BRAIN TISSUE SEGMENTATION IN MRI IMAGES USING RANDOM FOREST CLASSIFIER AND GOSSIP BASED NEIGHBORHOOD

摘要：

Considering the importance of careful segmentation of medical images for deleanition of unhelthy tissues or locating and tracing of the tumor growth, it has been the subject of interest in many medical researches. In this paper we propose a gossip-based region growing algorithm to extract more accurate spatial features.The features will then be imported to a random forest.The random forest classifier is an ensemble classifier derived from the decision tree idea but with accuracy rates comparable to most of currently used classifiers.Although being a very strong classifier, random forest has rarely been studied in the field of medical segmentation.The brain scans segmentation with random forest showed promising results using the gossip-based region growing algorithm.

关键词： Medical image segmentation Random forest classifier Gossiping algorithm

作者: SAEIDEH ESLAMI MORTEZA ZAHEDI REZA AZMI ROBAB ANBIAEE

作者单位: Shahrood University of Technology,Shahrood,Iran Shahrood University of Technology Alzahra University,Tehran,Iran Shahidbeheshti Medical Science University

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer Technology and Development(2011第三届计算机技术与发展国际会议 ICCTD2011)

会议地点: 成都

会议语种:英文

页码: 617-621

在线出版日期: 2011-11-25（万方平台首次上网日期，不代表论文的发表时间）