Skin lesion classification using decision trees and random forest algorithms
- PDF / 1,239,379 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 40 Downloads / 276 Views
ORIGINAL RESEARCH
Skin lesion classification using decision trees and random forest algorithms C. R. Dhivyaa1 · K. Sangeetha2 · M. Balamurugan3 · Sibi Amaran4 · T. Vetriselvi5 · P. Johnpaul6 Received: 14 August 2020 / Accepted: 3 November 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Any superficial skin growth that does not resemble the surrounding area is referred to as skin lesion. It can occur in the form of mole, bump, cyst, rash or other changes that can be classified either as primary or secondary lesion. While primary skin lesions correspond to those changes in color or texture, secondary lesions occur as a primary lesion progression. Skin lesion image segmentation and classification at the early stages can help the patients recover through proper medication and treatment. Many algorithms for segmentation and classification are available in the literature but they all fail to extract lesion boundaries perfectly and classify them with more accuracy. To improve the reliability of the skin image segmentation and classification, we propose to use decision trees and random forest algorithms in this works and compare them with different data sets. The proposed method can generate high-resolution feature maps that can help to preserve the spatial details of the image. While tested against the ISIC 2017 and HAM10000 dataset, we found that the proposed method is more accurate as compared to the existing algorithms in this domain and is also very robust to artifacts or hair fibers present in the skin images. Keywords Feature maps · Noise · Skin lesion · Artifacts · Filters · Melanoma · Asymmetry · Dermoscopy · Specificity · Lesion configuration · Entropy function · Texture · Decision trees
1 Introduction Skin lesions happens due to skin infection by viruses, bacteria, fungus or parasites. The primary skin lesions include blisters, macule, nodule, papule, pustule, rash and wheals while the secondary skin lesions include crusts, scale, ulcers, scars and skin atrophy. While most of them are harmless, some can be related to a skin cancer warning as well. So, it is imperative to identify them at the earliest possible and assist the medical practitioner in helping the patients with the right
medication and cure the disease. With this objective, many skin lesion image segmentation and classification algorithms are discussed in the literature. A typical skin lesion image with the affected region is shown in Fig. 1 below: Medical image processing assist the dermatologists in the skin lesion identification and classification. Image segmentation remains the first and key step in skin lesion diagnosis (Jain et al. 2015). It helps to improve the region of interest accuracy in dermoscopic images. As the human eyes are not so good in discriminating the lesion borders in terms of 1
Nandha College of Technology, Erode, India
2
Department of CSE, Kongu Engineering College, Perundurai, India
3
Department of Computer Science and Engineering, School of Engineering and Technology, CHRIST Dee
Data Loading...
