Lobule-wise quantitative T1 and T2* analysis of cerebellar grey matter in multiple sclerosis patients at 7T MRI
Yohan Boillat1, Kieran O'Brien2,3, Mário João Fartaria de Oliveira4,5, Guillaume Bonnier1,5,6, Gunnar Krueger5,7, Wietske van der Zwaag1,8, and Cristina Granziera1,5,6,9

1Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, 2Siemens Healthcare Pty Ltd., Brisbane, Australia, 3University of Queensland, St-Lucia, Australia, 4University of Lausanne, Lausanne, Switzerland, 5Advanced Clinical Imaging Technology Group, Siemens, Lausanne, Switzerland, 6Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland, 7Healthcare Sector IM&WS S, Siemens Schweiz AG, Lausanne, Switzerland, 8Spinoza Centre for Neuroimaging, Amsterdam, Switzerland, 9Massachusetts General Hospital and Harvard Medical School, Charlestown, MA, United States


We compared ultra-high field, high resolution quantitative T1 and T2* measurements in the cerebellum of MS patients to that of healthy controls. A correlation between the multiple sclerosis functional scale scores and local T2* values was found for several motor and cognitive related lobules. No significant differences between groups were found.

Target audience

Physisists, neuroscientists and clinicians interested in multiple sclerosis biomarkers, the cerebellum and quantitative MR.


We investigated quantitative gray matter cerebellar changes at a lobule level in multiple sclerosis (MS) patients and correlated these with clinical assessment. MS, a global brain pathology, does not spare the cerebellum, where both focal white and grey matter lesions have been observed1. In addition, more subtle changes already occur in normal-appearing tissue2. Quantitative imaging, such as T1 and T2* relaxometry, provide information about the macromolecular structure, water and iron content of brain tissue3,4, and may support metrics to monitor disease progression. High-resolution quantitative measurements at 7T were used to study the normal-appearing gray matter (NAGM) in MS patients.


Eighteen early-stage MS patients (<5 years disease duration) were enrolled (4 females, median EDSS 1, 5; range: 1-2), age=33±7 (mean±SD) and nine healthy controls (2 females, age=31±7). Patients and controls underwent cognitive and behavioral testing as in4, including the MS functional composite score5 (MSFC), which is a combined score of three individual measurements independently assessing motor and cognitive function. Whole-brain T1 and T2*maps were acquired using the MP2RAGE6 (TR/TE/TI1/TI2=6000/2.84/750/2350ms, matrix=300x320x160, 0.75x0.75x0.9mm3) and 3D multi gradient echo (MGE; TR=45ms TE1/ΔTE/TE9=4.59/4.59/41.3ms, matrix=300x320x160, 0.75x0.75x0.9mm3) sequences at 7T (Siemens Health, Germany) using a 32 channel head coil (Nova Medical, USA). Three dielectric pads were placed around the cerebellum to improve B1 homogeneity. A mono-exponential fit to the MGE data was used to obtain S0 and T2*maps. A SA2RAGE B1 map7 (TR/TE=2400/0.72ms, matrix=116x128x64, 2.3x2.3x4mm3) was used to correct the MP2RAGE for B1 homogeneity8. T2* was registered to the MP2RAGE using Elastix9. The SUIT toolbox10 was used to normalize the MP2RAGE T1-weighted images and segment the individual cerebelli. An in-house semi-automatically generated mask for the focal lesions was used to exclude lesion sites from the atlas. A T1-map based mask (<2400ms) was used to remove CSF voxels from the atlas. Final segmentation quality was visually inspected (see Figure 1 for an example). NAGM T1 and T2*values were lobule-wise extracted. MS patient values were compared to controls’ with a repeated measure MANOVA (lobules as within-subject factor;patients/controls as between-subject factor). The MFSC score was correlated with lobule T1 and T2*values in the MS patients (Spearman correlation, FDR-adjusted).


The multivariate analysis MS patients against controls did not reveal any main effects. The outcome was similar for tests performed at univariate level. There was a trend for longer T1 values in the vermis rather than in the lateral regions, but this did not differ between controls and patients (Figure 2). Vermis and anterior lobules (IV, V, VI) showed on average the highest T2*values (Figure 3). In MS patients, the mean T2*value of several lobules showed a negative correlation with the MSFC score (Figure 4). More precisely, for these lobules, a lower performance during the clinical assessment was related to longer T2*values. These lobules were: left Crus I (rs=-.7 p=.044 FDR-corrected), left VIIb (rs=-.6, p=.049 FDR-corrected) and right VIIb (rs=-.6, p=.036 FDR-corrected). In addition, several other lobules showed a weaker negative correlation (rs=-.5, p<.05 uncorrected) with T2*. They were: right Crus I, vermis Crus II, left VIIIa, right VIIIa and right X. In the T1 maps, only a weak negative correlation was obtained for vermis VIIIb and IX (p<.05 uncorrected, rs=-.5; Figure 5). No significant correlations were obtained for controls.


