- Email: [email protected]
Effect of visual experience on structural organization of the human brain: A voxel based morphometric study using DARTEL
European Journal of Radiology 81 (2012) 2811–2819
Contents lists available at SciVerse ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Effect of visual experience on structural organization of the human brain: A voxel based morphometric study using DARTEL Shilpi Modi 1 , Manisha Bhattacharya 1 , Namita Singh 1 , Rajendra Prasad Tripathi 1 , Subash Khushu ∗ NMR Research Centre, Institute of Nuclear Medicine and Allied Sciences (DRDO), Lucknow Road, Timarpur, Delhi, India
a r t i c l e
i n f o
Article history: Received 13 September 2011 Received in revised form 18 October 2011 Accepted 25 October 2011 Keywords: Voxel-Based Morphometry (VBM) DARTEL Blindness Atrophy Structural reorganization
a b s t r a c t Objective: To investigate structural reorganization in the brain with differential visual experience using Voxel-Based Morphometry with Diffeomorphic Anatomic Registration Through Exponentiated Lie algebra algorithm (DARTEL) approach. Materials and methods: High resolution structural MR images were taken in fifteen normal sighted healthy controls, thirteen totally blind subjects and six partial blind subjects. The analysis was carried out using SPM8 software on MATLAB 7.6.0 platform. Results: VBM study revealed gray matter volume atrophy in the cerebellum and left inferior parietal cortex in total blind subjects and in left inferior parietal cortex, right caudate nucleus, and left primary visual cortex in partial blind subjects as compared to controls. White matter volume loss was found in calcarine gyrus in total blind subjects and Thlamus-somatosensory region in partially blind subjects as compared to controls. Besides, an increase in Gray Matter volume was also found in left middle occipital and middle frontal gyrus and right entorhinal cortex, and an increase in White Matter volume was found in superior frontal gyrus, left middle temporal gyrus and right Heschl’s gyrus in totally blind subjects as compared to controls. Comparison between total and partial blind subjects revealed a greater Gray Matter volume in left cerebellum of partial blinds and left Brodmann area 18 of total blind subjects. Conclusion: Results suggest that, loss of vision at an early age can induce significant structural reorganization on account of the loss of visual input. These plastic changes are different in early onset of total blindness as compared to partial blindness. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Voxel-Based Morphometry (VBM), introduced by Ashburner and Friston offers rapid unbiased assessment of the whole brain [1]. This automated technique examines differences in local composition of tissue on a voxel-by-voxel basis after global differences in anatomy have been discounted. VBM has been used to characterize structural brain differences in a variety of disease conditions including Down’s syndrome, autism, schizophrenia, epilepsy, Alzheimer’s disease, Parkinsonism, type II diabetes mellitus [2], etc. VBM studies have also investigated the impact of learning and practice on brain structure of the healthy subjects [3]. These morphological studies clearly give evidence of morphological plasticity of the brain. Moreover, the morphology and the function seem to be inter-related. Morphological studies on diseased conditions have shown that
∗ Corresponding author. Tel.: +91 11 23905313/5336; fax: +91 11 23919509. E-mail addresses: modi [email protected] (S. Modi), [email protected] (M. Bhattacharya), [email protected] (N. Singh), [email protected] (R.P. Tripathi), [email protected] (S. Khushu). 1 Tel.: +91 11 23905336; fax: +91 11 23919509. 0720-048X/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2011.10.022
