Computer Aided Segmentation Approach for Melanoma Skin Cancer Detection
4 Pages Posted: 4 Jun 2019
Date Written: April 18, 2018
Skin cancer is one of the varieties of the most common type of cancers and nowadays, it alarms the human communities in rapid manner. In recent years, it has been realized of rapid increment in melanoma skin cancer patients. Melanoma, is a form of skin cancer, and must be diagnosed in early stage for eﬀective treatment. The early detection of melanoma highly depends on the accuracy of skin lesion segmentation. Still, the segmentation of the melanoma skin cancer lesion in traditional approach is a challenging task due to the number of false positives is large and time consuming in prediction. The innovative development of automated computer vision system is a vital tool to segment the skin lesion from the photograph (the area where cancer affected) of cancer patient, and which the method is currently available. Meanwhile this research work was carried out through image processing techniques, and some of these techniques are widely used in similar applications, as is the case of the canny edge detection for finding the lesion boundary. Other techniques are Watershed segmentation for segmenting the lesion from skin, multilevel thresholding for merge the lesion and Active contour for increasing the accuracy. Although the people, who are in medical domain had introduced new methodologies to improve the accuracy by addressing the challenges and mainly focusing on the accuracy but still the accuracy is a challenging grey area. Our approach was to introduce an optimal algorithm, to improve the accuracy and the result sounds well, which could achieve 97.54% sensitivity, 97.69% specificity, and 97.56% accuracy.
Keywords: Lesion, Segmentation, Canny edge, Thresholding, Watershed, optimal algorithm, cancer detection
Suggested Citation: Suggested Citation