Multimodal Registration of White Matter Brain Data via Optimal Mass Transport

Please use this identifier to cite or link to this publication:
The elastic registration of medical scans from different acquisition sequences is becoming an important topic for many research labs that would like to continue the post-processing of medical scans acquired via the new generation of high-field-strength scanners. In this note, we present a parameter-free registration algorithm that is well suited for this scenario as it requires no tuning to specific acquisition sequences. The algorithm encompasses a new numerical scheme for computing elastic registration maps based on the minimizing flow approach to optimal mass transport. The approach utilizes all of the gray-scale data in both images, and the optimal mapping from image A to image B is the inverse of the optimal mapping from B to A. Further, no landmarks need to be specified, and the minimizer of the distance functional involved is unique. We apply the algorithm to register the white matter folds of two different scans and use the results to parcellate the cortex of the target image. To the best of our knowledge, this is the first time that the optimal mass transport function has been applied to register large 3D multimodal data sets.

minus Review by Anonymous on 06-30-2008 for revision #1
starstarstarstarstar expertise: 4 sensitivity: 5

   In this paper the authors propose the use of the Optimal Mass Transport theory for identifying the cortical structures by mapping a brain atlas to the MRI scan of a patient.


   Most of the solution algorithm derivation and implementation described in this paper was presented in previous papers [1]. The authors contribute by proposing the use of a different direction for minimization and by enforcing the mass preserving property after each iteration step.


   The method requires as a pre-requisite a segmentation of the patient MRI scan into the major tissue classes - it is not specified in the paper how this problem is solved. It also requires that the intensities of the two input images (atlas and MRI) be normalized and rescaled to make sure that both have the same mass (intensity). This will certainly influence the results especially if the intensity distribution differs considerably between the atlas and the MRI scan (scanning sequences or sensors are very different).


   The authors present the difference in white matter intensity between the patient MRI and the re-sampled atlas image as a proof of accuracy. Nevertheless, this only proves that the algorithm can morph one image into the other, not that the folds are accurately aligned (as shown in the presented results, some atlas labels are propagated to the wrong regions). The matching of anatomical landmarks would be more appropriate for quantifying the accuracy of the method.




1.         Haker, S., Zhu, L., Tannenbaum, A., Angenent, S., 2004, Optimal mass transport for registration and warping. International Journal of Computer Vision 60 (3), 225–240.


Add a new review
Quick Comments

Download All

Statistics more
Global rating: starstarstarstarstar
Review rating: starstarstarstarstar [review]
Paper Quality: plus minus

Information more
Categories: Atlas-based segmentation, Data Representation, Derivatives and Integrals, Image, Linear Algebra, Mathematics, Mesh, Multi-modality registration, Optimization, PointSet, Registration, Registration metrics, Registration optimizers, Segmentation
Keywords: Optimal Mass Transport, Registration, Monge Kantorovich, Variational Methods, Fluid Mechanics
Export citation:


Linked Publications more
HL7 CCD - Clinical Interoperability to Improve Quality and  the Point-of-Care of EHR HL7 CCD - Clinical Interoperability to Improve Quality and the Point-of-Care of EHR
by Patel G.
The Surgical Assistant Workstation The Surgical Assistant Workstation
by Balazs V., Simon D., Anton D., Peter K., Rajesh K., Christopher H., Russell T.

View license
Loading license...

Send a message to the author
Powered by Midas