altered function is often associated with brain atrophy in the related areas [2]. On the other hand, training or acquisition of new skills often results in an increase in gray matter volume in the relevant regions [3]. Neuroimaging studies have demonstrated that the visual cortex of both the early and late blind subjects exhibits significant cross-modal functional plasticity [4]. However, fewer studies have dealt with structural plasticity associated with blindness using automated techniques like Voxel or Tensor Based Morphometry [4–8]. As the disuse related and compensatory mechanisms, both govern the structural plasticity in the brain, we hypothesized that the morphological changes would be different in totally blind subjects who had no vision left in them, as compared to the morphological changes in partial blind subjects who had partial vision left. In the present study we attempted to detect gray matter changes in the brain, with the extent of sightedness a person has, using Voxel Based Morphometry technique applied to three groups consisting of totally blind subjects, partially blind subjects and controls. To improve inter-subject registration of the MRI images, we applied Diffeomorphic Anatomic Registration Through Exponentiated Lie algebra algorithm (DARTEL) [9], which has been found to optimize the sensitivity of such analyses [10] by using
2812
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
Table 1 Clinical characteristics of the blind subjects. Subject IDa
Age (years)
Onset of blindness
Cause of blindness
Light sense
T1 T2
21 23
Birth 3.0
None None
T3 T4 T5 T6 T7 T8 T9
23 25 24 21 20 23 25
5.0 years Birth Birth 4.0 years Birth Birth 6.0 years
T10 T11 T12 T13 P1 P2 P3 P4 P5 P6
21 23 20 24 21 20 23 27 20 25
6.0 years Birth Birth 4.0 years 5.0 years Birth Birth Birth 5.0 years Birth
Congenital cataract Poor medical aid after conjunctivitis Eye infection Optic nerve atrophy Glaucoma Post iridocyclitis Congenital cataract Post iridocyclitis Poor medication when affected with smallpox Ophthalmitis Congenital cataract Optic nerve atrophy Ophthalmitis Congenital cataract Optic nerve atrophy Congenital cataract Congenital cataract Congenital cataract Post iridocyclitis
a
Table 2 Normalized gray matter volume (nGM), normalized white matter volume (nWM), normalized cerebrospinal fluid volume (nCSF) and total intracranial volume (TIV) for the three groups (mean ± SD).
Controls Total blinds Partial blinds
None Bright light None None Bright light Bright light None
TIV (ml)*
nGM*
nWM*
nCSF*
1445.77 ± 112.68 1403.66 ± 84.62 1358.68 ± 0.09
0.45 ± 0.01 0.45 ± 0.01 0.46 ± 0.01
0.34 ± 0.01 0.33 ± 0.01 0.33 ± 0.01
0.21 ± 0.01 0.22 ± 0.01 0.21 ± 0.01
* TIV, nGM, nWM and nCSF showed no significant difference between the three groups (p > 0.05).
age ± SD = 22.42 ± 1.78 years, group 2) and six right handed early partial blind males (20–25 years old, mean age ± SD = 22.86 ± 2.67 years, group 3) participated in the study. No statistically significant age difference was found between the three groups (p < 0.134 for groups 1 and 2; p < 0.222 for groups 1 and 3; p < 0.974 for groups 2 and 3). All the subjects chosen for the study were Indian natives and none of them had any clinical evidences of stroke, head injury, cardiovascular diseases, history of drug dependence, psychiatric disorder or cognitive impairment nor did they have any cortical infarctions on the T2 -weighted MR images. The reported onset of total or partial blindness for all the subjects ranged from birth to 6 years of age (Table 1). In ophthalmologic examinations, five of the totally blind subjects were found to be sensitive only to strong sunlight that too only from one of their eyes. The maximum distance to which partial blind subjects could perform finger counting ranged from ‘close to face’ to 2 m. Both the partially blind and totally blind subjects were students, undergoing either graduate or post graduate courses in various subjects of Arts (Political Science, History, Hindi, Programming, Sanskrit, Education, English or Buddhist Study). Further, all subjects gave their consent to participate in the study.
Bright light None Bright light None Yes Yes Yes Yes Yes Yes
T – total blind subjects; P – partial blind subjects.
the Levenberg–Marquardt strategy as compared to standard VBM. The current study on one hand explores the morphological changes using DARTEL, which to the best of our knowledge has never been applied in blind subjects, as compared to previous studies using other methods viz. optimized VBM [5], fast fluid registration [6], analysis of cortical thickness and surface area [8], etc. Secondly, it also compares the morphological changes in early total and partial blinds.
