Predicting Slice-to-Volume Transformation in Presence of Arbitrary Subject Motion
This paper aims to solve a fundamental problem in intensity-based 2D/3D registration, which concerns the limited capture range and need for very good initialization of state-of-the-art image registration methods. We propose a regression approach that learns to predict rotations and translations of arbitrary 2D image slices from 3D volumes, with respect to a learned canonical atlas co-ordinate system. To this end, we utilize Convolutional Neural Networks (CNNs) to learn the highly complex regression function that maps 2D image slices into their correct position and orientation in 3D space. Our approach is attractive in challenging imaging scenarios, where significant subject motion complicates reconstruction performance of 3D volumes from 2D slice data. We extensively evaluate the effectiveness of our approach quantitatively on simulated MRI brain data with extreme random motion. We further demonstrate qualitative results on fetal MRI where our method is integrated into a full reconstruction and motion compensation pipeline. With our CNN regression approach we obtain an average prediction error of 7 mm on simulated data, and convincing reconstruction quality of images of very young fetuses where previous methods fail. We further discuss applications to Computed Tomography (CT) and X-Ray projections. Our approach is a general solution to the 2D/3D initialization problem. It is computationally efficient, with prediction times per slice of a few milliseconds, making it suitable for real-time scenarios.
NVIDIA, Wellcome Trust/EPSRC iFIND , EPSRC EP/N024494/1.
- 1.Alzheimer’s Disease Neuroimaging Initiative (2017). http://adni.loni.usc.edu
- 3.Ghesu, F.C., Georgescu, B., Mansi, T., Neumann, D., Hornegger, J., Comaniciu, D.: An artificial agent for anatomical landmark detection in medical images. In: Ourselin, S., Joskowicz, L., Sabuncu, M.R., Unal, G., Wells, W. (eds.) MICCAI 2016, Part III. LNCS, vol. 9902, pp. 229–237. Springer, Cham (2016). doi: 10.1007/978-3-319-46726-9_27 CrossRefGoogle Scholar
- 4.Gholipour, A., et al.: Robust super-resolution volume reconstruction from slice acquisitions: application to fetal brain MRI. IEEE TMI 29(10), 1739–1758 (2010)Google Scholar
- 7.Jia, Y., et al.: Caffe: convolutional architecture for fast feature embedding. arXiv:1408.5093 (2014)
- 10.Keraudren, K., Kainz, B., Oktay, O., Kyriakopoulou, V., Rutherford, M., Hajnal, J.V., Rueckert, D.: Automated localization of fetal organs in MRI using random forests with steerable features. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015, Part III. LNCS, vol. 9351, pp. 620–627. Springer, Cham (2015). doi: 10.1007/978-3-319-24574-4_74 CrossRefGoogle Scholar
- 15.Wu, J.: ITK-based implementation of two-projection 2D/3D registration method with an application in patient setup for external beam radiotherapy. Insight J. 784 (2010). http://www.insight-journal.org/home/index