MR Fingerprinting
Dan Ma1

1Radiology, Case Western Reserve University, OH, United States

Synopsis

Quantitative MR measurements are essential to assess complex changes in the brain and monitor treatment outcomes. Although full quantitative multi-parametric acquisition has long been the goal of research in MR, the conventional methods typically provide information on a single parameter at a time, thus requiring significant scan time. The purpose of Magnetic Resonance Fingerprinting (MRF) is to introduce a new framework to data acquisition and post-processing that permits the simultaneous quantification of multiple tissue properties in a time efficient manner.

Target Audience

MR physicist and physicians interested in quantitative MRI

Objectives

  • Introduce basic framework of the MRF
  • Discuss various MRF sequence designs and results
  • Review clinical applications of MRF

Purpose

Quantitative MR measurements are essential to assess complex changes in the brain and monitor treatment outcomes. Although full quantitative multi-parametric acquisition has long been the goal of research in MR, the conventional methods typically provide information on a single parameter at a time, thus requiring significant scan time. The purpose of Magnetic Resonance Fingerprinting (MRF)(1) is to introduce a new framework to data acquisition and post-processing that permits the simultaneous quantification of multiple tissue properties in a time efficient manner.

Method

Here we will discuss three key components of the MRF framework: data acquisition, dictionary generation and pattern recognition. Instead of using a repeated, serial acquisition of data to quantify individual parameter of interest, MRF uses a variable acquisition that makes the signal from different tissues to have a unique signal evolution, which could be simultaneously a function of multiple parameters of interest. This is accomplished by varying parameters such as flip angle and phase of RF pulses, TR, TE and sampling patterns during the acquisition. The goal is to generate spatial and temporal incoherence signals with maximal sensitivity to the target tissue properties. After the data are acquired, the separation of the signal into different material or tissue types can be achieved through pattern recognition. In MRF, this pattern recognition can take place through many means. For example, a dictionary that contains signal evolutions from all foreseeable combination of tissues and system-related parameters can be constructed through Bloch Equation(1–3), extended phase graph(4,5), or other analytical or numerical models(6) depending on the sequence type and the parameters of interest. The pattern recognition algorithm is then used to match the signal to the dictionary, the result of which identifies the most likely parameter combination in each of the voxels, and can then be translated to multiple quantitative maps.

Discussions

This review will be focusing on the following main features of the MRF.

First, MRF is a full quantitative method with high accuracy, scan efficiency and repeatability(1,7). Because MRF is based on pattern recognition to a dictionary, where interactions between multiple tissue properties and system variations are predicted, MRF could be more accurate and time efficient to estimate quantitative values as compared to conventional quantitative methods. For the same reason, the pattern recognition basis of MRF also makes MRF less sensitive to errors during measurement. This allows fast acquisition using single shot non-Cartesian trajectories, because incoherent aliasing artifacts are largely ignored by the pattern recognition process. The high accuracy, scan efficiency and repeatability allow the MRF method to be applied to various clinical studies, seeking for quantitative evaluation of tissue properties and diseases. For example, MRF has been used to analyze age and gender related changes of T1 and T2 values in different brain regions(8), to distinguish between healthy and pathological brain tissue with tumor and stroke models(9), to differentiate tumor regions from gliomas to metastases(10), to differentiate prostate cancer from normal peripheral zone(11), and to identify differences in normal appearing brain tissue between multiple sclerosis (MS) patients and healthy controls(12).

Second, because there is no requirement of the signal shape, MRF brings many degrees of freedom into the data acquisition. In other words, there are near-infinite possibilities for MRF compatible pulse sequence designs. This feature allows researchers to explore and simulate any properties that could contribute to signal changes. These include various tissue properties, such as T1, T2, T2*(13), perfusion(14), diffusion(15), cerebral blood volume, mean vessel radius and blood oxygen saturation(6), as well as system-related parameters, such as B0(1) and B1(16,17). The sequences can be adapted to different body parts such as brain, cardiac(18), abdominal(19) and prostate(11), and can also be designed for additional purposes, such as for ultra-low or ultra-high field(17,20), to reduce SAR(21) and to improve patient comfort(22). MRF can also be incorporated with advanced acquisition techniques such as multi-band and parallel imaging(23–25).

Third, MRF is related to the concept of compressed sensing in the sense that the acquisition generates incoherent signals and the number of key parameters to be quantified is much less than the amount of data acquired during the scan. Therefore, the data can either be compressed using various CS methods, with the purpose of reducing the scan time and post-processing time, reducing memory consumption, and improving image quality(26–32), or further analyzed to exploit more underlying properties, such as partial volume(33,34) and chemical exchange(35).

In summary, MRF introduces a flexible framework for quantitative MR and offers tremendous opportunities for both researches and clinical applications.

