Extraction of skin lesion texture features based on independent component analysis

Background/purpose: During the recent years, many diagnostic methods have been proposed aiming at early detection of malignant melanoma. The texture of skin lesions is an important feature to differentiate melanoma from other types of lesions, and different techniques have been designed to quantify this feature. In this paper, we discuss a new approach based on independent component analysis (ICA) for extraction of texture features of skin lesions in clinical images.
Methods: After preprocessing and segmentation of the images, features that describe the texture of lesions and show high discriminative characteristics are extracted using ICA, and then these features, along with the color features of the lesions, are used to construct a classification module based on support vector machines for the recognition of malignant melanoma vs. benign nevus.
Results: Experimental results showed that combining melanoma and nevus color features with proposed ICA-based texture features led to a classification accuracy of 88.7%.
Conclusion: ICA can be used as an effective tool for quantifying the texture of lesions.     

Differentiation of melanoma from benign mimics using the relative‐color method

Background: Previous studies have successfully classified 86% of malignant melanomas using a relative‐color segmentation method, by feature extraction from photographic images in the automatic identification of skin tumors. These studies were extended by applying the relative‐color method to dermoscopic images of melanoma grouped with melanoma in situ and clark nevus lesions in dermoscopic images allow more control over lighting variations, which contribute to lesion misclassification. Dermoscopic images then enable a more detailed examination of the structure of skin lesions, provide much more structural detail within lesions, and contain visual information that cannot be seen in photographic images. This present work extends the previous studies by applying relative‐color feature extraction to dermoscopic images to differentiate among melanoma, seborrheic keratoses and Reed/Spitz nevi. Objective: To develop a method for automatically differentiating among malignant melanoma, seborrheic keratoses and Reed/Spitz nevi, using digitized, color, dermoscopic images. Methods: Images underwent preprocessing, tumor segmentation, feature extraction and tumor classification. The relative‐color method was used in the segmentation stage. Classification was accomplished by taking the inner products of model tumor feature vectors with test‐image tumor vectors followed by the nearest‐neighbor classification method. Results: The classification rates of melanoma, seborrheic keratoses and Reed/Spitz nevi images mixed together, were 60%, 58.3% and 80%, respectively. Classification of melanoma and Reed/Spitz nevi mixed, were 70% and 90%, respectively. Classification rates were the best when melanoma was being differentiated from seborrheic keratoses. These rates were 100% and 88.9%, respectively. Conclusion: Dermoscopic rather than photographic images were preprocessed, using a hair‐removal technique. They were then converted to relative‐color images, which were segmented using the principal components transform and median split, followed by morphological filtering. After processing, the multi‐dimensional tumor feature space described herein was used to differentiate the tumors. The high success rates for differentiating seborrheic keratoses from melanoma show that the use of dermoscopic images has a strong promise in enabling prescreening, as well as automated assistance and significant improvement in tumor diagnosis in clinics.     

Tissue counter analysis of tissue components in skin biopsies: evaluation using CART (Classification and Regression Trees)

In tissue counter analysis, complex histologic sections are overlaid with regularly distributed measuring masks of equal size and shape, and the digital contents of each mask (or tissue element) are evaluated by gray level, color, and texture parameters. In this study, the feasibility of tissue counter analysis and classification and regression trees for the quantitative evaluation of skin biopsies was assessed. From 100 randomly selected skin biopsies, a learning set of tissue elements was created, differentiating between cellular elements, collagenous elements of the reticular dermis, fatty elements and other tissue components. Classification and regression trees based on the learning set were used to automatically classify tissue elements in samples of normal skin, benign common nevi, malignant melanoma, molluscum contagiosum, seborrheic keratosis, epidermoid cysts, basal cell carcinoma, and scleroderma. The procedure yielded reproducible assessments of the relative amounts of tissue components in various diagnostic groups. Furthermore, a reliable diagnostic separation of molluscum contagiosum versus normal skin and epidermal cysts, benign common nevi versus malignant melanoma, and seborrheic keratosis versus basal cell carcinoma was possible. Tissue counter analysis combined with classification and regression trees may be a suitable approach to the fully automated analysis of histologic sections of skin biopsies.     

Pigmented seborrheic keratoses of the vulva clinically mimicking a malignant melanoma: a clinical, dermoscopic-pathologic case study

The diagnosis of seborrheic keratosis is, in general, a clinical one, but in some cases, the differential diagnosis between pigmented seborrheic keratosis and malignant melanoma is difficult. Dermoscopy may improve the early diagnosis of vulvar melanoma and thus play a role in the preoperative classification of pigmented lesions at this particular site. We report the first case of a pigmented seborrheic keratosis of the vulva clinically mimicking a malignant melanoma, whose dermoscopic features have been investigated together with their pathologic correlates. Dermoscopically our case shows the absence of comedo-like openings and the presence of the pseudo-network. Dermoscopy is therefore a useful method for the differential diagnosis of pigmented lesions even in the vulva.     