2.2. Scanning protocol 2. Materials and methods The MRI scans were acquired using 1.5 Tesla whole-body MRI system (Siemens Magnetom Vision, Erlangen, Germany) with a circularly polarized head coil and 25 mT/m actively shielded gradient system. T1 weighted 3D Magnetization Prepared Rapid Acquisition Gradient Echo (MPRAGE) sequence with 160 slices (1 mm slice thickness) covering the whole brain was performed in
2.1. Subjects Fifteen normal sighted right handed healthy males (21–27 years old, mean age ± SD = 23.61 ± 2.10 years, group 1), thirteen right handed early totally blind males (20–25 years old, mean Table 3 Regions of GM volume changes. Hemisphere
Localization of peak voxels
Controls minus Total blinds* L Lobule VIIIa (Hem), Cerebellum Lobule VIIa Crus II (Hem), Cerebellum R IPC (PF), Inferior Parietal Lobule L Controls minus Partial blinds# L IPC (PFm), Middle Temporal Gyrus Caudate Nucleus R BA 17, Cuneus L BA 17 L Partial blinds minus Total blinds@ L Lobule VIIIa (Hem) L Lobule VIIa Crus II (Hem) Total blinds minus Controls$ L Middle Occipital Gyrus (hOC5 (V5)) BA 6, Middle Frontal Gyrus L Hippocampus (EC) R Total blinds minus Partial blinds! L BA 18 MNI: Montreal Neurological Institute; BA: Brodmann area. * Small Volume Correction (FWE corrected), p < 0.012. # Small Volume Correction (FWE corrected), p < 0.011. @ Small Volume Correction (FWE corrected), p < 0.002. $ Small Volume Correction (FWE corrected), p < 0.01. ! Small Volume Correction (FWE corrected), p < 0.01.
MNI coordinates
Z value (peak voxel)
Cluster size (no. of voxels)
−27 47 −57
−58 −65 −39
−63 −57 43
4.26 3.98 3.37
4836 2898 369
−62 12 −12 −21
−54 15 −79 −64
19 16 16 0
3.58 3.67 3.75 3.95
473 319 302 235
−24 −42
−66 −49
−60 −51
3.49 3.48
460 301
−42 −50 20
−79 51 3
1 6 −35
4.36 4.32 3.28
1034 451 157
−5
−81
28
3.66
354
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
2813
Fig. 1. (a) Three dimensional rendered view and (b) overlay of the peak cluster on the gray matter template provided in SPM8, of regions of GM volume loss in (i) total blind subjects as compared to controls; (ii) partial blind subjects as compared to controls; (iii) total blind subjects as compared to partial blind subjects and regions of GM volume increase in (iv) total blind subjects as compared to controls; (v) total blind subjects as compared to partial blind subjects.
the sagittal plane with TR/TE/TI of 9.7 ms/4 ms/300 ms, flip angle of 12◦ , FOV = 256 mm × 256 mm, number of acquisitions = 1 and an in-plane resolution of 1 mm × 1 mm. All subjects also underwent an axial double spin echo sequence (TE1 = 20 ms, TE2 = 90 ms), which was used to exclude subjects with possible cortical infarctions.
2.3. Data processing Data were processed using Statistical Parametric Mapping (SPM8, Wellcome Department of Cognitive Neurology) implemented in MATLAB R2008a (version 7.6.0 (Mathworks, Sherborn, MA)). To improve the registration of the MRI images,
2814
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
Fig. 1.
DARTEL toolbox for SPM8 was used. All the steps were carried out as suggested by Ashburner, 2010 [www.fil.ion.ucl.ac.uk/∼john/misc/VBMclass10.pdf]. In brief, the anatomic images were first manually reoriented so that the mm coordinate of the anterior commissure matched the origin (0, 0, 0), and the orientation approximated Montreal Neurological Institute (MNI) space. Next, T1 -weighted images were classified into gray matter, white matter and cerebrospinal fluid (CSF) using the ‘new-segment’ routine implemented in SPM8, that gives both the native space versions and DARTEL imported versions
of the tissues. The DARTEL imported versions of gray and white matter were used to generate the flow fields (which encode the shapes), and a series of template images by running ‘DARTEL (create templates)’ routine. During this step, DARTEL increases the accuracy of inter-subject alignment by modeling the shape of each brain using millions of parameters (three parameters for each voxel). DARTEL works by aligning gray matter among the images, while simultaneously aligning white matter. This is achieved by generating increasingly crisp average template data, to which the data are iteratively aligned. The flow fields
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
2815
Fig. 2. (a) Three dimensional rendered view and (b) overlay of the peak cluster on the white matter template provided in SPM8, of regions WM volume loss in (i) total blind subjects as compared to controls; (ii) partial blind subjects as compared to controls and regions of WM volume increase in (iii) total blind subjects as compared to controls; (iv) partial blind subjects as compared to controls.
and the final template image created in the previous step are used to generate smoothed (10 mm FWHM), modulated, spatially normalized and Jacobian scaled gray and white matter images resliced to 1.5 mm × 1.5 mm × 1.5 mm voxel size, in MNI space. Total gray matter (GM) volume, white matter (WM) volume, CSF volume and Total Intracranial Volume were calculated using a MATLAB script downloaded from the SPM list [http://www.jiscmail.ac.uk/cgibin/wa.exe?A2=ind0807&L=SPM&P=R7075&1].