Acknowledgements

The author would like to acknowledge research support from Siemens Healthcare and NIH grants NIH 1R01EB016728-01A1 and NIH 5R01EB017219-02

References

1. Ma D, Gulani V, Seiberlich N, Liu K, Sunshine JL, Duerk JL, Griswold MA. Magnetic Resonance Fingerprinting. Nature 2013;495:187–192. doi: 10.1038/nature11971.

2. Bloch F, Rabi II, Millman S, Kusch P, Zacharias JR, Bloch ~ F, Hansen WW, Packard M. Nuclear Induction. 1946;70.

3. Hargreaves B. Bloch Equation Simulation.

4. Jiang Y, Ma D, Seiberlich N, Gulani V, Griswold M a. MR fingerprinting using fast imaging with steady state precession (FISP) with spiral readout. Magn. Reson. Med. 2014;0:n/a-n/a. doi: 10.1002/mrm.25559.

5. Weigel M. Extended phase graphs: Dephasing, RF pulses, and echoes - pure and simple. J. Magn. Reson. Imaging 2014;0. doi: 10.1002/jmri.24619.

6. Christen T, Pannetier N a, Ni WW, Qiu D, Moseley ME, Schuff N, Zaharchuk G. MR Vascular Fingerprinting: A New Approach to Compute Cerebral Blood Volume, Mean Vessel Radius, and Oxygenation Maps in the Human Brain. Neuroimage 2014;89:262–70. doi: 10.1016/j.neuroimage.2013.11.052.

7. Jiang Y, Ma D, Keenan KE, Stupic KF, Gulani V, Griswold MA. Repeatability of magnetic resonance fingerprinting T1 and T2 estimates assessed using the ISMRM/NIST MRI system phantom. Magn. Reson. Med. 2016;0:1–6. doi: 10.1002/mrm.26509.

8. Badve C, Yu A, Rogers M, Ma D, Liu Y, Schluchter M, Sunshine J, Griswold M, Gulani V. Simultaneous T 1 and T 2 Brain Relaxometry in Asymptomatic Volunteers Using Magnetic Resonance Fingerprinting. Tomography 2015;1. doi: 10.18383/j.tom.2015.00166.

9. Lemasson B, Pannetier N, Coquery N, Boisserand LSB, Collomb N, Schuff N, Moseley M, Zaharchuk G, Barbier EL, Christen T. MR Vascular Fingerprinting in Stroke and Brain Tumors Models. Sci Rep 2016;6:37071. doi: 10.1038/srep37071.

10. Badve C, Yu A, Dastmalchian S, et al. MR Fingerprinting of Adult Brain Tumors: Initial Experience. Am. J. Neuroradiol. 2016. doi: 10.3174/ajnr.A5035.

11. Yu A, Badve C, Ponsky L, et al. Combined MR Fingerprinting and Diffusion Examination for Prostate Cancer. Radiology 2017;0. doi: 10.1148/radiol.2017161599.

12. Nakamura K, Deshmane A, Guruprakash D, et al. A Novel Method for Quantification of Normal Appearing Brain Tissue in Multiple Sclerosis: Magnetic Resonance Fingerprinting. Neurology 2016;86:P4.158.

13. Rieger B, Zimmer F, Zapp J, Weingärtner S, Schad LR. Magnetic resonance fingerprinting using echo-planar imaging: Joint quantification of T1 and T2∗ relaxation times. Magn. Reson. Med. 2016. doi: 10.1002/mrm.26561.

14. Su P, Mao D, Liu P, Li Y, Pinho MC, Welch BG, Lu H. Multiparametric estimation of brain hemodynamics with MR fingerprinting ASL. Magn. Reson. Med. 2016. doi: 10.1002/mrm.26587.

15. Jiang Y, Hamilton J, Wright K, Ma D, Seiberlich N, Gulani V, Griswold M. Simultaneous Quantification of T1, T2 and Diffusion with Diffusion-weighted drive-equilibrium prepared Magnetic Resonance Fingerprinting. In: Proc. Intl. Soc. Mag. Reson. Med. ; 2016.

16. Buonincontri G, Sawiak SJ. MR fingerprinting with simultaneous B1 estimation. Magn. Reson. Med. 2015;0:1–9. doi: 10.1002/mrm.26009.

17. Cloos MA, Knoll F, Zhao T, Block K, Bruno M, Wiggins C, Sodickson D. Multiparametric imaging with heterogeneous radiofrequency field: Additional material. Nat. Commun. 2016. doi: 10.1017/CBO9781107415324.004.

18. Hamilton JI, Jiang Y, Chen Y, Ma D, Lo W-C, Griswold M, Seiberlich N. MR fingerprinting for rapid quantification of myocardial T1, T2, and proton spin density. Magn. Reson. Med. 2016;0:n/a-n/a. doi: 10.1002/mrm.26216.

19. Chen Y, Jiang Y, Pahva Shivani, Ma D, Lu L, Tweig MD, Wright KL, Seiberlich N, Griswold MA, Gulani V. MR Fingerprinting for Rapid Quantitative Abdominal Imaging. Radiology 2016;0:1–9.

