypical cells at the DEJ (28.7% vs 18.3% and 19.4% vs 3.7%; P<0.001, respectively). Focal pagetoid infiltration and ringed or clod patterns were more commonly seen in benign lesion. In conclusion, the correct interpretation of regressing lesions remains a challenge, assessing carefully the extent and characteristics of architectural and cytologic atypia on RCM is an additional piece of the complex puzzle of melanoma diagnosis.

A Deep Learning Approach to Vascular Structure Segmentation in Dermoscopy Colour Images.

Background Atypical vascular pattern is one of the most important features by differentiating between benign and malignant pigmented skin lesions. Detection and analysis of vascular structures is a necessary initial step for skin mole assessment; it is a prerequisite step to provide an accurate outcome for the widely used 7-point checklist diagnostic algorithm. Methods In this research we present a fully automated machine learning approach for segmenting vascular structures in dermoscopy colour images. The U-Net architecture is based on convolutional networks and designed for fast and precise segmentation of images. After preprocessing the images are randomly divided into 146516 patches of 64 × 64 pixels each. Results On the independent validation dataset including 74 images our implemented method showed high segmentation accuracy. For the U-Net convolutional neural network, an average DSC of 0.84, sensitivity 0.85, and specificity 0.81 has been achieved. Conclusion Vascular structures due to small size and similarity to other local structures create enormous difficulties during the segmentation and assessment process. The use of advanced segmentation methods like deep learning, especially convolutional neural networks, has the potential to improve the accuracy of advanced local structure detection.

Palpation as a useful diagnostic tool for skin lesions

Early identification and accurate diagnosis of malignant pigmented skin lesions is essential for effective management and cost containment. The aim was to investigate the additional value of tactile descriptive information from lesion palpation on the diagnostic accuracy of pigmented skin lesions by medical students using computer-driven learning. Sixth year medical students (n = 152) from the University of Adelaide were invited to participate in an online teaching module on pigmented skin lesions. Users were asked to describe, diagnose and manage 15 pigmented skin lesions in three separate case studies based on pertinent clinical history and visual images of the lesions. Tactile descriptive information was then provided and users were asked to reflect on their diagnosis and management. A total of 66 (43%) of the sixth year students successfully completed the online module. Diagnostic accuracy improved significantly with the provision of tactile descriptive information for seborrhoeic keratosis (p = 0.012), basal cell carcinoma (p = 0.001), squamous cell carcinoma (p = 0.02), and dysplastic naevi (p = 0.035). Tactile descriptive information was stated by 23% of medical students to be important in the clinical diagnosis of pigmented skin lesion. Students managed all malignant pigmented skin lesions with either appropriate biopsy or specialist referral. Palpation information about skin lesions offers useful information for improvement of diagnostic accuracy in an online computer learning setting for medical students.

SOX10 Expression in Malignant Melanoma, Carcinoma, and Normal Tissues

Sry-related HMg-Box gene 10 (SOX10) is a nuclear transcription factor that plays an important role in melanocytic cell differentiation. It has been shown to be a sensitive marker of melanoma including spindle and desmoplastic subtypes. We assessed its frequency of expression in melanoma, carcinoma, benign nevi, and non-neoplastic tissues with routine immunohistochemistry for SOX10. The 109 primary melanoma included 49 epithelioid, 19 spindle cell, 22 desmoplastic, and 19 mixed spindle cell/desmoplastic melanoma. All primary, except 8 desmoplastic melanoma, and 11 metastatic melanoma were strongly and diffusely nuclear SOX10-positive. Six desmoplastic melanoma had ≤10% cells positive, and 2 were <50% positive, all of 3+ intensity. Eighteen of 149 (12%) breast carcinoma were SOX10-positive. All 24 ovarian, 23 endometrial, 26 lung, and 25 colon carcinoma were SOX10-negative. All 43 benign nevi, 18 dysplastic nevi, 68 non-neoplastic and benign skins, and all 56 non-neoplastic breast tissue were SOX10-positive. The sensitivity and specificity for SOX10 in the diagnosis of melanoma are 1.0 and 0.93, respectively; the positive and negative predictive values are 0.87 and 1.0, respectively. SOX10 is a sensitive, specific marker for melanoma. As benign nevi also express SOX10, it cannot be used to differentiate between benign and malignant pigmented skin lesions. Only a small number of breast carcinoma (12%), and breast lobules, express SOX10; no carcinoma of the ovary, endometrium, lung, or colon expressed SOX10.

