- Email: [email protected]
Functional Magnetic Resonance Imaging of the Basal Ganglia and Cerebellum Using a Simple Motor Paradigm
Magnetic Resonance Imaging, Vol. 16, No. 3, pp. 281–287, 1998 © 1998 Elsevier Science Inc. All rights reserved. Printed in the USA. 0730-725X/98 $19.00 1 .00
PII S0730-725X(97)00276-2
● Original Contribution
FUNCTIONAL MAGNETIC RESONANCE IMAGING OF THE BASAL GANGLIA AND CEREBELLUM USING A SIMPLE MOTOR PARADIGM JU¨ RGEN R. REICHENBACH,* ROBERT FEIWELL,* KARTHIKEYAN KUPPUSAMY,*† MARK BAHN,* AND E. MARK HAACKE*† *Mallinckrodt Institute of Radiology, Washington University, School of Medicine, 510 S. Kingshighway, St. Louis, MO, and †Department of Electrical Engineering, Washington University, 1 Brookings Drive, St. Louis, MO, USA Activation of cortical and subcortical motor areas of the brain, including primary motor cortex, supplementary motor area, basal ganglia and cerebellum, were successfully investigated in seven right-handed, normal volunteers during a simple, rapid, thumb flexion-extension task using functional magnetic resonance imaging. A multi-slice echo-planar imaging sequence was used to cover the entire brain. A signal increase varying from 2% to 6% was observed for the different regions involved in the motor task. Moving the non-dominant thumb was associated with a more bilateral activation pattern in both putamen and cerebellar regions. This study demonstrates the capability of functional magnetic resonance imaging to delineate simultaneously many activated brain areas that are commonly thought to be involved in the performance of motor tasks. © 1998 Elsevier Science Inc. Keywords: Functional magnetic resonance imaging (fMRI); Motor cortex stimulation; Basal ganglia; Cerebellum; Self-paced movement.
biologic studies that, for example, the control of voluntary and even simple movements is a complex process that involves many different areas of the brain. It is widely believed that motor system control is achieved by a series of parallel systems formed by somatotopically organized, descending projections that link the various motor-related areas of the cortex more directly with spinal motor circuits.12–14 An important role is hereby played by subcortical structures, such as the basal ganglia and the cerebellum. The basal ganglia is a large, functionally heterogeneous structure arranged mainly in multiple parallel cortico-striato-thalamo-cortical circuits and involved in a wide variety of motor and affective behaviors, in sensorimotor integration, and in cognitive functions. With respect to motor behavior, the basal ganglia are believed to be involved in the determination of movement parameters, preparation for movement, enabling movement to become automatic, facilitation of sequential move-
INTRODUCTION In recent years, functional magnetic resonance imaging (fMRI) has become a versatile and important clinical as well as research tool to study non-invasively activation of the normal and diseased human brain. Since the observation that relaxation times depend on the blood oxygenation level1 (BOLD-contrast) and that changes of the latter can be measured in vivo,2 the possibility to examine cortical activity using fMRI methods has found widespread interest.3–10 Until recently, application of fMRI to map brain activation was restricted in practice to a limited number of slices covering only parts of the brain. With the advent of scanners having echo-planar imaging (EPI) capability with powerful and very rapid gradient systems, it has become possible to cover the whole brain using singleshot, multi-slice EPI acquisitions.11 This possibility has an important impact for a better understanding of functional neuroanatomy since it is well known from neuroRECEIVED 4/9/97; ACCEPTED 10/11/97. Address correspondence to Ju¨rgen R. Reichenbach, PhD, Institute of Diagnostic and Interventional Radiology, FriedrichSchiller-University Jena, Bachstrasse 18, 07740 Jena, Ger-
many. E-mail: [email protected] *Present address: Institute of Diagnostic and Interventional Radiology, Friedrich-Schiller-University, Jena, Germany 281
282
Magnetic Resonance Imaging ● Volume 16, Number 3, 1998
ment, inhibition of unwanted movements, adaptation to novel circumstances, facilitation of rewarded action, as well as motor learning and planning.14 The basal ganglia are usually taken to include the caudate nucleus and putamen (often jointly termed striatum), the globus pallidus (or pallidum), the subthalamic nucleus, and the substantia nigra. A motor control loop has been characterized involving the supplementary motor area (SMA), primary motor cortex, putamen, pallidum and ventrolateral thalamus.13,14 Damage of the basal ganglia can cause motion disorders, such as involuntary movements, muscular rigidity, and immobility without paralysis. The cerebellum plays a key role in movement and is considered the primary site of motor learning. It contributes primarily to motor coordination and control and receives input from virtually all brain areas. It is also involved in the control of posture, regulation of bodily function in response to a variety of stimuli, initiation of limb movements, adjustment of eye movements in response to hand movement, and fine manipulative movements. The cerebellum is one of the major sources of input, via the thalamus, to the primary motor cortex (area 4) and premotor cortex (lateral portion of area 6). However, its exact function in the timing of movement and motor learning is still unclear and controversial.15 The use of volume coverage fMRI using EPI offers a practical approach to more global studies of brain function because, based on the neuroanatomic knowledge, it can be expected that performing a motor task does not only activate areas in the motor cortex, but should also activate parts of the basal ganglia and the cerebellum. Whole-brain methods are thus essential to map simultaneously complex activation patterns that are spread across different areas. To date there have been limited activation studies of the basal ganglia using functional magnetic resonance imaging and these have been only local in nature.16,17 Recently, Bucher et al.16 have shown localized activation within the putamen and globus pallidus using high resolution, FLASH (fast low-angle shot) fMRI. Limitations of that study include prolonged imaging time and the limited number of slices acquired. The aim of this study was to examine the feasibility and reproducibility of visualizing simultaneous activation of motor cortex, SMA, basal ganglia and cerebellum during the performance of a simple, skilled, motor-specific task with a T2*-weighted, interleaved, echo-planar fMRI technique. This demonstrates the capability of fMRI in assessing global involvement of a task over the entire brain. MATERIALS AND METHODS Seven normal, right-handed, healthy subjects (six male, one female, mean age 31 years) participated in this
