Pixel Mask Analyzer (PMA): Noise removal in gland detection for colon cancer image

Abstract

It is well known that for any type of cancer, mortality rate can be reduced by early detection of the cancer. Detection of cancer markers from the time of taking blood from patients until microscopic examination of the biopsy tissues is not only laborious but time consuming. It is known that with the help of a computerized system, the diagnosis time can be shortened. In this paper, we proposed the method of gland edge detection and noise removal using Pixel Mask Analyzer (PMA) as image enhancement. PMA is based on kFill filter that uses 5x5 mask slide through entire 2D biopsies tissue images. Pixels within mask are analyzed on 8 sample images to remove noise. We compare the median filter and PMA and found that PMA gives better result on noise removal. After noise removal, Sobel and Canny edge detectors are applied on the images and we found that Canny provides better output for edge detection. The output of this detection is important for further analysis of colon cancer cells image detection. © 2011 Springer-Verlag.