| 講演抄録/キーワード |
| 講演名 |
2015-09-11 10:55
Levenberg-Marquardt法を用いた3次元頭部CT・MR画像の自動位置合わせ 木崎美桜・山村雄太郎・○金 亨燮・タン ジュークイ・石川聖二(九工大)・山本晃義(戸畑共立病院) LOIS2015-24 IE2015-56 EMM2015-39 |
| 抄録 |
(和) |
本稿では, サイバーナイフ治療に用いられる3次元頭部CT画像とMR画像のフュージョン画像を生成するための画像位置合わせを提案する. 提案法では,大まかな画像位置合わせ法であるグローバルマッチングの改良と処理時間の低減を実現するため,coarse-to-fine法に基づく初期画像位置合わせを行う. さらに,両画像上の複数の関心領域(VOI)を設置し, Levenberg-Marquardtの最適法を用いた各VOIの最適な変形量を求める. |
| (英) |
Medical imaging diagnosis and surgical planning often benefit from an image fusion technique. However, commonly used mouse-based registration yields variations in the accuracy of results due to manual operations. In addition, the increase of physical burdens to the operators and the requirement for a substantial investment of time are pending issues. Therefore, in recent years many researches have attempted to automatic image registration for multi-modal images. In this paper, a CAD system is developed to assist radiotherapy planning for CyberKnife treatment. We propose a new automatic image registration technique for image fusion of head CT and MR images. In our method, the process of the global matching is refined based on coarse-to-fine as an initial matching and the maximization of normalized mutual information on VOIs is performed based on Levenberg-Marquardt algorithm as the final matching. We evaluated our proposed method in 5 clinical cases and discussed the accuracy of the registration results with computational time. |
| キーワード |
(和) |
画像位置合わせ / Levenberg-Marquardt Algorithm / CT / MRI / サイバーナイフ / / / |
| (英) |
Image Registration / Levenberg-Marquardt Algorithm / CT / MRI / CyberKnife / / / |
| 文献情報 |
信学技報, vol. 115, no. 220, IE2015-56, pp. 51-54, 2015年9月. |
| 資料番号 |
IE2015-56 |
| 発行日 |
2015-09-03 (LOIS, IE, EMM) |
| ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
| PDFダウンロード |
LOIS2015-24 IE2015-56 EMM2015-39 |