Nodal Combined Blue Nevus and Benign Nevus Cells in Multiple Axillary Sentinel Nodes in a Patient With Breast Carcinoma

Combined blue nevus and benign nevus cells were identified in the same sentinel lymph node. Blue nevus alone was also present in an additional sentinel lymph node in the same axilla in a patient who underwent needle localization, wide local excision, and sentinel lymph node biopsy for her pT1cN1mi(sn)M(na) invasive duct carcinoma of the breast. Of the 4 sentinel lymph nodes, 1 showed micrometastasis and 2 other lymph nodes showed blue nevus involving the capsule and trabeculae of the nodes. The patient had no significant previous clinical history of any skin tumors and had a negative clinical examination for malignant melanoma or pigmented skin lesions after the diagnosis of nodal blue nevus. To our knowledge, this is the first case report of combined blue nevi involving multiple sentinel lymph nodes in the same axilla. An equally interesting finding is the presence of benign nonpigmented nevus cells in continuation with the blue nevus in the same node.

Prehistological evaluation of benign and malignant pigmented skin lesions with optical computed tomography

Discrimination of benign and malignant melanocytic lesions is a major issue in clinical dermatology. Assessment of the thickness of melanoma is critical for prognosis and treatment selection. We aimed to evaluate a novel optical computed tomography (optical-CT) system as a tool for three-dimensional (3-D) imaging of melanocytic lesions and its ability to discriminate benign from malignant melanocytic lesions while simultaneously determining the thickness of invasive melanoma. Seventeen melanocytic lesions, one hemangioma, and normal skin were assessed immediately after their excision by optical-CT and subsequently underwent histopathological examination. Tomographic reconstructions were performed with a back-propagation algorithm calculating a 3-D map of the total attenuation coefficient (AC). There was a statistically significant difference between melanomas, dysplastic nevi, and non-dysplastic nevi, as indicated by Kruskal-Wallis test. Median AC values were higher for melanomas compared with dysplastic and non-dysplastic nevi. No statistically significant difference was observed when thickness values obtained by optical-CT were compared with histological thickness using a Wilcoxon sighed rank test. Our results suggest that optical-CT can be important for the immediate prehistological evaluation of biopsies, assisting the physician for a rapid assessment of malignancy and of the thickness of a melanocytic lesion.

The Assessment of Melanoma Risk Using the Dynamic Infrared Imaging Technique

Melanoma is the deadliest form of skin cancer. Each year more than 53,600 people learn that they have melanoma, and around 8700 people die from melanoma in the United States. Early detection is the key to improving survival in patients with malignant melanoma. We developed a thermal (infrared) imaging system that allows accurate measurements of small temperature differences on the skin surface with the aim to diagnose malignant pigmented skin lesions at an early stage of the disease. The imaging method we developed relies on active infrared imaging and a multimodal image analysis strategy, including involuntary body/limb motion correction and interactive lesion segmentation for detecting malignant lesions. The imaging system described in the paper was tested in a pilot patient study in which patients who possess a pigmented lesion with a clinical indication for biopsy were selected to participate. The lesion and the surrounding healthy skin were cooled by air at 15 °C for 30–60 s, and the thermal recovery was imaged with the infrared camera after the removal of this cooling stress. We found that the benign lesions have a thermal recovery similar to normal skin, whereas the thermal recovery of the melanoma lesion is different. It was observed that a malignant skin lesion has a higher temperature than healthy skin during the thermal recovery process (up to 2.2 °C higher for a Clark’s level II melanoma). The present study shows the feasibility of dynamic thermal imaging in distinguishing malignant pigmented lesions from benign, look-alike pigmented lesions.

Melanoacanthoma Simulating Pigmented Spitz Nevus: an Unusual Dermoscopy Pitfall

The starburst pattern is the dermoscopic hallmark of pigmented Spitz nevus, although it has been rarely observed in melanoma as well. To describe a case of melanoacanthoma simulating pigmented Spitz nevus. Clinical, dermoscopic, and histopathologic examinations were performed for the occurrence of a 4-mm pigmented skin lesion on the hip of a 38-year-old Caucasian woman. Dermoscopy examination of the lesion disclosed a stereotypical starburst pattern characterized by pigmented streaks symmetrically distributed at the periphery. A preoperative diagnosis of pigmented Spitz nevus was made, and the lesion was excised. However, subsequent histopathologic examination revealed a melanoacanthoma. The starburst pattern, although diagnostic for pigmented Spitz nevus, can be rarely observed in other benign or malignant pigmented skin lesions. Accordingly, all lesions in adults exhibiting a starburst pattern or other spitzoid features should be excised for histopathologic evaluation.

