Classification of Dermoscopic Image of Skin Cancer Using the GLCM Method and Multi-SVM Algorithm

The development of abnormal skin pigment cells can cause a skin cancer called melanoma. Melanoma can be cured if diagnosed and treated in its early stages. Various studies using various technologies have been developed to conduct early detection of melanoma. This research was conducted to diagnose melanoma skin cancer with digital image processing techniques on the dermoscopic image of skin cancer. The diagnosis is made by classifying dermoscopic images based on the types of Common Nevus, Atypical Nevus or Melanoma. Pre-processing is done by changing the RGB image to grayscale (grayscaling), smoothing image using median filtering, and image segmentation based on binary images of skin lesions. The value of Contrast, Correlation, Energy and Homogeneity obtained from the texture feature extraction of the GLCM method is used in the next step, which is the classification process with the Multi-SVM algorithm. The proposed research method shows high accuracy results in diagnosing skin cancer

Sharing is Caring – A Systematic Review on how Frequently Dermoscopic Images are Shared in Inter-observer Investigations

Research interest in dermoscopy has accelerated, but the complete dermoscopic image sets used for inter-observer investigations for skin tumors are not often shared to the reader. The aim of this systematic review was to analyze what proportion of images depicting skin tumors are shared in the manuscripts of studies investigating inter-observer variation in the assessment of dermoscopic features and/or patterns. The Embase, MEDLINE, and Scopus databases were screened for eligible studies published from inception to July 2, 2020. For included investigations we extracted the proportion of lesion images presented in the manuscripts and or supplements. Overall, we included 61 studies (52 original investigations and 9 concise reports) in the time period of 1997 to 2020. These investigations combined, included 14,124 skin tumors of which 373 (3%) images were shared. Since data sharing must be promoted, this investigation should be a wake-up call for the dermatology research community and editorial offices.

Skin lesion segmentation method for dermoscopic images with convolutional neural networks and semantic segmentation

Melanoma skin cancer is one of the most dangerous forms of skin cancer because it grows fast and causes most of the skin cancer deaths. Hence, early detection is a very important task to treat melanoma. In this article, we propose a skin lesion segmentation method for dermoscopic images based on the U-Net architecture with VGG-16 encoder and the semantic segmentation. Base on the segmented skin lesion, diagnostic imaging systems can evaluate skin lesion features to classify them. The proposed method requires fewer resources for training, and it is suitable for computing systems without powerful GPUs, but the training accuracy is still high enough (above 95 %). In the experiments, we train the model on the ISIC dataset – a common dermoscopic image dataset. To assess the performance of the proposed skin lesion segmentation method, we evaluate the Sorensen-Dice and the Jaccard scores and compare to other deep learning-based skin lesion segmentation methods. Experimental results showed that skin lesion segmentation quality of the proposed method are better than ones of the compared methods.

Scale and Pustule on Dermoscopy of Rosacea: A Diagnostic Clue for Demodex Species

Background: Demodex mites are highly found in the skin of patients with rosacea.The diagnosis of Demodex can be made by standardized skin surface biopsy. Dermoscopy is a tool used in the noninvasive diagnosis of various dermatological diseases. Objectives: To determine whether dermoscopic features of demodicosis are associated with the result of standardized skin surface biopsy in patients with rosacea and to compare dermoscopic features of rosacea in Demodex-positive and negative samples and Demodex type. Methods: A total of 30 patients (7 male, 23 female) were included in the study. Dermoscopic examination was performed on both the clinically most severely affected areas and adjacent healthy skin. The skin surface biopsy sample was taken from the same place from where the dermoscopic image was taken. Results: A total of 83 (lesion n = 60, non-lesion n = 23) areas were evaluated. Demodex was detected in 60.2% (n = 50) of the samples. Half of these samples revealed only Demodex folliculorum, and the remaining half revealed D folliculorum and Demodex brevis. Of theDemodex-positive samples, 88% had Demodex tails (P =0.001) and68% Demodex follicular openings (P = 0.002) on dermoscopy. In D folliculorum+D brevis-positive samples, the rate of scale and pustule was higher than D folliculorum-positive samples (P = 0.017 and P = 0032,respectively). Conclusions: The sensitivity and specificity of Demodex tail are higher than Demodex follicular opening and scale and pustule detection with dermoscopy and may indicate the coexistence of both D folliculorum and D brevis.