Automatic identification of corresponding CT images having the same lymph node in longitudinal studies

摘要：

The study of lymph node features over time is of greatclinical significance. Identification of the corresponding Ctimage with the same lymph node is done manually in the currentclinical practice. In this paper, we propose a search scheme toautomate the process. Region of interest (ROIs) are located bymapping the center point of lymph node based on thetransformation found in the rigid registration. Similarity valuesbetween ROI of the template image and ROIs of repositoryimages are compared, the highest of which decides the bestmatch. Our method generated a success rate of 77.78% indetermining the correct match. We also investigated theefficiency of using different similarity measures, mutualinformation and cross correlation, in the scheme. The effect ofvarying sizes of ROIs on the identification results was examinedas well.

作者: Peicong Yu Chueh Loo Poh Kenneth Sheah

作者单位: Division of Bioengineering, SCBE Nanyang Technological University Singapore Department of Radiology, Changi General Hospital Radlink Diagnostic Imaging,Paragon Medical,Singapor

会议类型: 国际会议

会议名称: The 4th International Conference on Bioinformatics and Biomedical Engineering(第四届IEEE生物信息与生物医学工程国际会议 iCBBE 2010)

会议地点: 成都

会议语种:英文

页码: 1-4

在线出版日期: 2010-06-18（万方平台首次上网日期，不代表论文的发表时间）