Modern Functional & Diffusion Imaging
Mara Cercignani1

1Brighton & Sussex Medical School, United Kingdom


Diffusion and functional MRI are among the most popular MRI techniques used in neuroscience research. Although basic versions of these techniques have found their way to the clinic relatively soon after their introduction, almost none of the technological advances introduced in the past 20 years have been adopted outside the research setting. This lecture will review some of the latest flavours of diffusion and functional MRI and their potential clinical applications


Diffusion (dMRI) and functional (fMRI) MRI were introduced in the early 90s [1, 2]. They soon became extremely popular within the MRI community and started to dominate the methodological sessions of ISMRM. This keen interest has prompted a rapid evolution of both techniques, contributing to stimulate some of the most important MRI hardware developments of the past 20 years. dMRI was very soon adopted in clinical routine thanks to its unique sensitivity to cytotoxic oedema [3], making it the method of choice for diagnosing acute ischaemia. Later, both fMRI and dMRI tractography started to be used for pre-surgical mapping in patients with brain tumours [4, 5]. In parallel, both physicists and mathematicians continued to develop these techniques addressing their limitations and pitfalls, and developing increasingly complex models of tissue. Almost none of these developments have ever made it to the clinic. Although this is partly justified by the difficulty of implementing lengthy acquisition and processing methods in such a context, this lecture will discuss a few examples of advanced fMRI and dMRI that could ‘make a difference’ for clinical applications.

Diffusion MRI

dMRI is based on the random motion of water molecules within tissue. As such a motion is affected by tissue structure, dMRI can indirectly reflect tissue microstructure [6]. The early observation that diffusion was anisotropic in the white matter led to the development of diffusion tensor imaging (DTI), which became very popular also thanks to its ability to infer tissue orientation and thus to enable the reconstruction of the main white matter tracts with diffusion tractography [2, 7]. The assumption of Gaussianity of the tensor model allows the reconstruction of a single direction and a single water compartment per voxel. The realisation that this was too simplistic for describing the complex tissue microstructure of the brain led on the one hand to the development of high angular resolution methods (eg, [8, 9]) for the purpose of improving tractography, and on the other to the development of multicompartment models that allow intra and extra-cellular water to be separated based on their diffusion characteristics (eg., [10-12]). Recent work has demonstrated that dMRI can be used to infer the distribution of axonal radius and fiber density within an image voxel [13, 14]. The ability of resolving complex finer configurations with tractgraphy is directly relevant to pre-surgical applications as the white matter tracts most frequently investigated (motor and language pathways) are characterised by such configurations and the use of DTI might lead to a wrong or insufficient reconstruction (false negatives). This can have important consequences when applied to clinical cases. Higher order models can provide more accurate tractography information [15]. In addition, as most DTI-based tractography algorithms rely on anisotropy thresholds, often the presence of oedema can limit the applicability of tractography. In these cases tracts might appear to be “interrupted” by tumours, but “reappear” after oedema is absorbed. In this context, the use of diffusion models that include separate compartments for isotropic and anisotropic diffusion [12, 16] can help. Multicompartment models can also increase confidence in the localisation of the epileptogenic zone, invisible to other imaging modalities and particularly to DTI, in patients with focal cortical dysplasia [17]. The augmented microstructural information can reveal additional clinical information. Mathematical models of tumour tissue applied to dMRI have already been shown to be able to measure histologic features of preclinical colorectal and human prostate cancer in vivo [18]. While the complex microstructure of the brain prevents a direct translation of these models to brain tumours, there is scope for developing useful approaches based on the same concept. It is conceivable that such models might contain information relevant for diagnosis/prognosis (e.g., biomarkers for IDH-mutation in gliomas). Another area of potential application of sophisticated microstructural model is multiple sclerosis (MS). With the increases in spatial resolution achieved in recent years, it has become possible to differentiate MS lesions based on their dMRI properties, and link them to diverse pathological substrate with clinical relevance [19].

Functional MRI

fMRI exploits neurovascular coupling and the differing magnetic properties of oxy- and deoxy-haemoglobin to indirectly measure and localise neural activity. The most widely used methods is based on blood oxygenation level dependant (BOLD) contrast, which reflects a combination of changes in oxygen metabolism, blood flow and blood volume. The measured signal change is relative to an unknown baseline, making the interpretation of this signal and the quantification of brain activity very challenging. The well-known limitations of BOLD include the limited spatial specificity, caused by the fact that changes to blood flow and volume can occur within draining veins that are quite far from the activated area [20]. For clinical uses, a non-negligible element is the need for patient compliance: often patients are cognitively impaired or too unwell to perform the tasks required for an fMRI experiment. Resting state (rs) fMRI is typically used to measure brain connectivity [21], rather than activity. Nevertheless, recently the possibility of training an algorithm to predict activations from rsfMRI maps was explored [22, 23] with discrete success. The advantages of this approach are the possibility of obtaining diverse functional maps from a single acquisition, thus reducing scan time for patients, and that of obtaining useful data even if the patient finds the task difficult to perform. In order to address the limitations of BOLD fMRI mentioned above, other MRI methods, aiming at measuring more specific quantities (cerebral blood flow, cerebral blood volume, cerebrovascular reactivity, cerebral rate of oxygen metabolism, aka CMRO2 [24-28]) are under development. They hold potential for clinical applications because of their quantitative nature. These techniques, however, remain confined to research settings due to the complex set-up required for them, or the limited SNR currently achievable.