pilot study. Informed written consent was obtained from all subjects after the nature of the experiment had been fully explained and was approved by the Institutional Review Board and the Human Studies Committee. The activation state was a rapid (2–3 Hz), self-paced, flexionextension movement of the thumb (digit 1) of the dominant hand. Subjects had a practice session of about 15 min prior to the scans to ensure that they performed the task as consistently and rapidly as possible. They were also instructed to keep their eyes closed during the experiments. To investigate possible differences in the activation pattern for dominant vs. non-dominant exercise, three subjects were also imaged during non-dominant thumb movement. During acquisition of 50 images/slice, resting and active periods were alternated in an ‘‘off-on’’ manner for five cycles. Each cycle consisted of five resting and five activation images. A buzzer was used as an auditory trigger to indicate the onset of each ‘‘off’’ and ‘‘on’’ mode, respectively. The paradigm was repeated three times within the same scan session to investigate intra-subject variability. Subjects were placed supine on the MRI table. Foam cushions and pads were used to comfortably immobilize the head in the circular polarized head coil and ear plugs were used for noise reduction. Prior to imaging, shimming was performed to optimize the static field homogeneity. Usually, a gradient offset tolerance better than 0.001 mT/m was obtained. For orientation, sagittal anatomical scout measurements were performed. Based on these images and using the anterior commissure-posterior commissure (AC-PC) line as a rough neuroanatomic landmark, two-dimensional transaxial single-shot, multislice, echo-planar images covering the entire brain and cerebellum were acquired on a whole-body, 1.5 T scanner (Vision system, Siemens Medical Systems, Erlangen, Germany) with the following sequence parameters: repetition time 5 5 s, echo time 5 87 ms, flip angle a 5 90°, slice thickness TH 5 5 mm, 10 or 20 slices, matrix size 5 96 3 128 (sinc interpolated to 256 3 256), FOV 5 200 –230 mm. Corresponding T1-weighted images were collected with exactly the same orientation at the same slice locations to compare the anatomy with functional activation patterns using a three-dimensional magnetization-prepared rapid gradient echo (MP-RAGE) sequence. The imaging parameters were: repetition time 5 9.7 ms, echo time 5 4 ms, a 5 8°, slice thickness TH 5 2.5 mm, 80 partitions, matrix size 5 256 3 256. The field of view (FOV) was the same as for the functional studies. In this way, the EPI experiments could be anatomically correlated with the conventional T1weighted scans very easily. The subjects were visually monitored during the examination to confirm adherence to the experimental design. Intrasubject registration of the echo-planar data was
fMRI of basal ganglia and cerebellum ● J.R. REICHENBACH
Fig. 1. Overlay images showing areas of activation during dominant thumb movement in the primary motor cortex area and SMA. The maps are based on a correlation coefficient threshold of 0.3 (p , 0.02) and only show areas consisting of more than 20 clustered activated pixels. Pixels with correlation coefficients higher than the threshold are displayed in yellow. The images are displayed in standard radiologic orientation (PMC - primary motor cortex, SMA - supplementary motor area).
performed using the AIR 3.0 realignment software package developed by Roger Woods.18 The functional time courses were analyzed with the cross-correlation method and a box car reference waveform with the same period as the exercise cycles.19,20 The functional maps were generated pixelwise using a correlation coefficient threshold of 0.30 (p , 0.02, paired, two-tailed t-test) and activated pixels were clustered to at least 20 pixels to be considered significant with a high spatial consistency. These pixels were overlayed onto the EPI-images. The large cluster size is a very conservative means to increase the statistical significance of the observed activation. It was also chosen due to the fact that the EPI-images were sinc-interpolated from a low-resolution data set. RESULTS Clusters of activated pixels were found in all subjects. As an example, the functional images in Fig. 1 demonstrate the activation seen in consecutive slices containing the primary motor cortex and the SMA, while the subject performed the motor task with the dominant (right) hand. The response is clearly localized on the contralateral side with only minor activation on the ipsilateral hemisphere and can be followed throughout several slices. Ipsilateral activity, however, was not consistently observed across different subjects. Figure 2 displays the anatomical and functional images obtained in the basal ganglia and the
ET AL.
283
cerebellum for dominant and non-dominant thumb movement in one right-handed volunteer. For better illustration, anatomical and functional images were put side by side. For this volunteer, dominant thumb movement (second panel) resulted in activation of the contralateral putamen. In contrast, non-dominant thumb movement (third panel) yielded increased ipsilateral activity in association with the expected strong contralateral activation. The cerebellum shows ipsilateral, anterior hemispheric activation for both left and right thumb movement. However, with non-dominant motor activity the response is more bilaterally distributed. Figure 3 demonstrates the intra-subject repeatability of the observed functional activation. The volunteer repeated the paradigm during the same scan session. As can be clearly seen, the response in the second experiment (lower row) matches reasonably well the result obtained during the first experiment (upper row). Plots of fMRI time courses from regions of activated pixels in the primary motor cortex, supplementary motor area (SMA), putamen, and cerebellum are shown in Fig. 4. Typical signal increases of approximately 6% in the contralateral motor cortex, 2–3% in the contralateral putamen, and 4 –5% in the ipsilateral cerebellum were obtained, all of which are in agreement with the results of other groups19 –21 and the expected range from the BOLD effect at 1.5 Tesla.7 As can be seen from the time courses the signal in the activated states stays reasonably flat, especially in the cases of the motor cortex and cerebellum. Given this behavior, we use a boxcar model function in the correlation analysis. Recently, Kuppusamy et al.20 have shown that a box car will fit using a cross-correlation analysis even if the time course shows deviations from an ideal behavior. Table 1 summarizes our findings of the different activation foci among the subjects. Repeatable (both intra- and inter-subject) patterns were observed in the primary motor cortex, SMA, putamen and cerebellum. Activation in other cortical regions was also seen, more consistently in the superior frontal gyrus and insular cortex. Activation in the globus pallidus, caudate nucleus, and thalamus were also present in some cases, but the results showed high intraand inter-subject variability. Although activation was present, the patterns were small and scattered in multiple loci across the subjects. DISCUSSION The main advantage of multislice fMRI lies in the fact that it is possible to examine, non-invasively and within reasonable acquisition times, functionally related areas with high temporal resolution and sufficient spatial resolution that may have a wide anatomical distribution. It is well known that the motor and somatosensory cortices
284
Magnetic Resonance Imaging ● Volume 16, Number 3, 1998
Fig. 2. Anatomical (upper row) and functional activation images covering the basal ganglia and cerebellum for dominant (middle row) and non-dominant (lower row) thumb movement. The images are displayed in standard radiologic orientation (P - putamen, C - caudate nucleus, Cer - cerebellum).