Colour histogram analysis for melanoma discrimination in clinical images

Background: Malignant melanoma, the most deadly form of skin cancer, has a good prognosis if treated in the curable early stages. Colour provides critical discriminating information for the diagnosis of malignant melanoma.
Methods: This research introduces a three-dimensional relative colour histogram analysis technique to identify colours characteristic of melanomas and then applies these 'melanoma colours' to differentiate benign skin lesions from melanomas. The relative colour of a skin lesion is determined based on subtracting a representative colour of the surrounding skin from each lesion pixel. A colour mapping for 'melanoma colours' is determined using a training set of images. A percent melanoma colour feature, defined as the percentage of the lesion pixels that are melanoma colours, is used for discriminating melanomas from benign lesions. The technique is evaluated using a clinical image data set of 129 malignant melanomas and 129 benign lesions consisting of 40 seborrheic keratoses and 89 nevocellular nevi.
Results: Using the percent melanoma colour feature for discrimination, experimental results yield correct melanoma and benign lesion discrimination rates of 84.3 and 83.0%, respectively.
Conclusions: The results presented in this work suggest that lesion colour in clinical images is strongly related to the presence of melanoma in that lesion. However, colour information should be combined with other information in order to further reduce the false negative and false positive rates.     

Differential cytology of malignant melanoma

On the basis of 285 fine needle biopsies and smear preparations of malignant melanoma (both primaries and metastases), we present a catalogue of highly characteristic cytologic features: dissociation of melanoma cells, excessive dimension of the tumor cells, characteristic staining of the cytoplasm, structure of the nucleus, features of the nucleoli, aberrations of mitoses and amitoses. With the help of the cytologic catalogue, we are able to differentiate melanoma in a contrasting way from other diseases such as pigmented seborrheic keratosis, basal cell carcinoma, squamous cell carcinoma, and various lymphomas. Thus the differential cytology of malignant melanoma can be referred to for general criteria of malignancy on the one hand, as well as for the diagnosis of the specific tumor type and the epicrisis of uncertain histologic aspects on the other.     

Colour analysis of skin lesion regions for melanoma discrimination in clinical images

Background: Skin lesion colour is an important feature for diagnosing malignant melanoma. Colour histogram analysis over a training set of images has been used to identify colours characteristic of melanoma, i.e., melanoma colours. A percent melanoma colour feature defined as the percentage of the lesion pixels that are melanoma colours has been used as a feature to discriminate melanomas from benign lesions.
Methods: In this research, the colour histogram analysis technique is extended to evaluate skin lesion discrimination based on colour feature calculations in different regions of the skin lesion. The colour features examined include percent melanoma colour and a novel colour clustering ratio. Experiments are performed using clinical images of 129 malignant melanomas and 129 benign lesions consisting of 40 seborrheic keratoses and 89 nevocellular nevi.
Results: Experimental results show improved discrimination capability for feature calculations focused in the lesion boundary region. Specifically, correct melanoma and benign recognition rates are observed as high as 89 and 83%, respectively, for the percent melanoma colour feature computed using only the outermost, uniformly distributed 10% of the lesion's area.
Conclusions: The experimental results show for the features investigated that the region closest to the skin lesion boundary contains the greatest colour discrimination information for lesion screening. Furthermore, the percent melanoma colour feature consistently outperformed the colour clustering ratio for the different skin lesion regions examined. The clinical application of this result is that clustered colours appear to be no more significant than colours of arbitrary distribution within a lesion.     

Quantitative discrimination of pigmented lesions using three-dimensional high-resolution ultrasound reflex transmission imaging

High-resolution ultrasound-reflex transmission imaging is a non-invasive method that can be performed in vivo. We have adapted and refined this technique for skin imaging. Scans can be analyzed to produce objective parameters. Previous work has highlighted sonographic differences between benign and malignant lesions. The aim of this study was to produce and test numerical parameters from ultrasound skin images that would quantify the acoustic differences between common pigmented lesions, which may aid their discrimination from melanoma. We report our findings for randomly selected patients referred from primary care with suspected melanoma. Those subsequently classified as malignant melanoma (MM), seborrheic keratosis (SK), and benign nevi by a consultant dermatologist (n=87) were imaged by high-resolution ultrasound-reflex transmission imaging. Using surrounding normal skin as a control, numerical sonographic parameters were derived for each lesion giving a relative measure of surface sound reflectance, intra-lesional sound reflection, total sound attenuation, and the relative uniformity of each parameter across the tumor. Significant quantitative differences existed between benign and malignant pigmented lesions studied. Sufficient discrimination was produced between MM (n=25), SKs (n=24) and other benign-pigmented lesions (n=38) to potentially reduce the referral of benign tumors by 65% without missing melanoma.     