MS patients showed higher T2*values in several bilateral cerebellar NAGM subparts for lower performances in the MSFC assessment whereas controls did not show any correlations. The nature of the T2*contrast would suggest a loss of macromolecules or a local decrease of tissue iron content4. The different regions affected are involved in several functional domains11, namely motor, executive function and working memory and reflect the composite nature of the MFSC which jointly assesses motor function of arm and leg as well as processing speed and working memory. We did not observe any significant differences in T1 and T2* properties of the cerebellar cortex between patients and controls. This may be due to insufficient power or to the fact that we studied early stage and minimally impaired patients, whereas demyelination and changes in cerebellar cortex volume have been so far reported in patients at more advanced disease stages.


This study showed that, even at early MS stages, mild alterations in the cerebellar cortex, as measured by quantitative T2* at ultra-high field, are related to cognitive and motor performances in subjects with mild clinical impairment. Future studies should assess the value of quantitative T1 and T2* in larger cohorts of MS patients including more advanced stages and clinical deficits.


No acknowledgement found.


1. Calabreses, M., Filippi, M. & Gallo, P. Cortical lesions in multiple sclerosis. Nat. Rev Neurol 6, 438–444 (2010).

2. Ramió-Torrentà, L. et al. Abnormalities in normal appearing tissues in early primary progressive multiple sclerosis and their relation to disability: a tissue specific magnetisation transfer study. J. Neurol. Neurosurg. Psychiatry 77, 40–5 (2006).

3. Ogg, R. J. & Steen, R. G. Age-related changes in brain T1 are correlated with iron concentration. Magn. Reson. Med. 40, 749–53 (1998).

4. Bonnier, G. et al. Advanced MRI unravels the nature of tissue alterations in early multiple sclerosis. Ann. Clin. Transl. Neurol. 1, 423–432 (2014).

5. Fischer, J. S., Rudick, R. A., Cutter, G. R. & Reingold, S. C. The Multiple Sclerosis Functional Composite Measure (MSFC): an integrated approach to MS clinical outcome assessment. National MS Society Clinical Outcomes Assessment Task Force. Mutliple Scler. 5, 244–50 (1999).

6. Marques, J. P. et al. MP2RAGE, a self bias-field corrected sequence for improved segmentation and T1-mapping at high field. Neuroimage 49, 1271–81 (2010).

7. Eggenschwiler, F., Kober, T., Magill, A. W., Gruetter, R. & Marques, J. P. SA2RAGE: a new sequence for fast B1+ -mapping. Magn. Reson. Med. 67, 1609–19 (2012).

8. Marques, J. P. & Gruetter, R. New developments and applications of the MP2RAGE sequence--focusing the contrast and high spatial resolution R1 mapping. PLoS One 8, e69294 (2013).

9. Klein, S., Staring, M., Murphy, K., Viergever, M. a. & Pluim, J. elastix: A Toolbox for Intensity-Based Medical Image Registration. IEEE Trans. Med. Imaging 29, 196–205 (2010).

10. Diedrichsen, J. A spatially unbiased atlas template of the human cerebellum. Neuroimage 33, 127–138 (2006).

11. Stoodley, C. J. & Schmahmann, J. D. Functional topography in the human cerebellum: A meta-analysis of neuroimaging studies. Neuroimage 44, 489–501 (2009).


Figure 1 Mapping of the SUIT atlas over an MS patient MP2RAGE in the subject space. 28 different cerebellar subregions were identified and their T1 and T2* values extracted. Segmentation was improved manually where necessary. The block in the white matter are focal lesions.

Figure 2 T1 measurements of the 28 different cerebellar subregions. The controls are represented by the blue bars and the MS patients by the green bars. The error bars represent the standard deviation.

Figure 3 T2* measurements of the 28 different cerebellar subregions. The controls are represented by the blue bars and the MS patients by the green bars. The error bars represent the standard deviation.

Figure 4 Scatter plot representing the MSFC scores with the T2* values for each lobule. The red squares represent the lobule showing a Spearman correlation between the MFSC and the T2* at FDR-corrected p-value. The orange squares represent a significant correlation at uncorrected p-value. Patients=green squares and controls=blue circles.

Figure 5 Scatter plot representing the MSFC scores with the T1 values for each lobule. The orange squares represent the lobule showing a Spearman correlation between the MFSC and the T1 at uncorrected p-value. Patients=green squares and controls=blue circles.

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)