2.4. Statistical analysis The gray matter, white matter and CSF volumes of all the subjects were normalized by dividing the individual value by total intracranial volume (TIV) of the respective subjects (Table 2). Normalized GM (nGM) volume, normalized WM (nWM) volume, normalized CSF (nCSF) volume and TIV were compared among the three groups using the Student’s t test.
2816
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
Fig. 2.
The normalized, segmented, smoothed and modulated GM and WM data were analyzed using a voxel-wise statistical parametric mapping. An absolute threshold for masking of around 0.1 was used which computes the statistics in those areas that are above this threshold. ‘ANalysis Of VAriance (ANOVA)’ with gray matter volume of the individual subjects as a covariate, was used to find the regional changes in the gray matter volumes. Total Intra Cranial Volume was entered as a global variable to correct for the global brain volume of different subjects. A whole brain analysis was performed, with a significance level of p < 0.001, uncorrected for multiple comparisons. Since this threshold might have led to false positive results, in those areas which passed this threshold a small volume correction (SVC) was further applied, setting the cut off value for significance at FWE corrected p < 0.05 (at cluster level) and using a 8 mm radius. The anatomical representation of the clusters was related to cytoarchitectonic maps as implemented in the SPM Anatomy Toolbox [11].
3. Results Student’s t test did not reveal any statistically significant inter group differences in any of the volumetric parameters (Table 2).
3.1. VBM analysis 3.1.1. Changes in gray matter A significantly smaller gray matter volume was found bilaterally in cerebellum and left Inferior Parietal Cortex (PF) in the totally blind subjects relative to the controls (Fig. 1a(i), b(i) and Table 3). In partially blind subjects, gray matter (GM) volume loss was detected in left Inferior Parietal cortex (PFm), right caudate nucleus and left BA 17 as compared to the controls (Fig. 1a(ii), b(ii) and Table 3). Moreover, smaller gray matter volume was also observed in left cerebellum in totally blind subjects when compared to partial blinds (Fig. 1a(iii), b(iii) and Table 3). We also found GM volume increase in totally blind subjects as compared to controls in left middle occipital gyrus, left BA 6 and right hippocampus (EC) (Fig. 1a(iv), b(iv) and Table 3). Similarly a greater GM volume in totally blind subjects as compared to partial blinds was found in left BA 18 (Fig. 1a(v), b(v) and Table 3).
3.1.2. Changes in white matter A significantly smaller white matter volume was found bilaterally in calcarine gyrus in the totally blind subjects relative to the controls (Fig. 2a(i), b(i) and Table 4). In partially blind subjects, white matter volume loss was detected in left Th-somatosensory (region connecting thalamus with somatosensory cortex) as compared to the controls (Fig. 2a(ii), b(ii) and Table 4). No significant
Table 4 Regions of WM volume changes. Hemisphere
Localization of peak voxels
MNI Coordinates
Z value (peak voxel)
Cluster size (no. of voxels)
*
Controls minus Total blinds L Area 18, calcarine Gyrus L Area 17, calcarine Gyrus Area 17, calcarine Gyrus R Controls minus Partial blinds# L Th-somatosensory Total blinds minus Controls$ R Superior frontal gyrus L Superior frontal gyrus L Middle temporal gyrus TE 1.0, Heschl’s Gyrus R Partial blinds minus Controls@ R Superior frontal gyrus MNI: Montreal Neurological Institute; BA: Brodmann area. * Small Volume Correction (FWE corrected), p < 0.005. # Small Volume Correction (FWE corrected), p < 0.003. $ Small Volume Correction (FWE corrected), p < 0.007. @ Small Volume Correction (FWE corrected), p < 0.005.
