USING CLASTERING-BASED SEGMENTATION ALGORITHMS IN REGIONS OF INTEREST DETECTION ON THE MEDICAL IMAGES

Abstract

This article gives an overview of the existing clustering based algorithms of segmentation and their application in the medical image processing area. At the beginning of the article the importance of the research in this field was presented and approved by the analysis of the founded sources. The general clustering approach was described where were faced up the advantages and the disadvantages of this method. After that the general overview about the segmentation process was given and connected with the main direction of the clustering method. Several approaches were shown, such as hierarchical, fuzzy and etc. The steps of these algorithms were presented, the input and output data also was browsed. According to the received information several conclusion were prepared as a result of the analysis of the algorithms results. The general advices were given about the using the particular approach in a context of the real segmentation of the medical image. The full analysis was performed in application of these algorithms and as a result the benefits and looses of the approaches also were given in these article. A comparison of these methods was given in the main conclusion of the article.