QFib: Fast and Efficient Brain Tractogram Compression
- PDF / 3,488,179 Bytes
- 14 Pages / 595.224 x 790.955 pts Page_size
- 46 Downloads / 200 Views
ORIGINAL ARTICLE
QFib: Fast and Efficient Brain Tractogram Compression C. Mercier1,2 · S. Rousseau1
· P. Gori1 · I. Bloch1 · T. Boubekeur1
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Diffusion MRI fiber tracking datasets can contain millions of 3D streamlines, and their representation can weight tens of gigabytes of memory. These sets of streamlines are called tractograms and are often used for clinical operations or research. Their size makes them difficult to store, visualize, process or exchange over the network. We propose a new compression algorithm well-suited for tractograms, by taking advantage of the way streamlines are obtained with usual tracking algorithms. Our approach is based on unit vector quantization methods combined with a spatial transformation which results in low compression and decompression times, as well as a high compression ratio. For instance, a 11.5GB tractogram can be compressed to a 1.02GB file and decompressed in 11.3 seconds. Moreover, our method allows for the compression and decompression of individual streamlines, reducing the need for a costly out-of-core algorithm with heavy datasets. Last, we open a way toward on-the-fly compression and decompression for handling larger datasets without needing a load of RAM (i.e. in-core handling), faster network exchanges and faster loading times for visualization or processing. Keywords Compression · Diffusion MRI · Tractography · On-the-fly algorithms · Unit vectors
Introduction Diffusion magnetic resonance imaging (dMRI) tractography is currently the only technique able to noninvasively obtain the white matter architecture of the human brain. Tractography helps clinicians, neurosurgeons and researchers to understand the connections of the brain and is widely used for pre-operative planning, during clinical operations, and for research purposes. Fiber tracking datasets – called tractograms – are composed of 3D streamlines represented as 3D polylines with hundreds to thousands of ordered 3D points. Modern tractography algorithms can obtain up to several millions of these streamlines (Tournier et al. 2011), resulting in tens of gigabytes (GB) of data (Rheault et al. 2017). For instance, a file conC. Mercier, S. Rousseau contributed equally to this work C. Mercier
[email protected] S. Rousseau
[email protected] 1
LTCI, T´el´ecom Paris, Institut Polytechnique de Paris, Palaiseau, France
2
´ LTCI, T´el´ecom Paris and LIX, Ecole Polytechnique, Institut Polytechnique de Paris, Palaiseau, France
taining 1 million streamlines and obtained with a constant stepsize of 0.1mm can weight up to 8.7GB. This massive amount of data complicates visualization, processing, sharing or storage. In this article, we introduce a new compression algorithm for fiber tracking datasets, which is both fast and efficient. Existing methods, that propose a solution to this data size problem, can be divided into two different categories. They either compress the whole tractogram or the representa
Data Loading...
