Improving Arterial Spin Labeling by Using Deep Learning

Published Online:

Deep learning algorithms generated arterial spin labeling perfusion images successfully and showed superior performance over the conventional averaging method, regardless of the MR imager, labeling schemes, readout schemes, and presence of pathologic conditions.


To develop a deep learning algorithm that generates arterial spin labeling (ASL) perfusion images with higher accuracy and robustness by using a smaller number of subtraction images.

Materials and Methods

For ASL image generation from pair-wise subtraction, we used a convolutional neural network (CNN) as a deep learning algorithm. The ground truth perfusion images were generated by averaging six or seven pairwise subtraction images acquired with (a) conventional pseudocontinuous arterial spin labeling from seven healthy subjects or (b) Hadamard-encoded pseudocontinuous ASL from 114 patients with various diseases. CNNs were trained to generate perfusion images from a smaller number (two or three) of subtraction images and evaluated by means of cross-validation. CNNs from the patient data sets were also tested on 26 separate stroke data sets. CNNs were compared with the conventional averaging method in terms of mean square error and radiologic score by using a paired t test and/or Wilcoxon signed-rank test.


Mean square errors were approximately 40% lower than those of the conventional averaging method for the cross-validation with the healthy subjects and patients and the separate test with the patients who had experienced a stroke (P < .001). Region-of-interest analysis in stroke regions showed that cerebral blood flow maps from CNN (mean ± standard deviation, 19.7 mL per 100 g/min ± 9.7) had smaller mean square errors than those determined with the conventional averaging method (43.2 ± 29.8) (P < .001). Radiologic scoring demonstrated that CNNs suppressed noise and motion and/or segmentation artifacts better than the conventional averaging method did (P < .001).


CNNs provided superior perfusion image quality and more accurate perfusion measurement compared with those of the conventional averaging method for generation of ASL images from pair-wise subtraction images.

