Optimized fast GPU implementation of robust artificial-neural-networks for k-space interpolation (RAKI) reconstruction


Autoři: Chi Zhang aff001;  Seyed Amir Hossein Hosseini aff001;  Sebastian Weingärtner aff001;  Kâmil Uǧurbil aff002;  Steen Moeller aff002;  Mehmet Akçakaya aff001
Působiště autorů: Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN, United States of America aff001;  Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, United States of America aff002;  Department of Imaging Physics, Delft University of Technology, Delft, Netherlands aff003
Vyšlo v časopise: PLoS ONE 14(10)
Kategorie: Research Article
doi: 10.1371/journal.pone.0223315

Souhrn

Background

Robust Artificial-neural-networks for k-space Interpolation (RAKI) is a recently proposed deep-learning-based reconstruction algorithm for parallel imaging. Its main premise is to perform k-space interpolation using convolutional neural networks (CNNs) trained on subject-specific autocalibration signal (ACS) data. Since training is performed individually for each subject, the reconstruction time is longer than approaches that pre-train on databases. In this study, we sought to reduce the computational time of RAKI.

Methods

RAKI was implemented using CPU multi-processing and process pooling to maximize the utility of GPU resources. We also proposed an alternative CNN architecture that interpolates all output channels jointly for specific skipped k-space lines. This new architecture was compared to the original CNN architecture in RAKI, as well as to GRAPPA in phantom, brain and knee MRI datasets, both qualitatively and quantitatively.

Results

The optimized GPU implementations were approximately 2-to-5-fold faster than a simple GPU implementation. The new CNN architecture further improved the computational time by 4-to-5-fold compared to the optimized GPU implementation using the original RAKI CNN architecture. It also provided significant improvement over GRAPPA both visually and quantitatively, although it performed slightly worse than the original RAKI CNN architecture.

Conclusions

The proposed implementations of RAKI bring the computational time towards clinically acceptable ranges. The new CNN architecture yields faster training, albeit at a slight performance loss, which may be acceptable for faster visualization in some settings.

Klíčová slova:

Acceleration – Data acquisition – Interpolation – Knees – Magnetic resonance imaging – Machine learning – Machine learning algorithms – Memory


Zdroje

1. Sodickson DK, Manning WJ. Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays. Magn Reson Med. 1997;38(4):591–603. Epub 1997/11/05. doi: 10.1002/mrm.1910380414 9324327.

2. Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE: sensitivity encoding for fast MRI. Magn Reson Med. 1999;42(5):952–62. Epub 1999/11/05. doi: 10.1002/(SICI)1522-2594(199911)42:5<952::AID-MRM16>3.0.CO;2-S [pii]. 10542355.

3. Griswold MA, Jakob PM, Heidemann RM, Nittka M, Jellus V, Wang J, et al. Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med. 2002;47(6):1202–10. Epub 2002/07/12. doi: 10.1002/mrm.10171 12111967.

4. Chen F, Taviani V, Malkiel I, Cheng JY, Tamir JI, Shaikh J, et al. Variable-Density Single-Shot Fast Spin-Echo MRI with Deep Learning Reconstruction by Using Variational Networks. Radiology. 2018:336–73. Epub 2018/07/24. doi: 10.1148/radiol.2018180445 30040039.

5. Hammernik K, Klatzer T, Kobler E, Recht MP, Sodickson DK, Pock T, et al. Learning a variational network for reconstruction of accelerated MRI data. Magn Reson Med. 2018;79(6):3055–71. doi: 10.1002/mrm.26977 29115689; PubMed Central PMCID: PMC5902683.

6. Lee D, Yoo J, Tak S, Ye JC. Deep Residual Learning for Accelerated MRI Using Magnitude and Phase Networks. IEEE Trans Biomed Eng. 2018;65(9):1985–95. Epub 2018/04/02. doi: 10.1109/TBME.2018.2821699 29993390.

7. Mardani M, Gong E, Cheng JY, Vasanawala SS, Zaharchuk G, Xing L, et al. Deep Generative Adversarial Neural Networks for Compressive Sensing (GANCS) MRI. IEEE Trans Med Imaging. 2018;38(1):167–79. Epub 2018/07/23. doi: 10.1109/TMI.2018.2858752 30040634.

8. Yang G, Yu S, Dong H, Slabaugh G, Dragotti PL, Ye X, et al. DAGAN: Deep De-Aliasing Generative Adversarial Networks for Fast Compressed Sensing MRI Reconstruction. IEEE Trans Med Imaging. 2018;37(6):1310–21. doi: 10.1109/TMI.2017.2785879 29870361.

9. Quan TM, Nguyen-Duc T, Jeong WK. Compressed Sensing MRI Reconstruction Using a Generative Adversarial Network With a Cyclic Loss. IEEE Trans Med Imaging. 2018;37(6):1488–97. doi: 10.1109/TMI.2018.2820120 29870376.

10. Yang Y, Sun J, Li H, Xu Z. ADMM-Net: A Deep Learning Approach for Compressive Sensing MRI. 30th Conference on Neural Information Processing Systems (NIPS 2016)2016. p. 10–8.

