TY - GEN
T1 - Fast and Scalable Optimal Transport for Brain Tractograms
AU - Feydy, Jean
AU - Roussillon, Pierre
AU - Trouvé, Alain
AU - Gori, Pietro
N1 - Publisher Copyright:
© 2019, Springer Nature Switzerland AG.
PY - 2019/1/1
Y1 - 2019/1/1
N2 - We present a new multiscale algorithm for solving regularized Optimal Transport problems on the GPU, with a linear memory footprint. Relying on Sinkhorn divergences which are convex, smooth and positive definite loss functions, this method enables the computation of transport plans between millions of points in a matter of minutes. We show the effectiveness of this approach on brain tractograms modeled either as bundles of fibers or as track density maps. We use the resulting smooth assignments to perform label transfer for atlas-based segmentation of fiber tractograms. The parameters – blur and reach – of our method are meaningful, defining the minimum and maximum distance at which two fibers are compared with each other. They can be set according to anatomical knowledge. Furthermore, we also propose to estimate a probabilistic atlas of a population of track density maps as a Wasserstein barycenter. Our CUDA implementation is endowed with a user-friendly PyTorch interface, freely available on the PyPi repository (pip install geomloss) and at www.kernel-operations.io/geomloss.
AB - We present a new multiscale algorithm for solving regularized Optimal Transport problems on the GPU, with a linear memory footprint. Relying on Sinkhorn divergences which are convex, smooth and positive definite loss functions, this method enables the computation of transport plans between millions of points in a matter of minutes. We show the effectiveness of this approach on brain tractograms modeled either as bundles of fibers or as track density maps. We use the resulting smooth assignments to perform label transfer for atlas-based segmentation of fiber tractograms. The parameters – blur and reach – of our method are meaningful, defining the minimum and maximum distance at which two fibers are compared with each other. They can be set according to anatomical knowledge. Furthermore, we also propose to estimate a probabilistic atlas of a population of track density maps as a Wasserstein barycenter. Our CUDA implementation is endowed with a user-friendly PyTorch interface, freely available on the PyPi repository (pip install geomloss) and at www.kernel-operations.io/geomloss.
UR - https://www.scopus.com/pages/publications/85075649619
U2 - 10.1007/978-3-030-32248-9_71
DO - 10.1007/978-3-030-32248-9_71
M3 - Conference contribution
AN - SCOPUS:85075649619
SN - 9783030322472
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 636
EP - 644
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings
A2 - Shen, Dinggang
A2 - Yap, Pew-Thian
A2 - Liu, Tianming
A2 - Peters, Terry M.
A2 - Khan, Ali
A2 - Staib, Lawrence H.
A2 - Essert, Caroline
A2 - Zhou, Sean
PB - Springer Science and Business Media Deutschland GmbH
T2 - 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
Y2 - 13 October 2019 through 17 October 2019
ER -