Fig. 3. Activation of the basal ganglia and cerebellum during dominant thumb movement for two consecutive experiments performed on the same volunteer. The upper row shows the activation obtained during the first experiment, and the lower row during the second experiment. Note the similar location of the response for the two runs, indicating the intra-individual repeatability.
fMRI of basal ganglia and cerebellum ● J.R. REICHENBACH
Fig. 4. Typical time courses of fMRI signal intensity for the dominant thumb task. Data are taken from the activated cortical and subcortical areas in Figs. 2 and 3. The task conditions are shown at the bottom (white bars 5 resting state, black bars 5 activation state).
have strong anatomical connections to the basal ganglia, with the putamen acting as an input nucleus to the anterior and posterior lobes of the cerebellum via the pontine nuclei and the inferior olivary nucleus.22,23
ET AL.
285
To date, most functional experiments related to the motor systems and involving the basal ganglia and cerebellum have been performed using positron-emission tomography (PET).24 –29 Recent studies have demonstrated activation in these structures during motor sequence learning as well as many other tasks ranging from sensorimotor tasks to pure thinking. However, the results have been often contradictory and not conclusive. For example, when normal right-handed subjects performed freely selected joystick movements, Playford et al.30 observed significant activation in the left putamen and bilaterally in the cerebellum and thalamus, whereas Deiber et al.31 did not report any relative cerebral blood flow (rCBF) changes in the basal ganglia by applying the same paradigm. Our results (Figs. 1 and 2, Table 1) demonstrate the feasibility of whole brain coverage fMRI to detect reliably activation in multiple brain areas, such as primary motor cortex, SMA, putamen, cerebellum using a simple self-paced finger movement. These findings are in agreement with other investigations on functional responses for skilled motor movements.24,25,29,32,33 For example, Jenkins et al.29 have shown activation of the left putamen as well as significant activation bilaterally in the cerebellar hemispheres during a prelearned sequence of keypresses and Mazziotta et al.27 reported that simple finger tapping increases the metabolism in both the left basal ganglia and the left motor and premotor areas. In general, we observed larger clusters of activated pixels in the contralateral putamen and ipsilateral cerebellar hemisphere compared to the ipsilateral putamen and contralateral cerebellar area when subjects performed dominant thumb movement (Fig. 2). For the non-dominant task, the evoked response tends to be more bilaterally distributed. Furthermore, moving the nondominant thumb seems to evoke larger areas of activation in both putamen and cerebellum than the dominant thumb. The fact that a somewhat larger activation pattern in the contralateral putamen and a bilateral response in the cerebellum (Fig. 3) was observed in some cases during the second run (compared to the first run) reflects probably the normal intra-individual physiological fluctuations when repeating the task in a self-paced manner. However, it is also possible that the difference between the two runs reflects the effects of practice. In most cases, we observed also activation in the ipsilateral putamen when subjects moved their dominant thumb, but the pattern was not as consistent across the subjects as in the contralateral putamen. A similar result was reported by Kim et al.34 who observed a hemispheric asymmetry in the functional activation of the motor cortex during contralateral and ipsilateral movements, especially in righthanded subjects, at high field strength. However, further and more extensive studies are necessary to confirm the
Magnetic Resonance Imaging ● Volume 16, Number 3, 1998
286
Table 1. Regional activation in individual brain activation maps Dominant Thumb Subjects Hemisphere Cortex Primary Motor Cortex SMA Basal Ganglia Putamen Other Cerebellum
1
2
3
4
Non-Dominant Thumb
5
6
Total (n 5 7)
7
2
5
Total (n 5 3)
7
C
I
C
I
C
I
C
I
C
I
C
I
C
I
C
I
C
I
C
I
C
I
C
I
1 1
2 2
1 1
2 2
1 1
2 1
1 1
1 2
1 1
2 2
1 1
2 2
1 1
2 2
7 7
1 1
2 1
1 1
1 1
2 2
1 1
2 2
2 3
1 1
1
1
1
1
1
2
1
2
1
1
1
1
1
1
7
5
1
1
1
1
1
1
3
3
2
1
1
1
1
1
1
1
2
1
2
1
2
1
4
7
1
1
1
1
2
2
2
2
(1) 5 activation; (2) 5 no activation; C 5 contralateral; I 5 ipsilateral
present observation of a potential asymmetry in response associated with handedness. Ellerman et al.35 also reported very similar results in the cerebellum using fMRI. In their experiments, when subjects performed a series of alternating wrist flexion and extension movements against constant inertial loads, the strength of activation for right wrist movement was greater in the ipsilateral than in the contralateral cerebellum. However, for left hand movements, the activation was predominantly bilateral. In summary, in this preliminary study the ability of fMRI has been demonstrated to detect activation in multiple loci in the brain with a simple motor paradigm. Future applications of volume fMRI covering the entire brain could include patients suffering from movement disorders who could be asked to perform motor tasks to reveal abnormalities in patterns of cerebral activation. Acknowledgments—J.R.R. acknowledges financial support from the Deutsche Forschungsgemeinschaft (DFG, Re 1123/1-2). E.M.H. acknowledges support from Siemens Medical Systems. Special thanks to F.G.C. Hoogenraad for sharing his experience with the motion correction procedure and critical reading of the manuscript.
REFERENCES 1. Thulborn, K.R.; Waterton, J.C.; Matthews, P.M.; Radda, G.K. Oxygenation dependence of the transverse relaxation time of water protons in whole blood at high field. Biochim. Biophys. Acta 714:265–270; 1982. 2. Ogawa, S.; Lee, T.M.; Kay, A.R.; Tank, D.W. Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc. Natl. Acad. Sci. USA 87:9868 –9872; 1990. 3. Turner, R.; Le Bihan, D.; Moonen, C.T.W.; Despres, D.; Frank, J. Echo-planar time course MRI of cat brain oxygenation changes. Magn. Reson. Med. 22:159 –166; 1991.