20. Gao Y, Chen Y, Ma D, et al. Preclinical MR fingerprinting (MRF) at 7 T: effective quantitative imaging for rodent disease models. NMR Biomed. 2015;28:384–94. doi: 10.1002/nbm.3262.

21. Jiang Y, Ma D, Jerecic R, Duerk J, Seiberlich N, Gulani V, Griswold MA. MR fingerprinting using the quick echo splitting NMR imaging technique. Magn. Reson. Med. 2017;77:979–988. doi: 10.1002/mrm.26173.

22. Ma D, Pierre EY, Jiang Y, Schluchter MD, Setsompop K, Gulani V, Griswold MA. Music-based magnetic resonance fingerprinting to improve patient comfort during MRI examinations. Magn. Reson. Med. 2015. doi: 10.1002/mrm.25818.

23. Ye H, Ma D, Jiang Y, Cauley SF, Du Y, Wald LL, Griswold M a., Setsompop K. Accelerating magnetic resonance fingerprinting (MRF) using t-blipped simultaneous multislice (SMS) acquisition. Magn. Reson. Med. 2015;0:n/a-n/a. doi: 10.1002/mrm.25799.

24. Ye H, Cauley SF, Gagoski B, Bilgic B, Ma D, Jiang Y, Du YP, Griswold MA, Wald LL, Setsompop K. Simultaneous multislice magnetic resonance fingerprinting (SMS-MRF) with direct-spiral slice-GRAPPA (ds-SG) reconstruction. Magn. Reson. Med. 2016;0. doi: 10.1002/mrm.26271.

25. Jiang Y, Ma D, Bhat H, Ye H, Cauley SF, Wald LL, Setsompop K, Griswold MA. Use of pattern recognition for unaliasing simultaneously acquired slices in simultaneous multislice MR fingerprinting. Magn. Reson. Med. 2016. doi: 10.1002/mrm.26572.

26. Zhao B, Setsompop K, Ye H, Cauley SF, Wald LL. Maximum Likelihood Reconstruction for Magnetic Resonance Fingerprinting. IEEE Trans. Med. Imaging 2016;35:1812–1823. doi: 10.1038/nature11971.

27. Pierre EY, Ma D, Chen Y, Badve C, Griswold MA. Multiscale reconstruction for MR fingerprinting. Magn. Reson. Med. 2016;75:2481–2492. doi: 10.1002/mrm.25776.

28. McGivney D, Pierre E, Ma D, Jiang Y, Saybasili H, Gulani V, Griswold MA. SVD Compression for Magnetic Resonance Fingerprinting in the Time Domain. IEEE Trans. Med. Imaging 2014;62:1–13. doi: 10.1109/TMI.2014.2337321.

29. Cauley SF, Setsompop K, Ma D, Jiang Y, Ye H, Adalsteinsson E, Griswold M a, Wald LL. Fast Group Matching for MR Fingerprinting Reconstruction. Magn. Reson. Med. 2014;0:1–6. doi: 10.1002/mrm.25439.

30. Assländer J, Cloos MA, Knoll F, Sodickson DK, Hennig J, Lattanzi R. Low rank alternating direction method of multipliers reconstruction for MR fingerprinting. Magn. Reson. Med. 2017. doi: 10.1002/mrm.26639.

31. Davies M, Puy G, Vandergheynst P, Wiaux Y. A Compressed Sensing Framework for Magnetic Resonance Fingerprinting *. SIAM J. Imaging Sci. 2014;7:2623–2656. doi: 10.1137/130947246.

32. Wang Z, Zhang Q, Yuan J, Wang X. MRF denoising with compressed sensing and adaptive filtering. In: 2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI). IEEE; 2014. pp. 870–873. doi: 10.1109/ISBI.2014.6868009.

33. Deshmane A, Badve C, Rogers M, Yu A, Ma D, Sunshine J, Gulani V, Griswold M. Tissue Mapping in Brain Tumors with Partial Volume Magnetic Resonance Fingerprinting (PV-MRF). In: Proc. Intl. Soc. Mag. Reson. Med. ; 2015.

34. Mcgivney D, Deshmane A, Jiang Y, Ma D, Griswold M. A Bayesian Approach to the Partial Volume Problem in Magnetic Resonance Fingerprinting. In: Proc. Intl. Soc. Mag. Reson. Med. ; 2015.

35. Hamilton JI, Deshmane A, Hougen S, Griswold M, Seiberlich N. Magnetic Resonance Fingerprinting with Chemical Exchange (MRF-X) for Quantification of Subvoxel T1, T2, Volume Fraction, and Exchange Rate. In: Proc. Intl. Soc. Mag. Reson. Med. ; 2015.

Proc. Intl. Soc. Mag. Reson. Med. 25 (2017)