Adaptive Edge Detection Technique Towards Features Extraction from Mammogram Images
Cancer is one of the most dreaded diseases of modern world. Breast cancer is the second most type of cancer & the fifth most common cause of cancer related death so it’s a significant public health problem in the world especially for elderly females. Computer technology specifically computer aided diagnosis (CAD), relatively young interdisciplinary technology, has had a tremendous impact on medical diagnosis of cancer detection due to its accuracy and cost effectiveness. The accuracy of CAD to detect abnormalities on medical image analysis requires a robust segmentation algorithm. To achieve accurate segmentation, an efficient edge-detection algorithm is essential. The mammogram is a comparatively efficient and low cost diagnostic imaging technique for breast cancer detection. In this paper a robust mammogram enhancement and edge detection algorithm is proposed, using tree-based adaptive thresholding technique. The proposed technique has been compared with different classical edge-detection techniques yielding acceptable out come. The proposed edge-detection algorithm showing 0.07 p-values and 2.411 t-stat in one sample two tail t-test (α = 0.025). The edge is single pixeled and connected which is very significant for medical edge-detection.
Mammogram, CAD, Edge Detection, Full and Complete Binary Tree, Adaptive Threshold, Histogram, Average Bin Distance (ABD), Maximum Difference Threshold (MDT), Prominent Bins, t-Test
Full Text:Subscribers Only
- There are currently no refbacks.