Functional and diffusion MRI hold an enormous potential for clinical applications. Recent technological advances have radically changed the degree on information we can access non-invasively using these techniques; these advances however have not made their way to the clinic. A closer collaboration/communication between developers and users can really help to shape them in a clinically feasible and informative format.


No acknowledgement found.


1. Ogawa, S., et al., Oxygenation-sensitive contrast in magnetic resonance image of rodent brain at high magnetic fields. Magn Reson Med, 1990. 14(1): p. 68-78.

2. Basser, P.J., J. Mattiello, and D. LeBihan, MR diffusion tensor spectroscopy and imaging. Biophys J, 1994. 66(1): p. 259-67.

3. Moseley, M.E., et al., Early detection of regional cerebral ischemia in cats: comparison of diffusion- and T2-weighted MRI and spectroscopy. Magn Reson Med, 1990. 14(2): p. 330-46.

4. Clark, C.A., et al., White matter fiber tracking in patients with space-occupying lesions of the brain: a new technique for neurosurgical planning? Neuroimage, 2003. 20(3): p. 1601-8.

5. Atlas, S.W., et al., Functional magnetic resonance imaging of regional brain activity in patients with intracerebral gliomas: findings and implications for clinical management. Neurosurgery, 1996. 38(2): p. 329-38.

6. LeBihan, D., IVIM method measures diffusion and perfusion. Diagn Imaging (San Franc), 1990. 12(6): p. 133, 136.

7. Conturo, T.E., et al., Tracking neuronal fiber pathways in the living human brain. Proc Natl Acad Sci U S A, 1999. 96(18): p. 10422-7.

8. Wiegell, M.R., H.B. Larsson, and V.J. Wedeen, Fiber crossing in human brain depicted with diffusion tensor MR imaging. Radiology, 2000. 217(3): p. 897-903.

9. Frank, L.R., Characterization of anisotropy in high angular resolution diffusion-weighted MRI. Magn Reson Med, 2002. 47(6): p. 1083-99.

10. Clark, C.A. and D. Le Bihan, Water diffusion compartmentation and anisotropy at high b values in the human brain. Magnetic Resonance in Medicine, 2000. 44(6): p. 852-859.

11. Assaf, Y., et al., New modeling and experimental framework to characterize hindered and restricted water diffusion in brain white matter. Magnetic Resonance in Medicine, 2004. 52(5): p. 965-978.

12. Zhang, H., et al., NODDI: practical in vivo neurite orientation dispersion and density imaging of the human brain. Neuroimage, 2012. 61(4): p. 1000-16.

13. Alexander, D.C., et al., Orientationally invariant indices of axon diameter and density from diffusion MRI. Neuroimage, 2010. 52(4): p. 1374-89.

14. Assaf, Y., et al., AxCaliber: a method for measuring axon diameter distribution from diffusion MRI. Magn Reson Med, 2008. 59(6): p. 1347-54.

15. Farquharson, S., et al., White matter fiber tractography: why we need to move beyond DTI. J Neurosurg, 2013. 118(6): p. 1367-77.

16. Pasternak, O., et al., Free water elimination and mapping from diffusion MRI. Magn Reson Med, 2009. 62(3): p. 717-30. 17. Winston, G.P., et al., Advanced diffusion imaging sequences could aid assessing patients with focal cortical dysplasia and epilepsy. Epilepsy Res, 2014. 108(2): p. 336-9. 18. Panagiotaki, E., et al., Noninvasive quantification of solid tumor microstructure using VERDICT MRI. Cancer Res, 2014. 74(7): p. 1902-12.

19. Cercignani, M., et al. A multi-parametric study of MS lesions. in ISMRM. 2018. Paris, France.

20. Turner, R., BOLD localization: the implications of vascular architecture. Neuroimage, 2001. 13(6): p. S1011-S1011.

21. Dipasquale, O. and M. Cercignani, Network functional connectivity and whole-brain functional connectomics to investigate cognitive decline in neurodegenerative conditions. Functional Neurology, 2016. 31(4): p. 191-203.

22. Tavor, I., et al., Task-free MRI predicts individual differences in brain activity during task performance. Science, 2016. 352(6282): p. 216-220.

23. Parker Jones, O., et al., Resting connectivity predicts task activation in pre-surgical populations. Neuroimage Clin, 2017. 13: p. 378-385.

24. Detre, J.A., et al., Applications of arterial spin labeled MRI in the brain. J Magn Reson Imaging, 2012. 35(5): p. 1026-37.

25. Driver, I., et al., The change in cerebrovascular reactivity between 3 T and 7 T measured using graded hypercapnia. Neuroimage, 2010. 51(1): p. 274-9.

26. Lu, H. and P.C. van Zijl, A review of the development of Vascular-Space-Occupancy (VASO) fMRI. Neuroimage, 2012. 62(2): p. 736-42.

27. Fan, A.P., et al., Baseline oxygenation in the brain: Correlation between respiratory-calibration and susceptibility methods. Neuroimage, 2016. 125: p. 920-931.

28. Zhang, J., et al., Cerebral metabolic rate of oxygen (CMRO2 ) mapping with hyperventilation challenge using quantitative susceptibility mapping (QSM). Magn Reson Med, 2017. 77(5): p. 1762-1773.

Proc. Intl. Soc. Mag. Reson. Med. 26 (2018)