4. Kwong, K.K. Functional magnetic resonance imaging with echo planar imaging. Magn. Reson. Quart. 11:1–20; 1995. 5. Kim, S.G.; Ugurbil, K.; Strick, P.L. Activation of a cerebellar output nucleus during cognitive processing. Science 265:949 –951; 1994. 6. Rao, S.M.; Binder, J.R.; Bandettini, P.A., et al. Functional magnetic resonance imaging of complex human movements. Neurology 43:2311–2318; 1993. 7. Ogawa, S.; Menon, R.S.; Tank, D.W.; Kim, S.G.; Merkle, H.; Ellermann, J.M.; Ugurbil, K. Functional brain mapping by blood oxygenation level-dependent contrast magnetic resonance imaging: A comparison of signal characteristics with a biophysical model. Biophys. J. 64:803– 812; 1993. 8. Segebarth, C.; Belle, C.; Delon, R.; Massarelli, R.; Decety, J.; Le Bas, J.F.; Decorps, M.; Benabid, A.L. Functional MRI of the human brain: Predominance of signals from extracerebral veins. Neuroreport 5:813– 816; 1994. 9. Haacke, E.M.; Hopkins, A.L.; Lai, S., et al. 2D and 3D high-resolution gradient echo functional imaging of the brain: Venous contributions to signal in motor cortex studies. NMR Biomed. 7:54 – 62; 1994. 10. Haacke, E.M.; Lai, S.; Yablonskiy, D.A.; Lin, W. In vivo validation of the BOLD mechanism. A review of signal changes in gradient echo functional MRI in the presence of flow. Int. J. Imag. Sys. Techn. 6:153–163; 1995. 11. Mattay, V.S.; Frank, J.A.; Santha, A.K.S.; Pekar, J.J.; Duyn, J.H.; McLaughlin, A.C.; Weinberger, D.R. Wholebrain functional mapping with isotropic MR imaging. Radiology 201:399 – 404; 1996. 12. Alexander, G.E.; DeLong, M.R.; Strick, P.L. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Ann. Rev. Neurosci. 9:357–381; 1986. 13. Schell, G.R.; Strick, P.L. The origin of thalamic inputs to the arcuate premotor and supplementary motor areas. J. Neurosci. 4:539 –560; 1984. 14. Brooks, D.J. The role of the basal ganglia in motor control: contributions from PET. J. Neurological Sci. 128:1–13; 1995.
fMRI of basal ganglia and cerebellum ● J.R. REICHENBACH
15. Raymond, J.L.; Lisberger, S.G.; Mauk, M.D. The cerebellum: A neuronal learning machine? Science 272:1126 – 1131; 1996. 16. Bucher, S.F.; Seelos, K.C.; Stehling, M.; Oertel, W.H.; Paulus, W.; Reiser, M. High-resolution activation mapping of basal ganglia with functional magnetic resonance imaging. Neurology 45:180 –182; 1995. 17. Scholz, V.H.; Jenkins, B.G.; Keltner, J.R.; Chen, Y.I.; Kwong, K.K.; Rosen, B.R. Comparison of different motor tasks for fMRI studies of the basal ganglia: Reproducibility, specificity and artifacts. In: Book of Abstracts: Fourth Scientific Meeting of the International Society for Magnetic Resonance in Medicine, Vol. 1. New York: ISMRM; 1996:446. 18. Woods, R.P.; Cherry, S.R.; Mazziota, J.C. Rapid automated algorithm for aligning and reslicing PET images. J. Comp. Assist. Tomogr. 16:620 – 633; 1992. 19. Bandettini, P.A.; Wong, E.C.; Hinks, R.S.; Tikofsky, R.S.; Hyde, J.S. Time course EPI of human brain function during task activation. Magn. Reson. Med. 25:390 –397; 1992. 20. Kuppusamy, K.; Lin, W.; Haacke, E.M. Statistical assessment of crosscorrelation and variance methods and the importance of electrocardiogram in functional magnetic resonance imaging. Magn. Reson. Imaging 15:169 –181; 1997. 21. Kwong, K.K.; Belliveau, J.W.; Chesler, D.A., et al. Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. Proc. Natl. Acad. Sci. USA 89:5675–5679; 1992. 22. Glickstein, M.; May, J.G., III; Mercier, B.E. Corticopontine projection in the macaque: the distribution of labelled cortical cells after large injections of horseradish peroxidase in the pontine nuclei. J. Comp. Neurol. 235:343–359; 1985. 23. Kunzle, H. Bilateral projections from precentral motor cortex to the putamen and other parts of the basal ganglia. An autoradiographic study in macaca fascicularis. Brain Research 88:195–209; 1975. 24. Roland, P.E.; Meyer, E.; Shibasaki, H.; Yamamoto, Y.L. Regional cerebral blood flow changes in cortex and basal ganglia during voluntary movements in normal human volunteers. J. Neurophysiol. 48:467– 480; 1982.
ET AL.
287
25. Fox, P.T.; Fox, J.M.; Raichle, M.E.; Burde, R.M. The role of cerebral cortex in the generation of voluntary saccades: A positron emission tomography study. J. Neurophysiol. 54:348 –369; 1985. 26. Fox, P.T.; Raichle, M.E.; Thach, W.T. Functional mapping of human cerebellum with positron emission tomography. Proc. Natl. Acad. Sci. USA 82:7462–7466; 1985. 27. Mazziotta, J.C.; Phelps, M.E.; Wapenski, J.A. J. Cereb. Blood. Flow Metab. 5:213–214; 1985. 28. Seitz, R.J.; Roland, P.E.; Bohm, C.; Greitz, T.; StoneElander, S. Somatosensory Discrimination of shape: Tactile exploration and cerebral activation. Eur. J. Neurosci. 3:481– 492; 1991. 29. Jenkins, I.H.; Brooks, D.J.; Nixon, P.D.; Frackowiak, R.S.J.; Passingham, R.E. Motor Sequence Learning: A study with positron emission tomography. J. Neurosci. 14:3775–3790; 1994. 30. Playford, E.D.; Jenkins, I.H.; Passingham, R.E.; Nutt, J.; Frackowiak, R.S.J.; Brooks, D.J. Impaired mesial frontal and putamen activation in Parkinson’s disease: A PET study. Ann. Neurol. 32:151–161; 1992. 31. Deiber, M.-P.; Passingham, R.E.; Colebatch, J.G.; Friston, K.F.; Nixon, P.D.; Frackowiak, R.S.J. Cortical areas and the selection of movement: A study with positron emission tomography. Exp. Brain Res. 84:393– 402; 1991. 32. Seitz, R.J.; Roland, P.E.; Bohm, C.; Greitz, T.; StoneElander, S. Motor learning in man: A positron emission tomographic study. NeuroReport 1:57– 66; 1990. 33. Larsson, J.; Gulya´s, B.; Roland, P.E. Cortical representation of self-paced finger movement. NeuroReport 7:463– 468; 1996. 34. Kim, S.G.; Ashe, J.; Hendrich, K.; Ellermann, J.M.; Merkle, H.; Ugurbil, K.; Georgopoulos, A.P. Functional magnetic resonance imaging of motor cortex: Hemispheric asymmetry and handedness. Science 261:615– 617; 1993. 35. Ellermann, J.M.; Flament, D.; Kim, S.-G.; Fu, Q.-G.; Merkle, H.; Ebner, T.J.; Ugurbil, K. Spatial patterns of functional activation of the cerebellum investigated using high field (4 T) MRI. NMR Biomedicine 7:63– 68; 1994.
