Four-Class Classification of Skin Lesions With Task Decomposition Strategy

Abstract

This paper proposes a new computer-aided method for the skin lesion classification applicable to both melanocytic skin lesions (MSLs) and nonmelanocytic skin lesions (NoMSLs). The computer-aided skin lesion classification has drawn attention as an aid for detection of skin cancers. Several researchers have developed methods to distinguish between melanoma and nevus, which are both categorized as MSL. However, most of these studies did not focus on NoMSLs such as basal cell carcinoma (BCC), the most common skin cancer and seborrheic keratosis (SK) despite their high incidence rates. It is preferable to deal with these NoMSLs as well as MSLs especially for the potential users who are not enough capable of diagnosing pigmented skin lesions on their own such as dermatologists in training and physicians with different expertise. We developed a new method to distinguish among melanomas, nevi, BCCs, and SKs. Our method calculates 828 candidate features grouped into three categories: color, subregion, and texture. We introduced two types of classification models: a layered model that uses a task decomposition strategy and flat models to serve as performance baselines. We tested our methods on 964 dermoscopy images: 105 melanomas, 692 nevi, 69 BCCs, and 98 SKs. The layered model outperformed the flat models, achieving detection rates of 90.48%, 82.51%, 82.61%, and 80.61% for melanomas, nevi, BCCs, and SKs, respectively. We also identified specific features effective for the classification task including irregularity of color distribution. The results show promise for enhancing the capability of the computer-aided skin lesion classification.

A Mutual Bootstrapping Model for Automated Skin Lesion Segmentation and Classification

“…Traditionally, skin lesion segmentation methods mainly include clustering, thresholding, region growing, and active contour models [11]- [13]. Skin lesion classification methods focus mainly on extracting handcrafted features, including the color, texture, border irregularity, and asymmetry descriptors of lesions [5], [14]- [17] and using one or more of these features to train a classifier such as the K-nearest-neighbor [5], back propagation neural network [14], support vector machine [16], linear classifier [17], and logistic regression and product units [15]. Despite their prevalence, these methods rely heavily on handcrafted features and suffer a lot from less accuracy.…”

Bi-Directional Dermoscopic Feature Learning and Multi-Scale Consistent Decision Fusion for Skin Lesion Segmentation

“…We generate five sets of feature maps with five different dilation rates for the output of Block 5, i.e. (3,6,12,18,24), which are the extension of the four dilation rates of ASPP with large FOV [14]. To avoid the network growing too wide and reduce the computer consumption, we apply a 1 × 1 convolution layer before the dilated convolution to decrease the number of feature channels from 2048 to 512.…”

“…With the increase of the number of training images, the segmentation performance JA is progressively improved, since more general information (less overfitting to specific training data) can be learned from training samples with growing number and diversity. (3,6,12,18), the one used in ASPP [14] (6,12,18,24) produces better segmentation performance, i.e. improving the JA by 0.3%.…”

“…Timely detection and proper treatment are crucial for patient survival since melanoma can be cured with prompt excision [4]. Dermoscopy, widely used in the clinical examination of melanoma, is a noninvasive imaging tool that provides an accurate detail visualization of the pigmented skin lesion structure [5], [6]. But manual visual inspection based on dermoscopy is a time-consuming, hardly reproducible and subjective work.…”

Impact of morphological furrows as lubricant reservoir on creation of oleophilic and oleophobic behaviour of metallic surfaces in scuffing

“…That is why surface texturing is gaining a more and more important role in modern science and engineering. Its application covers such completely different areas as classifying skin lesions [2], identifying wood surface defects [3], analysing the lunar surface [4], evaluating concrete features [5] and even industrial robots learning to distinguish between different materials based on their texture [6]. In tribology, surface texturing is mostly used to prepare surfaces for cooperation with lubricants and/or in order to reduce friction (very popular in a design and manufacturing of cylinder liners [7,8] or bearing components [9,10]).…”