Diagnostic utility of magnetic resonance imaging in malignant melanoma.

BACKGROUND: Dermatologists sometimes have difficulty in making the diagnosis of a melanocytic tumor. OBJECTIVE: Our purpose was to establish a noninvasive method for the diagnosis of malignant melanoma. We investigated the diagnostic usefulness of magnetic resonance imaging (MRI) in 23 lesions of primary malignant melanoma, metastatic malignant melanoma, and benign pigmented skin tumors. METHODS: The morphologic characteristics and the signal intensity of the tumors were analyzed to differentiate malignant and benign pigmented skin tumors by means of the tumor-to-fat contrast ratio of the signal intensity. RESULTS: The morphologic characteristics of the tumors obtained by MRI were not an absolute criterion for diagnosis of malignant melanoma, but the signal intensity of the tumor assessed by the tumor-to-fat contrast ratio on T2-weighted images clearly differentiated between primary malignant melanoma and benign pigmented skin tumors. CONCLUSION: These findings indicated that MRI is useful for noninvasive diagnosis of malignant melanoma.     

A Possible Case of Eccrine Poroma Arising in Seborrheic Keratosis with Partial Intermediate Malignancy and Tumor of the Follicular Infundibulum

An eighty-year-old woman presented with a eccrine poroma that appeared to have arisen in a seborrheic keratosis lesion on the right side of her abdomen. The tumor consisted of a brownish, flat, elevated lesion with an eroded nodule at its periphery. Histopathological analysis of the flat, elevated lesion revealed benign basaloid and squamoid cells that showed acanthotic upward proliferation, suggesting a seborrheic keratosis. The eroded nodule showed irregular proliferation of slight atypical squamoid cells with, in parts, ductal structures in the upper dermis, suggesting that it was an eccrine poroma with partial intermediate malignancy. In addition to the eroded nodule, an occult tumor clinically accompanied the seborrheic keratosis. Histopathology of the occult tumor revealed fenestrated columns and cords of pale epithelial cells attached to the base of the seborrheic keratosis lesion, which was diagnosed as a tumor of the follicular infundibulum. Although reports of seborrheic keratosis transforming into a sweat gland tumor are rare, the present case suggests that seborrheic keratosis may develop into benign or malignant adnexal tumors, including sweat gland and follicular tumors. Thus, seborrheic keratosis should be considered as a possible precursor of adnexal tumor.     

水通道蛋白3在四种皮肤肿瘤中的表达

目的 探讨水通道蛋白3（AQP3）在四种皮肤肿瘤组织中的表达及意义。方法 应用免疫组织化学方法检测30例脂溢性角化病、15例Bowen病、32例鳞状细胞癌、17例恶性黑素瘤及15例正常人皮肤组织中AQP3的表达。结果 脂溢性角化病、Bowen病、鳞状细胞癌、恶性黑素瘤及正常人表皮组织中均存在AQP3蛋白的表达;脂溢性角化病皮损中AQP3表达水平与正常人对照组差异无统计学意义（P > 0.05）;Bowen病、鳞状细胞癌及恶性黑素瘤皮损中AQP3蛋白表达显著高于正常人对照组（P < 0.01）,其中鳞状细胞癌与恶性黑素瘤的表达最强,均显著高于Bowen病（P < 0.01）,但鳞状细胞癌与恶性黑素瘤比较差异无统计学意义（P > 0.05）。此外,在鳞状细胞癌中AQP3的表达与肿瘤的分化有显著相关性（P < 0.01）;在已转移恶性黑素瘤中AQP3的表达显著高于未转移恶性黑素瘤（P < 0.05）。结论 AQP3在皮肤恶性肿瘤中表达上调。     