© RSNA, 2017


  • 1. Detre JA, Leigh JS, Williams DS, Koretsky AP. Perfusion imaging. Magn Reson Med 1992;23(1):37–45. Crossref, MedlineGoogle Scholar
  • 2. Williams DS, Detre JA, Leigh JS, Koretsky AP. Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci U S A 1992;89(1):212–216 Crossref, MedlineGoogle Scholar
  • 3. Alsop DC, Detre JA, Golay X, et al. Recommended implementation of arterial spin-labeled perfusion MRI for clinical applications: A consensus of the ISMRM perfusion study group and the European consortium for ASL in dementia. Magn Reson Med 2015;73(1):102–116. Crossref, MedlineGoogle Scholar
  • 4. Bibic A, Knutsson L, Ståhlberg F, Wirestam R. Denoising of arterial spin labeling data: wavelet-domain filtering compared with Gaussian smoothing. MAGMA 2010;23(3):125–137. Crossref, MedlineGoogle Scholar
  • 5. Hu WT, Wang Z, Lee VM, Trojanowski JQ, Detre JA, Grossman M. Distinct cerebral perfusion patterns in FTLD and AD. Neurology 2010;75(10):881–888 Crossref, MedlineGoogle Scholar
  • 6. Wells JA, Thomas DL, King MD, Connelly A, Lythgoe MF, Calamante F. Reduction of errors in ASL cerebral perfusion and arterial transit time maps using image de-noising. Magn Reson Med 2010;64(3):715–724 Crossref, MedlineGoogle Scholar
  • 7. Frouin F, Duteil S, Lesage D, Carlier PG, Herment A, Leroy-Willig A. An automated image-processing strategy to analyze dynamic arterial spin labeling perfusion studies. Application to human skeletal muscle under stress. Magn Reson Imaging 2006;24(7):941–951. Crossref, MedlineGoogle Scholar
  • 8. Matsugu M, Mori K, Mitari Y, Kaneda Y. Subject independent facial expression recognition with robust face detection using a convolutional neural network. Neural Netw 2003;16(5-6):555–559. Crossref, MedlineGoogle Scholar
  • 9. LeCun Y, Bengio Y, Hinton G. Deep learning. Nature 2015;521(7553):436–444. Crossref, MedlineGoogle Scholar
  • 10. Greenspan H, van Ginneken B, Summers RM. Deep Learning in Medical Imaging: Overview and Future Promise of an Exciting New Technique. IEEE Trans Med Imaging. 2016;35(5):1153–1159. CrossrefGoogle Scholar
  • 11. Gondara L. Medical Image Denoising Using Convolutional Denoising Autoencoders. 2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW), 2016; 241–246. Google Scholar
  • 12. Chen H, Zhang Y, Zhang W, et al. Low-dose CT denoising with convolutional neural network. ArXiv e-prints [serial online]. 2016; vol 1610. Accessed October 1, 2016. Google Scholar
  • 13. Han YS, Yoo J, Ye JC. Deep Residual Learning for Compressed Sensing CT Reconstruction via Persistent Homology Analysis. ArXiv e-prints [serial online]. 2016; vol 1611. Accessed November 1, 2016. Google Scholar
  • 14. Ioffe S, Szegedy C. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. arXiv preprint. 2015;arXiv:1502.03167. Google Scholar
  • 15. Hinton GE, Srivastava N, Krizhevsky A, Sutskever I, Salakhutdinov RR. Improving neural networks by preventing co-adaptation of feature detectors. ArXiv e-prints [serial online]. 2012; vol 1207. Accessed July 1, 2012. Google Scholar
  • 16. Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv preprint. 2014;arXiv:1409.1556. Google Scholar
  • 17. Havaei M, Davy A, Warde-Farley D, et al. Brain tumor segmentation with Deep Neural Networks. Med Image Anal 2017;35:18–31 Crossref, MedlineGoogle Scholar
  • 18. Kamnitsas K, Ledig C, Newcombe VF, et al. Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation. Med Image Anal 2017;36:61–78 Crossref, MedlineGoogle Scholar
  • 19. Yu F, Koltun V. Multi-Scale Context Aggregation by Dilated Convolutions. arXiv preprint. 2015;arXiv:1511.07122. Google Scholar
  • 20. He K, Zhang X, Ren S, Sun J. Deep Residual Learning for Image Recognition. ArXiv e-prints [serial online]. 2015; vol 1512. Accessed December 1, 2015. Google Scholar
  • 21. Kim J, Lee JK, Lee KM. Accurate Image Super-Resolution Using Very Deep Convolutional Networks. ArXiv e-prints [serial online]. 2015; vol 1511. Accessed November 1, 2015. Google Scholar
  • 22. Han PK, Choi SH, Park SH. Investigation of control scans in pseudo-continuous arterial spin labeling (pCASL): Strategies for improving sensitivity and reliability of pCASL. Magn Reson Med 2017;78(3):917–929. Crossref, MedlineGoogle Scholar
  • 23. van Dyk DA, Meng XL. The art of data augmentation. J Comput Graph Stat 2001;10(1):1–50. CrossrefGoogle Scholar
  • 24. Wells JA, Lythgoe MF, Gadian DG, Ordidge RJ, Thomas DL. In vivo Hadamard encoded continuous arterial spin labeling (H-CASL). Magn Reson Med 2010;63(4):1111–1118. Crossref, MedlineGoogle Scholar
  • 25. Wang DJ, Alger JR, Qiao JX, et al. Multi-delay multi-parametric arterial spin-labeled perfusion MRI in acute ischemic stroke - Comparison with dynamic susceptibility contrast enhanced perfusion imaging. Neuroimage Clin 2013;3:1–7. Crossref, MedlineGoogle Scholar
  • 26. Dai W, Robson PM, Shankaranarayanan A, Alsop DC. Reduced resolution transit delay prescan for quantitative continuous arterial spin labeling perfusion imaging. Magn Reson Med 2012;67(5):1252–1265. Crossref, MedlineGoogle Scholar
  • 27. Vedaldi A, Lenc K. MatConvNet - Convolutional Neural Networks for MATLAB. ArXiv e-prints [serial online]. 2014; vol 1412. Accessed December 1, 2014. Google Scholar
  • 28. Sinha NK, Griscik MP. Concerning Comments on a Stochastic Approximation Method. IEEE Trans Syst Man Cybern B Cybern 1973;SMC3(5):526. Google Scholar
  • 29. van Osch MJ, Teeuwisse WM, Chen Z, Suzuki Y, Helle M, Schmid S. Advances in arterial spin labelling MRI methods for measuring perfusion and collateral flow. J Cereb Blood Flow Metab 2017 Jan 1:271678X17713434 [Epub ahead of print]. CrossrefGoogle Scholar
  • 30. Ledig C, Theis L, Huszar F, et al. Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network. arXiv preprint. 2016;arXiv:1609.04802. Google Scholar

Article History

Received May 18, 2017; revision requested July 20; revision received September 8; accepted September 29; final version accepted October 12.
Published online: Dec 21 2017
Published in print: May 2018