0 −14 23
−76 −60 −64
7 4 7
4.71 4.52 3.82
2515
−17
−19
−8
3.33
285
17 −15 −44 47
38 54 −52 −22
43 15 −2 12
4.79 4.59 4.01 3.78
479 431 380 308
21
53
25
237
3.95
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
difference was found in the white matter volume between totally blind subjects when compared to partial blinds. We also found an increase in WM volume in totally blind subjects as compared to controls bilaterally in superior frontal gyrus, left middle temporal gyrus and right Heschl’s gyrus (Fig. 2a(iii), b(iii) and Table 4). Similarly a greater WM volume in partial blind subjects as compared to controls was found in right superior frontal gyrus (Fig. 2a(iv), b(iv) and Table 4).
4. Discussion Many earlier neuroimaging studies have focused on the functional reorganization in blind subjects. Activations of the primary and associative visual cortices are frequently observed in the blinds for non-visual tasks like auditory/tactile perception, language processing [4], etc. It has been suggested that blinds often employ different cognitive mechanisms than sighted subjects as a compensatory mechanism to overcome limitations of sight loss. A functional connectivity analysis study has shown decreased functional connectivities within the visual cortices as well as between the visual cortices and various other brain networks [12]. However, the introduction of Braille earlier in life and for longer daily practice times produced stronger functional connectivities between visual and somatosensory and language areas. Thus, the findings support both the general loss and compensatory plasticity hypothesis [12]. Significant alterations in average diffusivity and relative anisotropy in the white matter of the occipital lobe of the blind subjects has also been obtained [4]. Similarly, a diffusion tensor tractography study in early blinds by Shu et al. [13] showed decreased degree of connectivity, a reduced global efficiency, and an increased characteristic path length in their brain anatomical network, especially in the visual cortex. Morphological studies on blinds provide an opportunity to understand the effects of visual deprivation on structural brain development. Earlier morphometric studies [4–6] have shown gray-white matter changes primarily in the visual, somatosensory, and motor cortices in the blind subjects as compared to controls. Ptito et al. [7] have shown a significant atrophy of the geniculo-striate system, inferior longitudinal tract and in the posterior part of the corpus callosum indicating that the afferent projections to the visual cortex in congenitally blind subjects are largely atrophied. However, if the blind subjects included in the study are categorized into two groups as partial and total, then the results of these studies can give further insight into the effect of graded visual experience in the shaping up of human brain. Moreover, as compared to earlier morphological studies on blind subjects, the use of DARTEL could provide better information as it has been shown to outperform a number of other widely used and fully automated inter subject registration algorithms [10]. Total blinds in our study have shown a reduction in gray matter volume in cerebellum and inferior parietal cortex (IPC (PF)). Cerebellum is involved in fine motor coordination. Cerebellum has also been shown to play a role in visual attention [14]. The cerebellum is also involved in maintaining saccadic accuracy, and saccade-related activation in the human cerebellum has been shown previously using positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) [15]. Moreover, clinical, animal studies, and imaging data provide support for a role for the cerebellum in spatial functions [14]. The cerebellum receives strong inputs from the parietal lobe, and cerebellar damage leads to spatial deficits [14]. Area PF is the largest region of inferior parietal cortex. Functional imaging studies in human subjects showed involvement of the IPC in location matching and in the mental rotation of two and three-dimensional objects as well as of the body. Furthermore, it was shown that object-oriented action and object recognition activates the IPC, suggesting that some form of within-object spatial
2817
analysis has to be processed by this area [16]. In a study by Jäncke et al. [16], authors found the left fronto-parietal network involvement while subjects imagined constructing the previously palpated three-dimensional object. Authors concluded that the left Inferior Parietal Lobule is involved in the somatic perception of hand-object interactive movement and suggest that the underlying mechanism is the somatic integration of internal information about the body and external information about the object. Similarly, using a fMRI study on kinesthetic illusion task, Naito and Ehrsson [17] provided direct physiological evidence for a dominant role of the left hemisphere in the somatic perception of hand-object interaction. The left parietal areas are also found to be strongly associated with the control of object manipulation and tool use in human imaging studies [17]. A gray matter volume loss in the cerebellar and IPC may reflect structural reorganization in these regions on account of disuse related mechanism arising from the absence of the visual input. Blind