11. Wang S, Su Z, Ying L, Peng X, Zhu S, Liang F, et al. Accelerating magnetic resonance imaging via deep learning. 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI). Prague, Czech Republic: IEEE; 2016: 514–517.

12. Qin C, Hajnal JV, Rueckert D, Schlemper J, Caballero J, Price AN. Convolutional Recurrent Neural Networks for Dynamic MR Image Reconstruction. IEEE Trans Med Imaging. 2018;38(1):280–90. Epub 2018/08/06. doi: 10.1109/TMI.2018.2863670 30080145.

13. Eo T, Jun Y, Kim T, Jang J, Lee HJ, Hwang D. KIKI-net: cross-domain convolutional neural networks for reconstructing undersampled magnetic resonance images. Magn Reson Med. 2018;80(5):2188–201. Epub 2018/04/06. doi: 10.1002/mrm.27201 29624729.

14. Han Y, Yoo J, Kim HH, Shin HJ, Sung K, Ye JC. Deep learning with domain adaptation for accelerated projection-reconstruction MR. Magn Reson Med. 2018;80(3):1189–205. Epub 2018/02/04. doi: 10.1002/mrm.27106 29399869.

15. Aggarwal HK, Mani MP, Jacob M. MoDL: Model Based Deep Learning Architecture for Inverse Problems. IEEE Trans Med Imaging. 2018, Epub 2018/08/13. doi: 10.1109/TMI.2018.2865356 30106719.

16. Kwon K, Kim D, Park H. A parallel MR imaging method using multilayer perceptron. Med Phys. 2017;44(12):6209–24. Epub 2017/10/23. doi: 10.1002/mp.12600 28944971.

17. Schlemper J, Caballero J, Hajnal JV, Price AN, Rueckert D. A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstruction. IEEE Trans Med Imaging. 2018;37(2):491–503. doi: 10.1109/TMI.2017.2760978 29035212.

18. Akcakaya M, Moeller S, Weingartner S, Ugurbil K. Scan-specific robust artificial-neural-networks for k-space interpolation (RAKI) reconstruction: Database-free deep learning for fast imaging. Magn Reson Med. 2019;81(1):439–53. doi: 10.1002/mrm.27420 30277269; PubMed Central PMCID: PMC6258345.

19. Qian N. On the momentum term in gradient descent learning algorithms. Neural Netw. 1999;12(1):145–51. 12662723.

20. Abadi M, Barham P, Chen J, Chen Z, Davis A, Dean J, et al. Tensorflow: a system for large-scale machine learning. OSDI; 2016.

21. Zhang C, Weingärtner S, Moeller S, Uğurbil K, Akçakaya M, editors. Fast GPU Implementation of a Scan-Specific Deep Learning Reconstruction for Accelerated Magnetic Resonance Imaging. 2018 IEEE International Conference on Electro/Information Technology (EIT); 2018 3–5 May 2018.

22. Kingma D, Ba J. Adam: A method for stochastic optimization. the 3rd International Conference on Learning Representations (ICLR 2015)2015.

23. Zbontar J, Knoll F, Sriram A, Muckley MJ, Bruno M, Defazio A, et al. fastMRI: An Open Dataset and Benchmarks for Accelerated MRI preprint. 2018:arXiv:1811.08839.

24. Zhao R, Song W, Zhang W, Xing T, Lin J-H, Srivastava M, et al. Accelerating Binarized Convolutional Neural Networks with Software-Programmable FPGAs. Proceedings of the 2017 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays; Monterey, California, USA. 3021741: ACM; 2017. p. 15–24.

25. Li C, Yang Y, Feng M, Chakradhar S, Zhou H. Optimizing Memory Efficiency for Deep Convolutional Neural Networks on GPUs. SC '16: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis; 2016 13–18 Nov. 2016.

26. Caruana R. Multitask Learning. Machine Learning. 1997;28(1):41–75. doi: 10.1023/a:1007379606734

27. Hussein S, Cao K, Song Q, Bagci U. Risk Stratification of Lung Nodules Using 3D CNN-Based Multi-task Learning2017; Cham: Springer International Publishing.

28. Zhang L, Karanikolas GV, Akçakaya M, Giannakis GB,s. Fully Automatic Segmentation of the Right Ventricle Via Multi-Task Deep Neural Networks. 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP); 2018 15–20 April 2018.

29. Han Y, Yoo J, Kim HH, Shin HJ, Sung K, Ye JC. Deep learning with domain adaptation for accelerated projection-reconstruction MR. Magnetic Resonance in Medicine. 2018;80(3):1189–205. doi: 10.1002/mrm.27106 29399869

30. Dar SUH, Cukur T. Transfer learning for reconstruction of accelerated MRI acquisitions via neural networks. Proceedings of the 26th Scientific Meeting of ISMRM; 2018 June; Paris: Proceedings of the 26th Scientific Meeting of ISMRM.

31. Eldar YC, A. O. Hero I, Deng L, Fessler J, Kovacevic J, Poor HV, et al. Challenges and Open Problems in Signal Processing: Panel Discussion Summary from ICASSP 2017 [Panel and Forum]. IEEE Signal Processing Magazine. 2017;34(6):8–23. doi: 10.1109/MSP.2017.2743842


Článek vyšel v časopise

PLOS One


2019 Číslo 10