Landmark-Based Evaluation of a Block-Matching Registration Framework on the RESECT Pre- and Intra-operative Brain Image Data Set
In this paper, we describe the application of an established block-matching based registration method to the CuRIOUS 2019 MICCAI registration challenge. Directional and symmetric approaches with different parameters are evaluated to select the most suitable setting of this fully automatic and general registration method. The results can be used as a baseline, for example when evaluating methods specialised in ultrasound (US) to MRI registration or registration of different interventional US (iUS) data. This work is a continuation of our contribution to the CuRIOUS 2018 challenge. We provide a more extensive analysis of main parameters as well as add pre- to post-resection iUS registration to the previous MRI-iUS registration. The proposed approach achieves an average target registration error of 2.68 mm and 1.92 mm for the MR-iUS and the iUS-iUS task respectively.
KeywordsBlock-matching Symmetric registration Resection Brain shift Fully automatic MRI iUS
This work is supported by the UCL EPSRC Centre for Doctoral Training in Medical Imaging [EP/L016478/1], the Wellcome/EPSRC Centre for Interventional and Surgical Sciences [NS/A000050/1], the Wellcome/EPSRC Centre for Medical Engineering [WT 203148/Z/16/Z] and EPSRC [NS/A000027/1]. We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Quadro P6000 used for this research. This research was supported by the NIHR BRC based at GSTT and KCL.
- 1.Drobny, D., Vercauteren, T., Ourselin, S., Modat, M.: Registration of MRI and iUS data to compensate brain shift using a symmetric block-matching based approach. In: Stoyanov, D., et al. (eds.) POCUS/BIVPCS/CuRIOUS/CPM -2018. LNCS, vol. 11042, pp. 172–178. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01045-4_21CrossRefGoogle Scholar
- 5.Niftyreg github page. https://github.com/KCL-BMEIS/niftyreg/wiki. Accessed 29 July 2019
- 7.Xiao, Y., et al.: Evaluation of MRI to ultrasound registration methods for brain shiftcorrection: the CuRIOUS2018 challenge. IEEE Trans. Med. Imaging (2019). https://doi.org/10.1109/TMI.2019.2935060