PII S0730-725X(97)00276-2
● Original Contribution
FUNCTIONAL MAGNETIC RESONANCE IMAGING OF THE BASAL GANGLIA AND CEREBELLUM USING A SIMPLE MOTOR PARADIGM JU¨ RGEN R. REICHENBACH,* ROBERT FEIWELL,* KARTHIKEYAN KUPPUSAMY,*† MARK BAHN,* AND E. MARK HAACKE*† *Mallinckrodt Institute of Radiology, Washington University, School of Medicine, 510 S. Kingshighway, St. Louis, MO, and †Department of Electrical Engineering, Washington University, 1 Brookings Drive, St. Louis, MO, USA Activation of cortical and subcortical motor areas of the brain, including primary motor cortex, supplementary motor area, basal ganglia and cerebellum, were successfully investigated in seven right-handed, normal volunteers during a simple, rapid, thumb flexion-extension task using functional magnetic resonance imaging. A multi-slice echo-planar imaging sequence was used to cover the entire brain. A signal increase varying from 2% to 6% was observed for the different regions involved in the motor task. Moving the non-dominant thumb was associated with a more bilateral activation pattern in both putamen and cerebellar regions. This study demonstrates the capability of functional magnetic resonance imaging to delineate simultaneously many activated brain areas that are commonly thought to be involved in the performance of motor tasks. © 1998 Elsevier Science Inc. Keywords: Functional magnetic resonance imaging (fMRI); Motor cortex stimulation; Basal ganglia; Cerebellum; Self-paced movement.
biologic studies that, for example, the control of voluntary and even simple movements is a complex process that involves many different areas of the brain. It is widely believed that motor system control is achieved by a series of parallel systems formed by somatotopically organized, descending projections that link the various motor-related areas of the cortex more directly with spinal motor circuits.12–14 An important role is hereby played by subcortical structures, such as the basal ganglia and the cerebellum. The basal ganglia is a large, functionally heterogeneous structure arranged mainly in multiple parallel cortico-striato-thalamo-cortical circuits and involved in a wide variety of motor and affective behaviors, in sensorimotor integration, and in cognitive functions. With respect to motor behavior, the basal ganglia are believed to be involved in the determination of movement parameters, preparation for movement, enabling movement to become automatic, facilitation of sequential move-
INTRODUCTION In recent years, functional magnetic resonance imaging (fMRI) has become a versatile and important clinical as well as research tool to study non-invasively activation of the normal and diseased human brain. Since the observation that relaxation times depend on the blood oxygenation level1 (BOLD-contrast) and that changes of the latter can be measured in vivo,2 the possibility to examine cortical activity using fMRI methods has found widespread interest.3–10 Until recently, application of fMRI to map brain activation was restricted in practice to a limited number of slices covering only parts of the brain. With the advent of scanners having echo-planar imaging (EPI) capability with powerful and very rapid gradient systems, it has become possible to cover the whole brain using singleshot, multi-slice EPI acquisitions.11 This possibility has an important impact for a better understanding of functional neuroanatomy since it is well known from neuroRECEIVED 4/9/97; ACCEPTED 10/11/97. Address correspondence to Ju¨rgen R. Reichenbach, PhD, Institute of Diagnostic and Interventional Radiology, FriedrichSchiller-University Jena, Bachstrasse 18, 07740 Jena, Ger-
many. E-mail: [email protected] *Present address: Institute of Diagnostic and Interventional Radiology, Friedrich-Schiller-University, Jena, Germany 281
282
Magnetic Resonance Imaging ● Volume 16, Number 3, 1998
ment, inhibition of unwanted movements, adaptation to novel circumstances, facilitation of rewarded action, as well as motor learning and planning.14 The basal ganglia are usually taken to include the caudate nucleus and putamen (often jointly termed striatum), the globus pallidus (or pallidum), the subthalamic nucleus, and the substantia nigra. A motor control loop has been characterized involving the supplementary motor area (SMA), primary motor cortex, putamen, pallidum and ventrolateral thalamus.13,14 Damage of the basal ganglia can cause motion disorders, such as involuntary movements, muscular rigidity, and immobility without paralysis. The cerebellum plays a key role in movement and is considered the primary site of motor learning. It contributes primarily to motor coordination and control and receives input from virtually all brain areas. It is also involved in the control of posture, regulation of bodily function in response to a variety of stimuli, initiation of limb movements, adjustment of eye movements in response to hand movement, and fine manipulative movements. The cerebellum is one of the major sources of input, via the thalamus, to the primary motor cortex (area 4) and premotor cortex (lateral portion of area 6). However, its exact function in the timing of movement and motor learning is still unclear and controversial.15 The use of volume coverage fMRI using EPI offers a practical approach to more global studies of brain function because, based on the neuroanatomic knowledge, it can be expected that performing a motor task does not only activate areas in the motor cortex, but should also activate parts of the basal ganglia and the cerebellum. Whole-brain methods are thus essential to map simultaneously complex activation patterns that are spread across different areas. To date there have been limited activation studies of the basal ganglia using functional magnetic resonance imaging and these have been only local in nature.16,17 Recently, Bucher et al.16 have shown localized activation within the putamen and globus pallidus using high resolution, FLASH (fast low-angle shot) fMRI. Limitations of that study include prolonged imaging time and the limited number of slices acquired. The aim of this study was to examine the feasibility and reproducibility of visualizing simultaneous activation of motor cortex, SMA, basal ganglia and cerebellum during the performance of a simple, skilled, motor-specific task with a T2*-weighted, interleaved, echo-planar fMRI technique. This demonstrates the capability of fMRI in assessing global involvement of a task over the entire brain. MATERIALS AND METHODS Seven normal, right-handed, healthy subjects (six male, one female, mean age 31 years) participated in this
