Liver image registration with constraint applied for accurate tumor localization(International Forum on Medical Imaging in Asia 2009 (IFMIA 2009))
スポンサーリンク
概要
- 論文の詳細を見る
The main purpose of liver image registration of 3D ultrasound and XCT images is in accurate tumor localization in an US image by fusing the corresponding XCT image. The existing approach for US-XCT liver registration aims to align visible anatomical structures such as vessels and diaphragm between two images. In this paper, we try to improve the performance of non-rigid registration for accurate tumor localization. By examining the motions of tumors, vessels, and diaphragm and adopting thin-plate spline modeling of vessels, we apply a constraint to the non-rigid transformation. Experimental results show the improvement of registration accuracy in tumor area.
- 社団法人電子情報通信学会の論文
- 2009-01-12
著者
-
Ra Jong
Kaist Daejeon Kor
-
Nam Woo
School Of Electrical Engineering And Computer Science Kaist
-
LEE Duhgoon
School of Electrical Engineering and Computer Science, KAIST
-
RA Jong
School of Electrical Engineering and Computer Science, KAIST
-
Lee Duhgoon
School Of Electrical Engineering And Computer Science Kaist