通信市場網--中國通信市場(CTM)權威信息發布門戶網站

東南大學影像科學與技術實驗室)

摘要：互信息法因其具有較高的準確性和較好的魯棒性在醫學圖像配準中得到廣泛的應用。對經典互信息法進行改進，使其在計算中包含圖像的空間信息是近年的一個熱點。本文在考慮圖像的空間信息以及不同形式的熵在超聲圖像配準中的效果后，結合了條件互信息法和Tsallis熵的特性，提出了一種改進型互信息方法。實驗表明該方法在超聲-磁共振多模態圖像配準中較經典互信息法有更好的效果。

關鍵字：互信息；多模態圖像配準；超聲圖像；相似度測量；Tsallis熵

Ultrasound-MR image registration based on improved conditional mutual information

Li Wen, Shu Huazhong

(Laboratory of Image Science and Technology, Southeast University, Nanjing, 210096)

Abstract: General mutual information (GMI) is a popular similarity measure for medical image registration due to its high accuracy and robustness. Extending GMI with spatial distribution is an active field of research. We propose a novel approach: improved mutual information, which combines the quality of conditional mutual information and Tsallis entropy. The experimental result demonstrates that proposed method significantly outperforms GMI for ultrasound-MR image registration.

Key words: Mutual information; multimodal Image registration; Ultrasound image; Similarity measures; Tsallis entropy

參考文獻：

[1] Roche, A., Pennec, X., Malandain, et al.: Rigid registration of 3D ultrasound with MR images: a new approach combining intensity and gradient information[J], IEEE Transactions on Medical Imaging , 2001, Vol.20, pp.1038–1049.

[2] Leroy, A., Mozer, P., Payan, et al.: Rigid registration of freehand 3D ultrasound and CT-Scan kidney images [J]. MICCAI Proceedings, 2004, pp. 837-844.

[3]Wein, W., Ro per, B., Navab: Integrating diagnostic B-mode ultrasonography into CT-based radiation treatment planning [J].IEEE Transactions on Medical Imaging, 2007, pp.866–879.

[4] Ji S, Wu Z, Hartov A, et al: Mutual-information-based image to patient re-registration using intraoperative ultrasound in image-guided neurosurgery [J], Medical physics, 2008, pp.4612-4624

[5] A. Collignon, F. Maes, D. Delaere, D. Vandermeulen, et al.: Automated multi-modality image registration based on information theory [J], Computational Imaging and Vision, 1995, Vo. 3, pp.263–274.

[6] P. Viola and W. M. Wells: Alignment by maximization of mutual information [J], international journal of computer vision, 1997, Vo. 24, pp.137-154.

[7] F. Maes, D. Vandermeulen, and P. Suetens: Medical image registration using mutual information [J], Proceedings of the IEEE, 2003, vol. 91, no. 10, pp.1699–1722.

[8] Pluim, J.P.W., Maintz, J.B.A., Viergecer, M.A.: Image registration by maximization of combined mutual information and gradient information [J]. IEEE Trans. Med. Imaging, 2000, Vol. 19 pp.809-814

[9] Rueckert, D., Clarkson, M.J.,Hill, et al.: Non-rigid registration using higher-order mutual information[A]. Medical Imaging 2000: Image Processing Proc. SPIE[C] 3979(2000), pp.438-447

[10] Russakoff, D., C. Tomasi, et al.: Image similarity using mutual information of regions [J]. Computer Vision-ECCV, 2004, pp. 596-607.

[11] Loeckx, D., P. Slagmolen, et al.: Nonrigid image registration using conditional mutual information[J], Medical Imaging, IEEE Transactions on, 2010, Vo. 29(1), pp.19-29.

[12] Wachowiak, M.P., Smolikova, R., Tourassi, G.D., et al: Similarity metrics based on non-additive entropies for 2D-3D multimodal biomedical image registration [A], Medical Imaging 2003: Image Processing[C], pp.1090-1100.

[13]楊金寶，劉常春，胡順波：廣義信息熵測度在醫學圖像配準中的應用[J]，計算機工程與應用，2008，Vo.44，pp.34-36.

《通信市場》中國·北京·復興路49號通信市場(100036) 點擊查看具體位置
電話:86-10-6820 7724, 6820 7726

◆建議使用 Microsoft IE4.0 以上版本 800*600瀏覽如果您有什么建議和意見請與管理員聯系