pilot study. Informed written consent was obtained from all subjects after the nature of the experiment had been fully explained and was approved by the Institutional Review Board and the Human Studies Committee. The activation state was a rapid (2–3 Hz), self-paced, flexionextension movement of the thumb (digit 1) of the dominant hand. Subjects had a practice session of about 15 min prior to the scans to ensure that they performed the task as consistently and rapidly as possible. They were also instructed to keep their eyes closed during the experiments. To investigate possible differences in the activation pattern for dominant vs. non-dominant exercise, three subjects were also imaged during non-dominant thumb movement. During acquisition of 50 images/slice, resting and active periods were alternated in an ‘‘off-on’’ manner for five cycles. Each cycle consisted of five resting and five activation images. A buzzer was used as an auditory trigger to indicate the onset of each ‘‘off’’ and ‘‘on’’ mode, respectively. The paradigm was repeated three times within the same scan session to investigate intra-subject variability. Subjects were placed supine on the MRI table. Foam cushions and pads were used to comfortably immobilize the head in the circular polarized head coil and ear plugs were used for noise reduction. Prior to imaging, shimming was performed to optimize the static field homogeneity. Usually, a gradient offset tolerance better than 0.001 mT/m was obtained. For orientation, sagittal anatomical scout measurements were performed. Based on these images and using the anterior commissure-posterior commissure (AC-PC) line as a rough neuroanatomic landmark, two-dimensional transaxial single-shot, multislice, echo-planar images covering the entire brain and cerebellum were acquired on a whole-body, 1.5 T scanner (Vision system, Siemens Medical Systems, Erlangen, Germany) with the following sequence parameters: repetition time 5 5 s, echo time 5 87 ms, flip angle a 5 90°, slice thickness TH 5 5 mm, 10 or 20 slices, matrix size 5 96 3 128 (sinc interpolated to 256 3 256), FOV 5 200 –230 mm. Corresponding T1-weighted images were collected with exactly the same orientation at the same slice locations to compare the anatomy with functional activation patterns using a three-dimensional magnetization-prepared rapid gradient echo (MP-RAGE) sequence. The imaging parameters were: repetition time 5 9.7 ms, echo time 5 4 ms, a 5 8°, slice thickness TH 5 2.5 mm, 80 partitions, matrix size 5 256 3 256. The field of view (FOV) was the same as for the functional studies. In this way, the EPI experiments could be anatomically correlated with the conventional T1weighted scans very easily. The subjects were visually monitored during the examination to confirm adherence to the experimental design. Intrasubject registration of the echo-planar data was
fMRI of basal ganglia and cerebellum ● J.R. REICHENBACH
Fig. 1. Overlay images showing areas of activation during dominant thumb movement in the primary motor cortex area and SMA. The maps are based on a correlation coefficient threshold of 0.3 (p , 0.02) and only show areas consisting of more than 20 clustered activated pixels. Pixels with correlation coefficients higher than the threshold are displayed in yellow. The images are displayed in standard radiologic orientation (PMC - primary motor cortex, SMA - supplementary motor area).
performed using the AIR 3.0 realignment software package developed by Roger Woods.18 The functional time courses were analyzed with the cross-correlation method and a box car reference waveform with the same period as the exercise cycles.19,20 The functional maps were generated pixelwise using a correlation coefficient threshold of 0.30 (p , 0.02, paired, two-tailed t-test) and activated pixels were clustered to at least 20 pixels to be considered significant with a high spatial consistency. These pixels were overlayed onto the EPI-images. The large cluster size is a very conservative means to increase the statistical significance of the observed activation. It was also chosen due to the fact that the EPI-images were sinc-interpolated from a low-resolution data set. RESULTS Clusters of activated pixels were found in all subjects. As an example, the functional images in Fig. 1 demonstrate the activation seen in consecutive slices containing the primary motor cortex and the SMA, while the subject performed the motor task with the dominant (right) hand. The response is clearly localized on the contralateral side with only minor activation on the ipsilateral hemisphere and can be followed throughout several slices. Ipsilateral activity, however, was not consistently observed across different subjects. Figure 2 displays the anatomical and functional images obtained in the basal ganglia and the
ET AL.
283
cerebellum for dominant and non-dominant thumb movement in one right-handed volunteer. For better illustration, anatomical and functional images were put side by side. For this volunteer, dominant thumb movement (second panel) resulted in activation of the contralateral putamen. In contrast, non-dominant thumb movement (third panel) yielded increased ipsilateral activity in association with the expected strong contralateral activation. The cerebellum shows ipsilateral, anterior hemispheric activation for both left and right thumb movement. However, with non-dominant motor activity the response is more bilaterally distributed. Figure 3 demonstrates the intra-subject repeatability of the observed functional activation. The volunteer repeated the paradigm during the same scan session. As can be clearly seen, the response in the second experiment (lower row) matches reasonably well the result obtained during the first experiment (upper row). Plots of fMRI time courses from regions of activated pixels in the primary motor cortex, supplementary motor area (SMA), putamen, and cerebellum are shown in Fig. 4. Typical signal increases of approximately 6% in the contralateral motor cortex, 2–3% in the contralateral putamen, and 4 –5% in the ipsilateral cerebellum were obtained, all of which are in agreement with the results of other groups19 –21 and the expected range from the BOLD effect at 1.5 Tesla.7 As can be seen from the time courses the signal in the activated states stays reasonably flat, especially in the cases of the motor cortex and cerebellum. Given this behavior, we use a boxcar model function in the correlation analysis. Recently, Kuppusamy et al.20 have shown that a box car will fit using a cross-correlation analysis even if the time course shows deviations from an ideal behavior. Table 1 summarizes our findings of the different activation foci among the subjects. Repeatable (both intra- and inter-subject) patterns were observed in the primary motor cortex, SMA, putamen and cerebellum. Activation in other cortical regions was also seen, more consistently in the superior frontal gyrus and insular cortex. Activation in the globus pallidus, caudate nucleus, and thalamus were also present in some cases, but the results showed high intraand inter-subject variability. Although activation was present, the patterns were small and scattered in multiple loci across the subjects. DISCUSSION The main advantage of multislice fMRI lies in the fact that it is possible to examine, non-invasively and within reasonable acquisition times, functionally related areas with high temporal resolution and sufficient spatial resolution that may have a wide anatomical distribution. It is well known that the motor and somatosensory cortices
284
Magnetic Resonance Imaging ● Volume 16, Number 3, 1998
Fig. 2. Anatomical (upper row) and functional activation images covering the basal ganglia and cerebellum for dominant (middle row) and non-dominant (lower row) thumb movement. The images are displayed in standard radiologic orientation (P - putamen, C - caudate nucleus, Cer - cerebellum).