subjects in our study also showed an increase in GM in area V5, Brodmann area 6 and hippocampus (EC) as compared to controls (Table 3), which might be attributed to compensatory mechanisms and nature of sensory inputs on which they rely to perform various tasks. Noppeney [4] has shown the involvement of extrastriate cortices in higher level semantic retrieval processes following early visual deprivation. Left hemispheric occipito-temporal activation in blind subjects has been demonstrated for: reading Braille relative to rest; tactile discrimination relative to a non-discrimination task; haptic exploration of scenes; sound imagery relative to listening to noise stimuli and other verbal tasks such as verb generation and sentence processing [4]. Individuals with no visual experience rely on supramodal brain areas within the extrastriate visual cortex to acquire knowledge about shape, movement, and localization of objects, through non-visual sensory modalities, including touch and hearing [18]. The Supplementary Motor Area (medial Brodmann area (BA) 6) and the premotor cortex (lateral BA 6) are both thought to play a role in the planning of complex, coordinated movements [6]. As the frontal lobes control executive function, increase in those areas may be related to a reorganization of those functions that now need to be generated using information from one less sense [6]. The increased GM in hippocampus (CA) is supported by a recent study by Leporé et al. [19] who found that the anterior right hippocampus was significantly larger, and the posterior right hippocampus significantly smaller in blind individuals. However, in our study blind subjects showed a greater GM volume than controls in entorhinal cortex of hippocampal formation. Specifically, the entorhinal cortex (EC) may be necessary for long-term spatial memory storage. The EC is a major input and output structure for the hippocampus, receiving sensory information from association cortices and sending projections to all hippocampal subfields [20]. Blind subjects must memorize extensive spatial information to compensate for their lack of vision. Increased demands on spatial cognition and memory may have resulted in the obtained GM volume increase in EC. On the other hand, partial blind subjects showed a slightly greater reduction in GM volume in inferior parietal cortex as compared to totally blind subjects when both groups were separately compared with controls. However, on direct comparison between partial and total blinds, IPC did not show any significant volumetric differences. Partial blind subjects also showed gray matter volume loss in cuneus (BA 17) and caudate nucleus. Thus, partial blinds showed gray matter reduction in primary visual cortex which was preserved in total blind subjects as compared to controls. Gray matter volume reduction observed in primary visual cortex (BA17) in partially blind subjects could be a manifestation of adaptive responses evoked by disuse related mechanisms originating from loss of peripheral visual input as compared to controls [5].
2818
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
The absence of GM reduction in totally blind subjects in primary visual cortex and a greater GM volume in middle occipital gyrus and BA18 (as compared to partial blind subjects) may be because of cross-modal functional plasticity in totally blind subjects. As in the absence of visual input, occipital lobe in blind subjects is involved in processing of other higher order cognitive tasks viz. semantic processing, auditory processing, Braille reading, tactile exploration of objects, etc., disuse related mechanism might not play a role in these regions in totally blind subjects. Moreover partial blind subjects had a greater GM volume when compared to total blinds in cerebellum. Thus, activity dependent mechanisms might play an important role in causing the structural changes. Our results suggest that since partial blinds have greater extent of visual experience and they have better motor coordination as compared to totally blind individuals, the structural reorganization is different in partial blind subjects as compared to total blind subjects. Along with GM changes, our study also revealed WM changes between the study groups. In totally blind subjects, a WM volume loss was observed bilaterally in calcarine gyrus (BA 17, 18). Thus, even though our study showed GM preservation in primary visual cortex and a greater GM volume in area V5 in totally blind subjects, WM atrophy was observed in these regions. In partial blind subjects, WM volume loss was found in Th-somatosensory. It is that portion of the thalamus that sends touch and proprioceptive information to the primary somatosensory cortex [http://www.fzjuelich.de/inm/inm-1/DE/Forschung/ docs/SPMAnantomy Toolbox/SPMAnantomyToolbox node.html]. White matter hypertrophy was also found in few regions both in total and partial blind subjects as compared to controls (Table 4). An increase in WM in superior frontal gyrus and primary auditory cortex (TE 1.0) may be on account of compensatory mechanisms as blind individuals may rely on other senses in the absence of visual input. It has been shown by a recent diffusion tensor tractography study that the precentral gyrus had more or strengthened connections