Melanocytes express 3G5 surface antigen

The 3G5-reactive ganglioside antigen (3G5 antigen) is expressed on the surface of various cell types including pericytes, pancreatic islet cells, thyroid follicular cells, and cells of the pituitary and the adrenal medulla. Expression on melanocytes has not yet been reported. We examined 148 5-microm cryosections of 12 normal skin samples and 45 skin tumors (21 melanocytic nevi, 8 malignant melanoma primaries, 4 metastases of malignant melanoma, 3 basal cell carcinomas, and 9 pigmented seborrheic keratoses) by triple fluorescence technique with the monoclonal antibody 3G5, DNA fluorochrome, and the anti-melanocytic antibody A103 (Anti-Melan-A). In normal skin, 3G5 reactivity was detected in epidermal melanocytes of 4 of 12 cases with 14.8 +/- 24.1% positive melanocytes; 20 of 21 nevi (72.2 +/- 29.1% positive nevus cells, mean +/- SD), 8 of 8 primary melanomas (83.9 +/- 12.3% positive melanoma cells), and 4 of 4 melanoma metastases (82.5 +/- 6.5% positive melanoma cells) expressed the 3G5 antigen. All tumor cells of investigated basal cell carcinoma or seborrheic keratosis were 3G5 negative. This is the first report of 3G5 antigen expression in melanocytes. The data demonstrate high expression of this ganglioside in the aggregated melanocytes of malignant or benign tumors but low or absent expression in singular melanocytes (normal epidermis, seborrheic keratoses) reflecting a different biologic state.     

Seborrheic keratosislike melanoma with folliculotropism

BACKGROUND: Seborrheic keratosislike melanoma could be one of the most problematic melanoma simulators, and it may be incorrectly treated by electrocautery or cryotherapy. Dermoscopic examination of pigmented tumors improves the diagnostic accuracy in these challenging lesions. In these tumors, numerous comedolike openings are present. OBSERVATIONS: A 34-year-old man was seen for a conspicuous pigmented lesion on his back that clinically resembled a seborrheic keratosis because of the presence of multiple comedolike openings. Findings from dermoscopic examination showed distinct melanoma criteria (atypical pigmented network, asymmetric globules and dots, and a blue-whitish veil), in addition to multiple comedolike openings. Histopathological examination confirmed a peculiar melanoma variant characterized by prominent folliculotropism and minimal radial spreading. This tumor was not associated with chronic sun-damaged skin. CONCLUSION: Dermoscopy was useful in identifying a particular case of seborrheic keratosislike melanoma with folliculotropism, thus avoiding incorrect treatment.     

Cloudy and starry milia‐like cysts: how well do they distinguish seborrheic keratoses from malignant melanomas?

Background Seborrheic keratoses are the most common skin lesions known to contain small white or yellow structures called milia‐like cysts (MLCs). Varied appearances can sometimes make it difficult to differentiate benign lesions from malignant lesions such as melanoma, the deadliest form of skin cancer found in humans. Objective The purpose of this study was to determine the statistical occurrence of MLCs in benign vs. malignant lesions. Methods A medical student with 10 months experience in examining approximately 1000 dermoscopy images and a dermoscopy‐naïve observer analysed contact non‐polarized dermoscopy images of 221 malignant melanomas and 175 seborrheic keratoses for presence of MLCs. Results The observers found two different types of MLCs present: large ones described as cloudy and smaller ones described as starry. Starry MLCs were found to be prevalent in both seborrheic keratoses and melanomas. Cloudy MLCs, however, were found to have 99.1% specificity for seborrheic keratoses among this group of seborrheic keratoses and melanomas. Conclusion Cloudy MLCs can be a useful tool for differentiating between seborrheic keratoses and melanomas.     

Pigmented malignant hidroacanthoma simplex mimicking irritated seborrheic keratosis

Pigmented variant of malignant hidroacanthoma simplex (PMHS) is very rare. We are aware of only two reported cases, all arising in pigmented hidroacanthoma simplex (HS). We report the third case of PMHS arising in a pigmented HS. A 71-year-old-woman presented with a well-demarcated pigmented hyperkeratotic tumor on the right knee resembling irritated seborrheic keratosis. Histopathologic examination of the excised tumor revealed intraepidermal proliferation of atypical polygonal poroid cells forming large, sharply demarcated nests with colonization of dendritic melanocytes. In addition, there were focal changes of a benign pigmented HS and syringofibroadenoma. The key diagnostic features of ductal structures and intracytoplasmic lumina were highlighted by carcinoembryonic antigen and epithelial membrane antigen immunostaining. PMHS should be differentiated from irritated seborrheic keratosis, melanoacanthoma, Bowen's disease and malignant melanoma both clinically and pathologically.     