Melanoacanthoma Simulating Pigmented Spitz Nevus

BACKGROUND The starburst pattern is the dermoscopic hallmark of pigmented Spitz nevus, although it has been rarely observed in melanoma as well. OBJECTIVE To describe a case of melanoacanthoma simulating pigmented Spitz nevus. MATERIAL AND METHODS Clinical, dermoscopic, and histopathologic examinations were performed for the occurrence of a 4-mm pigmented skin lesion on the hip of a 38-year-old Caucasian woman. RESULTS Dermoscopy examination of the lesion disclosed a stereotypical starburst pattern characterized by pigmented streaks symmetrically distributed at the periphery. A preoperative diagnosis of pigmented Spitz nevus was made, and the lesion was excised. However, subsequent histopathologic examination revealed a melanoacanthoma. CONCLUSION The starburst pattern, although diagnostic for pigmented Spitz nevus, can be rarely observed in other benign or malignant pigmented skin lesions. Accordingly, all lesions in adults exhibiting a starburst pattern or other spitzoid features should be excised for histopathologic evaluation.

Digital dermoscopy analysis of atypical pigmented skin lesions: a stepwise logistic discriminant analysis approach.

Digital microscopy is a non-invasive diagnostic technique enabling determination of characteristics that cannot be appreciated by direct observation. If correctly applied, this technique can be useful for the diagnosis of pigmented skin lesions. To evaluate the utility of digital microscopy for analysing atypical benign and malignant pigmented skin lesions exploiting digital numerical filtering and automatic measurements. Forty-eight parameters were identified as possible discriminating variables, and were grouped in four categories: geometries, colours, textures, and islands of colour. Statistical analysis was used to identify the variables with the highest discriminating power. The high quality of the digital image made it possible to observe diagnostic signs in pigmented skin lesion images, acquired by the present technique, in great detail. Specially designed filtering enhanced certain diagnostic patterns. Stepwise discriminant analysis selected only 10 variables (the means of these variables were higher in melanomas than in nevi). The combined use of digital dermoscopy and stepwise logistic discriminant analysis made it possible to single out the best objective variables for distinguishing atypical nevi and early melanoma.

Diagnostic potential of fluorescence of formalin-fixed paraffin-embedded malignant melanoma and pigmented skin lesions: quantitative study of fluorescence intensity using fluorescence microscope and digital imaging.

The background for this study was reports in the literature of stronger fluorescence observed visually for melanomas compared with benign naevi in formalin-fixed paraffin-embedded sections. Our objective was to carry out a quantitative study of the phenomenon and to investigate if such an approach could be used in the detection of melanomas. Microscopic digital imaging was used to measure quantitatively the fluorescence intensity in specimens from 50 malignant melanomas, four basal cell carcinomas and 58 benign lesions. The mean fluorescence intensity of the melanomas was considerably higher than of the other lesions. For melanomas, the intensity depended both on the distance from the skin surface and the distance from the centre of the lesion. A simple algorithm based on the intensity threshold correctly classified the melanomas with a sensitivity of 74% and a specificity of 59%. Quantitative measurements of the fluorescence of the pigmented skin lesions fixed with formalin and embedded in paraffin can be a useful auxiliary tool for differentiating melanoma from other pigmented lesions histopathologically.

Confocal scanning laser microscopy of benign and malignant melanocytic skin lesions in vivo

Background: The ability of physicians for early diagnosis of cutaneous melanomas is less than perfect, prompting research into noninvasive methods for diagnosis. Objective: Our purpose was to evaluate confocal scanning laser microscopy (CSLM) for noninvasive imaging of benign and malignant melanocytic lesions in vivo. Methods: Forty pigmented skin lesions (including adjacent normal skin as control) in vivo were imaged with near-infrared CSLM. The confocal images were correlated to histopathology. Results: Nuclear, cellular, and architectural detail in the epidermis and superficial dermis is imaged with high resolution and contrast. Melanin causes the cytoplasm of pigmented cells to appear bright. Melanocytic nevi had cohesive nests of uniformly circular cells and increased microvascular blood flow. Melanomas had a polymorphous cytologic structure, containing atypical, pleomorphic cells in disarray and irregular dendritic cells. Conclusion: CSLM is capable of identifying distinct patterns and cytologic features of benign and malignant pigmented skin lesions in vivo. CSLM may be useful to noninvasively discriminate benign and malignant lesions in vivo. (J Am Acad Dermatol 2001;45:365-76.)

