Automated Segmentation of the Melanocytes in Skin Histopathological Images

Abstract

Abstract-In the diagnosis of skin melanoma by analyzing histopathological images, the detection of the melanocytes in the epidermis area is an important step. However, the detection of melanocytes in the epidermis area is difficult because other keratinocytes that are very similar to the melanocytes are also present. This paper proposes a novel computer-aided technique for segmentation of the melanocytes in the skin histopathological images. In order to reduce the local intensity variant, a mean-shift algorithm is applied for the initial segmentation of the image. A local region recursive segmentation algorithm is then proposed to filter out the candidate nuclei regions based on the domain prior knowledge. To distinguish the melanocytes from other keratinocytes in the epidermis area, a novel descriptor, named local double ellipse descriptor (LDED), is proposed to measure the local features of the candidate regions. The LDED uses two parameters: region ellipticity and local pattern characteristics to distinguish the melanocytes from the candidate nuclei regions. Experimental results on 28 different histopathological images of skin tissue with different zooming factors show that the proposed technique provides a superior performance.Index Terms-Histopathological image analysis, image segmentation, local descriptor, object detection, pattern recognition.

Computer assisted diagnosis of basal cell carcinoma using Z-transform features

“…The authors in [17][18] addressed the problem of melanocyte detection in epidermis area which is necessary for automated melanoma diagnosis systems. The approach in [17] is based on morphological operation and adaptive thresholding for segmentation of cells nuclei.…”

“…Melanocytes are detected by analyzing the gradient vectors of the pixels located on radial lines originating from centroid of the cells nuclei. The method in [18] works by performing mean shift algorithm to decompose the image into homogenous regions and selecting nucleus candidates using a recursive split and merge algorithm. From these candidate regions, false positives are filtered out by an ellipse fitting approach.…”

“…All of the above procedure can be easily applied to the imaginary part of P in equation (7) to obtain the following approximation: (18) which is obtained from the fact that in equation (9) (19) …”

Localization of Diagnostically Relevant Regions of Interest in Whole Slide Images: a Comparative Study

“…One of the major research directions in histopathological image analysis is to develop image features for different problems and image types. Commonly used image features include low-level features (color [9,10,15,16,18,21,[27][28][29][30][31], texture [10-14, 18, 28]), object level features (shape [32][33][34][35][36][37], topology [8,11,14,18,26,31]), and semantic features (statistics [19,26], histograms [28,32], bag-of-words [28]). …”

Epidermis segmentation in skin histopathological images based on thickness measurement and k-means algorithm

“…However, the early diagnosis of skin cancers such as cutaneous melanoma is not trivial, as the malignant melanoma and benign tumors may have similar appearance in their early stages. Although many techniques have been developed for melanoma diagnosis, e.g., epiluminescence microscopy [1] and confocal microscopy [3], which can provide initial diagnosis, the histopathological examination of a whole slide image (WSI) by pathologists remains the gold standard for the diagnosis [4] as the histopathology slides provide a cellular level view of the disease [5].…”