皮肤科医师与深度卷积神经网络诊断色素痣和脂溢性角化病皮肤镜图像比较

【摘要】 目的 比较深度卷积神经网络（CNN）与皮肤科医师对色素痣和脂溢性角化病的诊断准确率。方法 使用5 094幅色素痣和脂溢性角化病（SK）的皮肤镜图像对CNN网络ResNet?50通过迁移学习进行训练,建立CNN二分类模型,并应用该模型对30幅色素痣和30幅SK的皮肤镜图像进行自动分类。同时,95位经过皮肤镜培训的有经验的皮肤科医师结合临床皮损图片对上述CNN自动分类的60幅皮肤镜图像进行判读。比较二者的诊断准确率,并对错误分类的图像做进一步统计分析。结果 CNN自动分类模型对色素痣和SK的皮肤镜图像的分类准确率分别为100%（30/30）和76.67%（23/30）,总准确率为88.33%（53/60）;95位皮肤科医师的诊断准确率平均值分别为82.98%（25.8/30）和85.96%（24.9/30）,总准确率为84.47%（50.7/60）。CNN自动分类模型与95位皮肤科医师对色素痣和SK的诊断准确率差异无统计学意义（χ2 = 0.38,P > 0.05）。CNN错误分类的皮肤镜图像被分为3类,即特殊类型（如皮损色素含量多、角化明显）,具有典型特征但存在干扰因素,具有典型特征尚找不到错误分类的原因。结论 CNN自动分类模型在色素痣和SK皮肤镜图像的二分类任务中的表现与有经验的皮肤科医师水平相当。CNN错误分类的原因仍需皮肤科医师与人工智能专业人员共同探索。     

Skin lesion classification using relative color features

BACKGROUND/PURPOSE: Clinically, it is difficult to differentiate the early stage of malignant melanoma and certain benign skin lesions due to similarity in appearance. Our research uses image analysis of clinical skin images and relative color-based pattern recognition techniques to enhance the images and improve differentiation of these lesions. METHODS: First, the relative color images were created from digitized photographic images. Then, they were segmented into objects and morphologically filtered. Next, the relative color features were extracted from the objects to form two different feature spaces; a lesion feature space and an object feature space. The two feature spaces serve as two data models to be analyzed to determine the best features. Finally, we used a statistical analysis model of relative color features, which better classifies the various types of skin lesions. RESULTS/CONCLUSIONS: The best features found for differentiation of melanoma and benign skin lesions from this study are area, mean in the red and blue bands, standard deviation in the red and green bands, skewness in the green band, and entropy in the red band. With the relative color feature algorithm developed, the best results were obtained with a multi-layer perceptron neural network model. This showed an overall classification success of 79%, with 70% of the benign lesions successfully classified, and 86% of malignant melanoma successfully classified.     

Regressing seborrheic keratosis - clinically and dermoscopically mimicking a regressing melanoma

The diagnosis of seborrheic keratosis is a clinical diagnosis. In a certain percentage of cases, differential diagnosis between seborrheic keratosis and malignant melanoma is difficult. We describe a case of regressing seborrheic keratosis simulating malignant melanoma. Clinical, dermoscopic and histopathologic examinations were performed for the occurrence of an asymmetric, irregularly demarcated, irregularly pigmented lesion measuring 1.3 x 1.5 cm on the right part of the abdomen in a 76-year-old male Caucasian. In order not to miss melanoma, the excision and histopathologic examination of the lesion with peppering is essential.     

Seborrheic keratosis

Seborrheic keratosis is one of the most common skin tumors. Because this tumor is benign, treatment is not mandatory. However, the lesions are often removed especially for cosmetic reasons. Despite its frequency, many aspects of seborrheic keratosis remain elusive. In the last years new molecular genetic insights into seborrheic keratoses have been gained. The current knowledge about seborrheic keratosis with respect to epidemiology, pathogenesis, diagnosis and therapy is summarized.     

Seborrheic keratoses, solar lentigines, and lichenoid keratoses. Dermatoscopic features and correlation to histology and clinical signs

Evaluation of the three benign lesions discussed here form the basis for dermoscopic evaluation of other pigmented skin lesions. The features of seborrheic keratosis, including [figure: see text] the various forms of fissures, comedo-like openings, and milia-like cysts, often allow easy interpretation of seborrheic keratosis; however, similar structures are commonly associated with melanocytic neoplasms, notably congenital nevi. Understanding solar lentigo and its dermoscopy features allows for the appreciation of pigment networks common in lentiginous melanocytic nevi and melanoma. The lichenoid keratosis is the model for lichenoid inflammation elsewhere, notably in halo nevi, regressing melanoma, and other melanocytic neoplasms with significant host inflammatory reactions.