TYPICAL DERMOSCOPIC PATTERNS OF BENIGN MELANOCYTIC NEVI

Dermoscopy often is an effective diagnostic tool for differentiating benign and malignant pigmented skin lesions. The observed dermoscopic structures and patterns of lesions provide valuable sources of information that are not accessible by routine clinical examination. Recognition and interpretation of these dermoscopic structures and patterns require training and experience. Mentioned here are some fundamental guidelines for differentiating malignant and benign lesions. Some of the classic patterns associated with a list of benign melanocytic nevi also have been illustrated. Additional research is needed to identify other patterns of these benign lesions. More importantly, analysis of the prevalence of various patterns associated with the various benign lesions will be helpful to clinicians having the responsibility of making in vivo diagnoses of benign compared with malignant melanocytic neoplasms.

IN VIVO CONFOCAL MICROSCOPY IN DERMATOLOGY

Confocal microscopy is an optical imaging tool that allows for high resolution, noninvasive imaging in vivo. Thin sections of human tissue can be imaged allowing visualization of cellular and nuclear detail without biopsy. This technique recently has been used to image benign and malignant pigmented skin lesions, nonmelanoma skin cancer, inflammatory skin conditions, and dynamic skin processes.

Precision of automatic measurements of pigmented skin lesion parameters with a MelaFind(TM) multispectral digital dermoscope.

The purpose of this study was to assess the precision of automatic computerized measurement of parameters that may be useful in the differentiation of malignant melanoma from benign pigmented skin lesions, and also to determine the feasibility of quantitative monitoring of skin lesions over time. Ten independent sequences of images were acquired with a MelaFind multispectral digital dermoscope for each of 12 benign or malignant pigmented skin lesions. The sequences of images were processed automatically to provide 10 independent measurements of the various parameters for each lesion. Parameters included lesion area, greatest 'diameter', perimeter, reflectance and asymmetry. The precision of each parameter determination was computed from the mean and standard deviation of the 10 measurements of that parameter. The relative errors in determining the lesion area, 'diameter' and perimeter were found to be 6%, 3% and 4%, respectively. Other lesion parameters that are used in differentiating melanomas from benign skin lesions were also analysed as a function of wavelength. In the blue band (about 430 nm) the relative error was about 7% for the mean lesion reflectance and about 7% for the asymmetry parameter. These results demonstrate the feasibility of using MelaFind for objective quantitative monitoring of changes in pigmented skin lesions over time. As suggested by some studies, such information is useful in the early detection of malignant melanoma. The results show that parameters obtained automatically from MelaFind images are sufficiently precise to allow pertinent parameters to be used to classify pigmented skin lesions.

Using the Internet to teach melanoma management guidelines to primary care physicians.

The aim of this study was to determine whether a brief Internet-based education programme could improve physicians' abilities to manage pigmented skin lesions. A pre-test-post-test assessment was used of subjects' knowledge of skin cancer, confidence in their management abilities and actual ability to recommend appropriate treatment for 20 hypothetical patients with pigmented skin lesions. The setting was the general medicine service of an academic medical centre. Seventeen volunteer medical students, house officers and faculty members took part in the study. Following the pre-test, subjects completed a 1-hour computer-based educational programme, distributed via the Internet, presenting a guideline for recognizing and managing potentially malignant pigmented skin lesions. The guideline was based on the ABCD rule and the Glasgow seven-point checklist. The educational programme had a positive effect on the subjects' overall skin cancer knowledge and had significantly positive effects on their confidence and ability to apply the management guideline. Based on the guideline criteria, the subjects made the correct management decision on the clinical scenarios 63.2% of the time before the programme and 74.1% of the time after the programme (P = 0.002). We were able to teach melanoma management guidelines to physicians and medical students using a brief, interactive computer programme distributed via the Internet. Such an approach is more cost-effective than classroom teaching and could be used to improve the clinical skills of practising physicians to recognize and manage early melanomas. This approach to distributed learning could also be used to teach other clinical guidelines to physicians.

Accuracy of epiluminescence microscopy among practical dermatologists: a study from the Emilia-Romagna region of Italy.

The accuracy of epiluminescence microscopy (ELM) for pigmented skin lesions (PSLs) is still unclear. The large scale utilization of the technique is generally discouraged. The present study was aimed at comparing the accuracy of ELM with that of clinical examination alone in a group of 20 practical dermatologists. Thirty digital clinical images of benign and malignant PSLs and their digital ELM counterparts were used. A set of cumulative accuracy measures reported in the combined clinical/ELM diagnosis was compared with that reported in the clinical diagnosis alone. The proportion of nonmelanocytic lesions (NMLs), melanocytic nevi, and melanomas correctly identified, the predictive value of such diagnoses, and the proportion of melanomas detected by referral for biopsy (irrespective of the diagnosis reported) increased significantly. The cases referred for biopsy despite a "benign" impression decreased significantly among NMLs. The observed tendency towards a greater accuracy suggests that even in the routine dermatology practice ELM has the potential to improve the clinical examination of PSLs.