Fig. 3. Activation of the basal ganglia and cerebellum during dominant thumb movement for two consecutive experiments performed on the same volunteer. The upper row shows the activation obtained during the first experiment, and the lower row during the second experiment. Note the similar location of the response for the two runs, indicating the intra-individual repeatability.
fMRI of basal ganglia and cerebellum ● J.R. REICHENBACH
Fig. 4. Typical time courses of fMRI signal intensity for the dominant thumb task. Data are taken from the activated cortical and subcortical areas in Figs. 2 and 3. The task conditions are shown at the bottom (white bars 5 resting state, black bars 5 activation state).
have strong anatomical connections to the basal ganglia, with the putamen acting as an input nucleus to the anterior and posterior lobes of the cerebellum via the pontine nuclei and the inferior olivary nucleus.22,23
ET AL.
285
To date, most functional experiments related to the motor systems and involving the basal ganglia and cerebellum have been performed using positron-emission tomography (PET).24 –29 Recent studies have demonstrated activation in these structures during motor sequence learning as well as many other tasks ranging from sensorimotor tasks to pure thinking. However, the results have been often contradictory and not conclusive. For example, when normal right-handed subjects performed freely selected joystick movements, Playford et al.30 observed significant activation in the left putamen and bilaterally in the cerebellum and thalamus, whereas Deiber et al.31 did not report any relative cerebral blood flow (rCBF) changes in the basal ganglia by applying the same paradigm. Our results (Figs. 1 and 2, Table 1) demonstrate the feasibility of whole brain coverage fMRI to detect reliably activation in multiple brain areas, such as primary motor cortex, SMA, putamen, cerebellum using a simple self-paced finger movement. These findings are in agreement with other investigations on functional responses for skilled motor movements.24,25,29,32,33 For example, Jenkins et al.29 have shown activation of the left putamen as well as significant activation bilaterally in the cerebellar hemispheres during a prelearned sequence of keypresses and Mazziotta et al.27 reported that simple finger tapping increases the metabolism in both the left basal ganglia and the left motor and premotor areas. In general, we observed larger clusters of activated pixels in the contralateral putamen and ipsilateral cerebellar hemisphere compared to the ipsilateral putamen and contralateral cerebellar area when subjects performed dominant thumb movement (Fig. 2). For the non-dominant task, the evoked response tends to be more bilaterally distributed. Furthermore, moving the nondominant thumb seems to evoke larger areas of activation in both putamen and cerebellum than the dominant thumb. The fact that a somewhat larger activation pattern in the contralateral putamen and a bilateral response in the cerebellum (Fig. 3) was observed in some cases during the second run (compared to the first run) reflects probably the normal intra-individual physiological fluctuations when repeating the task in a self-paced manner. However, it is also possible that the difference between the two runs reflects the effects of practice. In most cases, we observed also activation in the ipsilateral putamen when subjects moved their dominant thumb, but the pattern was not as consistent across the subjects as in the contralateral putamen. A similar result was reported by Kim et al.34 who observed a hemispheric asymmetry in the functional activation of the motor cortex during contralateral and ipsilateral movements, especially in righthanded subjects, at high field strength. However, further and more extensive studies are necessary to confirm the
Magnetic Resonance Imaging ● Volume 16, Number 3, 1998
286
Table 1. Regional activation in individual brain activation maps Dominant Thumb Subjects Hemisphere Cortex Primary Motor Cortex SMA Basal Ganglia Putamen Other Cerebellum
1
2
3
4
Non-Dominant Thumb
5
6
Total (n 5 7)
7
2
5
Total (n 5 3)
7
C
I
C
I
C
I
C
I
C
I
C
I
C
I
C
I
C
I
C
I
C
I
C
I
1 1
2 2
1 1
2 2
1 1
2 1
1 1
1 2
1 1
2 2
1 1
2 2
1 1
2 2
7 7
1 1
2 1
1 1
1 1
2 2
1 1
2 2
2 3
1 1
1
1
1
1
1
2
1
2
1
1
1
1
1
1
7
5
1
1
1
1
1
1
3
3
2
1
1
1
1
1
1
1
2
1
2
1
2
1
4
7
1
1
1
1
2
2
2
2
(1) 5 activation; (2) 5 no activation; C 5 contralateral; I 5 ipsilateral
present observation of a potential asymmetry in response associated with handedness. Ellerman et al.35 also reported very similar results in the cerebellum using fMRI. In their experiments, when subjects performed a series of alternating wrist flexion and extension movements against constant inertial loads, the strength of activation for right wrist movement was greater in the ipsilateral than in the contralateral cerebellum. However, for left hand movements, the activation was predominantly bilateral. In summary, in this preliminary study the ability of fMRI has been demonstrated to detect activation in multiple loci in the brain with a simple motor paradigm. Future applications of volume fMRI covering the entire brain could include patients suffering from movement disorders who could be asked to perform motor tasks to reveal abnormalities in patterns of cerebral activation. Acknowledgments—J.R.R. acknowledges financial support from the Deutsche Forschungsgemeinschaft (DFG, Re 1123/1-2). E.M.H. acknowledges support from Siemens Medical Systems. Special thanks to F.G.C. Hoogenraad for sharing his experience with the motion correction procedure and critical reading of the manuscript.
REFERENCES 1. Thulborn, K.R.; Waterton, J.C.; Matthews, P.M.; Radda, G.K. Oxygenation dependence of the transverse relaxation time of water protons in whole blood at high field. Biochim. Biophys. Acta 714:265–270; 1982. 2. Ogawa, S.; Lee, T.M.; Kay, A.R.; Tank, D.W. Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc. Natl. Acad. Sci. USA 87:9868 –9872; 1990. 3. Turner, R.; Le Bihan, D.; Moonen, C.T.W.; Despres, D.; Frank, J. Echo-planar time course MRI of cat brain oxygenation changes. Magn. Reson. Med. 22:159 –166; 1991.
