Lesion border detection in dermoscopy images

Abstract

Background-Dermoscopy is one of the major imaging modalities used in the diagnosis of melanoma and other pigmented skin lesions. Due to the difficulty and subjectivity of human interpretation, computerized analysis of dermoscopy images has become an important research area. One of the most important steps in dermoscopy image analysis is the automated detection of lesion borders.Methods-In this article, we present a systematic overview of the recent border detection methods in the literature paying particular attention to computational issues and evaluation aspects.Conclusion-Common problems with the existing approaches include the acquisition, size, and diagnostic distribution of the test image set, the evaluation of the results, and the inadequate description of the employed methods. Border determination by dermatologists appears to depend upon higher-level knowledge, therefore it is likely that the incorporation of domain knowledge in automated methods will enable them to perform better, especially in sets of images with a variety of diagnoses.

A Fully Automatic Random Walker Segmentation for Skin Lesions in a Supervised Setting

“…We begin by selecting 100 images that pose a challenge to segmentation methods, and call this imageset 'challenging'. These represent images that are often excluded from other studies [3]. An image is considered challenging if one or more of the following conditions is met: 1) the contrast between the skin and lesion is low, 2) there is significant occlusion by either oil or hair, 3) the entire lesion is not visible, 4) the lesion contains variegated colours or 5) the lesion border is not clearly defined.…”

“…As will be seen in section 3, our method employs two of these concepts from [3]: the use of textural information in segmentation and the use of Fisher's separability criterion. Our application of these concepts, however, is substantially different.…”

“…al. [3]. They reduce a dermoscopic image to 20 distinct colour groups, and assign labels to pixels based on the group to which they belong.…”

Depth Data Improves Skin Lesion Segmentation

“…Here, the colour representation of lesions combines the results from different channels of different colour spaces. It includes 1) the Saturation of HSV , 2) a * of CIE Lab, 3) the normalized blue of RGB as lesions are often more prominent in this channel [1] and 4) the Hue of HSV . Hence, each image position is associated with a colour-valued feature vector, as f (x) = (I saturation , I a * , I blue , I hue ) T .…”

“…First, the lesion boundary provides important information for accurate diagnosis. Second, the extraction of other clinical features critically depends on the accuracy of the boundary [1]. Due to reasons such as low contrast between the lesion and its background, artifact inference, etc., segmentation is a very challenging task.…”