Detection of Regions of Interest in Mammograms by Using Local Binary Pattern and Dynamic K-Means Algorithm
Abdelali Elmoufidi, Khalid El Fahssi, Said Jai Andaloussi, Abderrahim Sekkaki · 2015
This paper presents a method for the detection of regions of interest (ROI) in mammograms by using a dynamic K-means clustering algorithm. This method is used to partition automatically an image into a set of regions (clusters or classes). Our method consists of three phases: firstly, preprocessing images by using thresholding and filtering methods; secondly, generating range of number of clusters by using Local Binary Pattern (LBP) and Applying k-means with its features to automatically generating the optimal number of clusters ( thereafter k is The number of clusters generating); thirdly, partition the mammograms images into k clusters by applying the dynamic k-means clustering algorithm, we end by detecting the regions of interest (ROI) in mammograms images. To demonstrate the results of our proposed method we used the Mini-MIAS (Mammogram Image Analysis Society, UK) database, consisting of 322 mammograms. Our method’s performance is evaluated using Free response ROC (FROC) curves. The archived results are 2.84 false positives per image (FPpI) and sensitivity of 85%.