4. Kwong, K.K. Functional magnetic resonance imaging with echo planar imaging. Magn. Reson. Quart. 11:1–20; 1995. 5. Kim, S.G.; Ugurbil, K.; Strick, P.L. Activation of a cerebellar output nucleus during cognitive processing. Science 265:949 –951; 1994. 6. Rao, S.M.; Binder, J.R.; Bandettini, P.A., et al. Functional magnetic resonance imaging of complex human movements. Neurology 43:2311–2318; 1993. 7. Ogawa, S.; Menon, R.S.; Tank, D.W.; Kim, S.G.; Merkle, H.; Ellermann, J.M.; Ugurbil, K. Functional brain mapping by blood oxygenation level-dependent contrast magnetic resonance imaging: A comparison of signal characteristics with a biophysical model. Biophys. J. 64:803– 812; 1993. 8. Segebarth, C.; Belle, C.; Delon, R.; Massarelli, R.; Decety, J.; Le Bas, J.F.; Decorps, M.; Benabid, A.L. Functional MRI of the human brain: Predominance of signals from extracerebral veins. Neuroreport 5:813– 816; 1994. 9. Haacke, E.M.; Hopkins, A.L.; Lai, S., et al. 2D and 3D high-resolution gradient echo functional imaging of the brain: Venous contributions to signal in motor cortex studies. NMR Biomed. 7:54 – 62; 1994. 10. Haacke, E.M.; Lai, S.; Yablonskiy, D.A.; Lin, W. In vivo validation of the BOLD mechanism. A review of signal changes in gradient echo functional MRI in the presence of flow. Int. J. Imag. Sys. Techn. 6:153–163; 1995. 11. Mattay, V.S.; Frank, J.A.; Santha, A.K.S.; Pekar, J.J.; Duyn, J.H.; McLaughlin, A.C.; Weinberger, D.R. Wholebrain functional mapping with isotropic MR imaging. Radiology 201:399 – 404; 1996. 12. Alexander, G.E.; DeLong, M.R.; Strick, P.L. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Ann. Rev. Neurosci. 9:357–381; 1986. 13. Schell, G.R.; Strick, P.L. The origin of thalamic inputs to the arcuate premotor and supplementary motor areas. J. Neurosci. 4:539 –560; 1984. 14. Brooks, D.J. The role of the basal ganglia in motor control: contributions from PET. J. Neurological Sci. 128:1–13; 1995.
fMRI of basal ganglia and cerebellum ● J.R. REICHENBACH
15. Raymond, J.L.; Lisberger, S.G.; Mauk, M.D. The cerebellum: A neuronal learning machine? Science 272:1126 – 1131; 1996. 16. Bucher, S.F.; Seelos, K.C.; Stehling, M.; Oertel, W.H.; Paulus, W.; Reiser, M. High-resolution activation mapping of basal ganglia with functional magnetic resonance imaging. Neurology 45:180 –182; 1995. 17. Scholz, V.H.; Jenkins, B.G.; Keltner, J.R.; Chen, Y.I.; Kwong, K.K.; Rosen, B.R. Comparison of different motor tasks for fMRI studies of the basal ganglia: Reproducibility, specificity and artifacts. In: Book of Abstracts: Fourth Scientific Meeting of the International Society for Magnetic Resonance in Medicine, Vol. 1. New York: ISMRM; 1996:446. 18. Woods, R.P.; Cherry, S.R.; Mazziota, J.C. Rapid automated algorithm for aligning and reslicing PET images. J. Comp. Assist. Tomogr. 16:620 – 633; 1992. 19. Bandettini, P.A.; Wong, E.C.; Hinks, R.S.; Tikofsky, R.S.; Hyde, J.S. Time course EPI of human brain function during task activation. Magn. Reson. Med. 25:390 –397; 1992. 20. Kuppusamy, K.; Lin, W.; Haacke, E.M. Statistical assessment of crosscorrelation and variance methods and the importance of electrocardiogram in functional magnetic resonance imaging. Magn. Reson. Imaging 15:169 –181; 1997. 21. Kwong, K.K.; Belliveau, J.W.; Chesler, D.A., et al. Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. Proc. Natl. Acad. Sci. USA 89:5675–5679; 1992. 22. Glickstein, M.; May, J.G., III; Mercier, B.E. Corticopontine projection in the macaque: the distribution of labelled cortical cells after large injections of horseradish peroxidase in the pontine nuclei. J. Comp. Neurol. 235:343–359; 1985. 23. Kunzle, H. Bilateral projections from precentral motor cortex to the putamen and other parts of the basal ganglia. An autoradiographic study in macaca fascicularis. Brain Research 88:195–209; 1975. 24. Roland, P.E.; Meyer, E.; Shibasaki, H.; Yamamoto, Y.L. Regional cerebral blood flow changes in cortex and basal ganglia during voluntary movements in normal human volunteers. J. Neurophysiol. 48:467– 480; 1982.
ET AL.
287
25. Fox, P.T.; Fox, J.M.; Raichle, M.E.; Burde, R.M. The role of cerebral cortex in the generation of voluntary saccades: A positron emission tomography study. J. Neurophysiol. 54:348 –369; 1985. 26. Fox, P.T.; Raichle, M.E.; Thach, W.T. Functional mapping of human cerebellum with positron emission tomography. Proc. Natl. Acad. Sci. USA 82:7462–7466; 1985. 27. Mazziotta, J.C.; Phelps, M.E.; Wapenski, J.A. J. Cereb. Blood. Flow Metab. 5:213–214; 1985. 28. Seitz, R.J.; Roland, P.E.; Bohm, C.; Greitz, T.; StoneElander, S. Somatosensory Discrimination of shape: Tactile exploration and cerebral activation. Eur. J. Neurosci. 3:481– 492; 1991. 29. Jenkins, I.H.; Brooks, D.J.; Nixon, P.D.; Frackowiak, R.S.J.; Passingham, R.E. Motor Sequence Learning: A study with positron emission tomography. J. Neurosci. 14:3775–3790; 1994. 30. Playford, E.D.; Jenkins, I.H.; Passingham, R.E.; Nutt, J.; Frackowiak, R.S.J.; Brooks, D.J. Impaired mesial frontal and putamen activation in Parkinson’s disease: A PET study. Ann. Neurol. 32:151–161; 1992. 31. Deiber, M.-P.; Passingham, R.E.; Colebatch, J.G.; Friston, K.F.; Nixon, P.D.; Frackowiak, R.S.J. Cortical areas and the selection of movement: A study with positron emission tomography. Exp. Brain Res. 84:393– 402; 1991. 32. Seitz, R.J.; Roland, P.E.; Bohm, C.; Greitz, T.; StoneElander, S. Motor learning in man: A positron emission tomographic study. NeuroReport 1:57– 66; 1990. 33. Larsson, J.; Gulya´s, B.; Roland, P.E. Cortical representation of self-paced finger movement. NeuroReport 7:463– 468; 1996. 34. Kim, S.G.; Ashe, J.; Hendrich, K.; Ellermann, J.M.; Merkle, H.; Ugurbil, K.; Georgopoulos, A.P. Functional magnetic resonance imaging of motor cortex: Hemispheric asymmetry and handedness. Science 261:615– 617; 1993. 35. Ellermann, J.M.; Flament, D.; Kim, S.-G.; Fu, Q.-G.; Merkle, H.; Ebner, T.J.; Ugurbil, K. Spatial patterns of functional activation of the cerebellum investigated using high field (4 T) MRI. NMR Biomedicine 7:63– 68; 1994.