Comparison between K mean and fuzzy C-mean methods for segmentation of near infrared fluorescent image for diagnosing prostate cancer

Abstract

In each year there are thousands of people die due to prostate cancer. Near-infrared (NIRF) optical imaging is a new technique that uses the high absorption of hemoglobin in prostate's cancer cells for early detection. We use Image segmentation method to segment and extract the cancer region in the prostate's infrared images. In this paper, two image segmentation methods: K-means algorithm and fuzzy c-means (FCM) algorithms are discussed and compared. The extracted cancer clusters by two algorithms are compared using Student t-test and we found that the K-mean is more accurate approach than FCM in extracting the exact shape of tumors.