Automatic Skin Tumour Segmentation Using Prioritized Patch Based Region – A Novel Comparative Technique
A. Ashwini, S. Murugan
Abstract
Skin tumour detection plays a key factor in medical research. Nowadays, tumour detection process is of crucial importance as the number of persons affected is increasing at a larger rate. The aim of this research work is to develop a new approach for efficient image enhancement and tumour detection from other unaffected regions on computed tomographic skin images. This work is mainly related to medical application methods on computed tomography (CT) skin tumour images that have been designed and implemented efficiently. The initial method, which was based on the quality of image, enhanced the medical image performance. Normally, these images are very noise sensitive and create difficulty in handling procedures. Proper care has to be taken which involves the introduction of pre-processing algorithms like enhancement techniques and filters. According to this, Anisotropic Diffusion Filtering (ADF) followed by Recursive Mean Separate Histogram Equalization (RMSHE) algorithm was introduced to improve the contrast of tumour images. In the second method, Public Contour Metric Based Segmentation (PCMBS) Mapping and Prioritized Patch Based Region Segmentation (PPBRS) Algorithm is proposed for skin tumour segmentation. These techniques are performed in CT skin tumour image which are benign or malignant. Overall accuracy of 98.5% and 95.4% is obtained for various benign and malignant tumours, respectively, in MATLAB 2018a software.