Image registration and tracking

A core problem in biomedical, terrestrial or physical applications is to align several images or volumes of the same object that contain random or systematic differences.

At MIC/MeVis, we develop efficient methods for high-accuracy estimation of correspondences and deformation maps between multiple datasets. Challenges include large datasets (usually 3D or 4D volumes) and the highly ambiguous/ill-posed nature of the registration problem, which requires modelling of application-specific knowledge.

Algorithms developed at MIC have won several awards at large conferences including CVPR and BVM, and rank among the highest at several benchmark competitions, such as the Empire10 and CLUST challenge. They are currently licensed and in production by a number of companies.

Keywords

  • Image registration
  • Object tracking
  • Motion estimation and optical flow
  • Variational methods
  • Energy minimization

Applications 

  • Intra- and interpatient alignment of CT/MRI/PET/SPECT data for quantitative analysis, surgical intervention and radiation planning
  • Cursor synchronization
  • Tracking of structures in histological data
  • Change detection for follow-up diagnostics in radiology
  • Motion correction and compensation of breathing motion
  • Registration of lung CT images for ventilation modelling

People

Software

Selected Publications

  1. Rühaak, J., König, L., Tramnitzke, F., Köstler, H., and Modersitzki, J.: A Matrix-Free Approach to Efficient Affine-Linear Image Registration on CPU and GPU. Journal of Real-Time Image Processing, 2016 
  2. Lotz, J., Olesch, J., Müller, B., Polzin, T., Galuschka, P., Lotz, J. M., Heldmann, S., Laue, H., Warth, A., Lahrmann, B., Grabe, N., Sedlaczek, O., Breuhahn, K., and Modersitzki, J.: Patch-Based Nonlinear Image Registration for Gigapixel Whole Slide Images. IEEE Transactions on Biomedical Engineering, vol. 63, no. 9, pp. 1812 - 1819, 2015
  3. Luca, V. D., Benz, T., Kondo, S., König, L., Lübke, D., Rothlübbers, S., Somphone, O., Allaire, S., Bell, M. A. L., F, D. Y., Cifor, A., Grozea, C., Günther, M., Jenne, J., Kipshagen, T., Kowarschik, M., Navab, N., Rühaak, J., Schwaab, J., and Tanner, C.: The 2014 liver ultrasound tracking benchmark. Physics in Medicine and Biology, vol. 60, no. 14, , 2015
  4. Weiss, N., Lotz, J., and Modersitzki, J.: Multimodal Image Registration in Digital Pathology Using Cell Nuclei Densities. Bildverarbeitung für die Medizin 2015, Algorithmen – Systeme – Anwendungen, Proceedings des Workshops vom 15. bis 17. März 2015 in Lübeck, Handels, H., Deserno, T. M., Meinzer, H., and Tolxdorff, T. (Ed.), Springer Vieweg, Berlin, Heidelberg, 2015. Best Scientific Contribution at BVM 2015.
  5. Olesch, J.: Bildregistrierung für die navigierte Chirurgie. Springer Vieweg, Aktuelle Forschung Medizintechnik, 2014 
  6. Ruthotto, L., Mohammadi, S., Heck, C., Modersitzki, J., and Weiskopf, N.: Hyperelastic Susceptibility Artifact Correction of DTI in SPM: Bildverarbeitung für die Medizin 2013, Meinzer, H., Deserno, T. M., Handels, H., and Tolxdorff, T. (Ed.), Springer Berlin Heidelberg, 2013. 1st Price for Best Presentation at BVM 2013
  7. Polzin, T., Rühaak, J., Werner, R., Strehlow, J., Heldmann, S., Handels, H., and Modersitzki, J.: Combining Automatic Landmark Detection and Variational Methods for Lung CT Registration. Proc. Fifth International MICCAI Workshop on Pulmonary Image Analysis (PIA 2013), Nagoya, Japan, September 2013
  8. Modersitzki, J.: FAIR: Flexible Algorithms for Image Registration. SIAM, Philadelphia, 2009
  9. Modersitzki, J.: Numerical Methods for Image Registration. Oxford University Press, New York, 2004