Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Oct:9902:1-9.
doi: 10.1007/978-3-319-46726-9_1. Epub 2016 Oct 2.

Learning-Based Multimodal Image Registration for Prostate Cancer Radiation Therapy

Xiaohuan Cao  1   2 , Yaozong Gao  2   3 , Jianhua Yang  1 , Guorong Wu  2 , Dinggang Shen  2
Affiliations

Learning-Based Multimodal Image Registration for Prostate Cancer Radiation Therapy

Xiaohuan Cao et al. Med Image Comput Comput Assist Interv. 2016 Oct.

Abstract

Computed tomography (CT) is widely used for dose planning in the radiotherapy of prostate cancer. However, CT has low tissue contrast, thus making manual contouring difficult. In contrast, magnetic resonance (MR) image provides high tissue contrast and is thus ideal for manual contouring. If MR image can be registered to CT image of the same patient, the contouring accuracy of CT could be substantially improved, which could eventually lead to high treatment efficacy. In this paper, we propose a learning-based approach for multimodal image registration. First, to fill the appearance gap between modalities, a structured random forest with auto-context model is learnt to synthesize MRI from CT and vice versa. Then, MRI-to-CT registration is steered in a dual manner of registering images with same appearances, i.e., (1) registering the synthesized CT with CT, and (2) also registering MRI with the synthesized MRI. Next, a dual-core deformation fusion framework is developed to iteratively and effectively combine these two registration results. Experiments on pelvic CT and MR images have shown the improved registration performance by our proposed method, compared with the existing non-learning based registration methods.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
Pelvic CT and MRI. From left to right: CT, labeled CT, labeled MRI and MRI.
Fig. 2
Fig. 2
The framework of proposed learning-based MRI-to-CT image registration.
Fig. 3
Fig. 3
Classic random forest (top) and SRF (bottom).
Fig. 4
Fig. 4
Iterative refinement of synthesized MRI by the auto-context model.
Fig. 5
Fig. 5
Comparison of MRI-to-CT non-rigid registration results. (a) The mean DSC of prostate, bladder and rectum by different number of ACM layers in image synthesis. (b) Registration results of D. Demons (left) and SyN (right) with respect to different DDF iterations in Algorithm 1. Note that, 3-iter DDF is applied in (a), and 2-layer ACM is used in (b).
Fig. 6
Fig. 6
Demonstration of synthesized images and SyN registration results. (e) Result 1: direct registration of MRI to CT using MI; (f) Result 2: registration with our proposed method. Yellow contours: original CT labels of 3 organs. Red contours: warped MRI labels of 3 organs. (Color figure online)
See this image and copyright information in PMC

References

    1. Sotiras A, Davatzikos C, Paragios N. Deformable medical image registration: a survey. IEEE Trans Med Imaging. 2013;32(7):1153–1190. - PMC - PubMed
    1. Pluim JP, Maintz JA, Viergever MA. Mutual-information-based registration of medical images: a survey. IEEE Trans Med Imaging. 2003;22(8):986–1004. - PubMed
    1. Huynh T, et al. Estimating CT image from MRI data using structured random forest and auto-context model. IEEE Trans Med Imaging. 2015;35(1):174–183. - PMC - PubMed
    1. Tu Z, Bai X. Auto-context and its application to high-level vision tasks and 3d brain image segmentation. IEEE Trans Pattern Anal Mach Intell. 2010;32(10):1744–1757. - PubMed
    1. Viola P, Jones MJ. Robust real-time face detection. Int J Comput Vision. 2004;57(2):137–154.

MeSH terms