with other brain regions, therefore predicting better performance in information transfer and interaction than in sighted subjects [13]. This finding along with ours supports experience-dependent compensatory plasticity. In the absence of visual experience, early blind subjects need more practice to perform the same routine activities of the sighted subjects, and they engage in much finer finger movements, such as tactile exploration of objects and Braille reading. The enhanced motor activity in early blind subjects may increase the number, diameter and in particular myelination of the relevant axons, especially during the critical early period of neurodevelopment [13]. Previous studies have suggested that early blind individuals attend to shifting textural patterns formed by the different Braille letters, and possibly process an overall orthography similar to the holistic interpretations of letters in reading print. Such holistic analyses would require the integrative mechanisms more likely in higher visual/language areas in posterior/middle temporal cortex [4]. Compensatory plasticity in the auditory cortex of the blind subjects has also been shown by a study on the analysis of regional cortical thickness and surface area [8]. We did not find any region with significant difference in the WM volumes between total and partial blind subjects. Almost all the earlier morphometric studies [4–6,8] on early total blind individuals have reported GM atrophy in visual areas viz. BA 17/18, which is contradictory to our findings. In our study, we did not find any GM changes in BA 17/18 in totally blind individuals as compared to controls. Moreover, we found an increase in GM in area V5 of the occipital lobe in them. We attribute the GM preservation in the occipital lobe in total blind subjects to cross modal functional and compensatory plasticity as shown in earlier studies [4,8,13]. A Transcranial Magnetic Stimulation (TMS) study concluded that blindness from an early age can cause the visual cortex to play a role in somatosensory processing. Authors proposed
that this cross-modal plasticity may account in part for the superior tactile perceptual abilities of blind subjects [4]. These studies give evidence that since occipital lobe is functional in the blind individuals for processing of higher order cognitive functions; disuse related mechanism for structural reorganization might be less dominant. However, the use of a registration method (DARTEL), which has not been used earlier in morphometric studies in blind subjects, might also be the result of this discrepancy. Moreover, the inconsistency may also be associated with the age group of the subjects. The mean age in our study was 23.61 years while that in above studies was around 36 years. In conclusion, the study suggests that loss of vision at an early age can induce significant morphological changes in the cerebellar, parieto-occipital, temporal and the hippocampal regions due to either disuse-related or compensatory mechanisms originating on account of loss of peripheral visual input. Further, the morphological changes are different between totally blind and partial blind subjects because of the difference in the extent of visual experience. However, more studies using different image registration algorithms and more number of subjects are needed for generalizing the results. Conflict of interest None. Acknowledgements We thank the volunteers from ‘Seva Kutir’, Hostel for College Going Blind Students, Kingsway Camp, Delhi for participating in the study. We also thank Dr. Asha Saxena, Ophthalmologist from Aruna Asaf Ali Hospital, Delhi for performing the ophthalmologic examinations on the blind subjects and categorizing them as total and partial blinds. References [1] Ashburner J, Friston KJ. Voxel-based morphometry-the methods. Neuroimage 2000;11:805–21. [2] Kakeda S, Korogi Y. The efficacy of a voxel-based morphometry on the analysis of imaging in schizophrenia, temporal lobe epilepsy, and Alzheimer’s disease/mild cognitive impairment: a review. Neuroradiology 2010;52(8):711–21. [3] Draganski B, Gaser C, Busch V, Schuierer G, Bogdahn U, May A. Neuroplasticity: changes in gray matter induced by training. Nature 2004;427:311–2. [4] Noppeney U. The effects of visual deprivation on functional and structural organization of the human brain. Neuroscience and Biobehavioral Reviews 2007;31:1169–80. [5] Pan WJ, Wu G, Li CX, Lin F, Sun J, Lei H. Progressive atrophy in the optic pathway and visual cortex of early blind Chinese adults: a voxel-based morphometry magnetic resonance imaging study. Neuroimage 2007;37(1):212–20. [6] Leporé N, Voss P, Lepore F, et al. Brain structure changes visualized in earlyand late-onset blind subjects. Neuroimage 2010;49(1):134–40. [7] Ptito M, Schneider FC, Paulson OB, Kupers R. Alterations of the visual pathways in congenital blindness. Experimental Brain Research 2008;187(1):41–9. [8] Park H, Lee JD, Kim EY, et al. Morphological alterations in the congenital blind based on the analysis of cortical thickness and surface area. NeuroImage 2009;47:98–106. [9] John Ashburner. A fast diffeomorphic image registration algorithm. NeuroImage 2007;38(1):95–113. [10] Klein A, Andersson J, Ardekani B, et al. Evaluation of 14 nonlinear deformation algorithms applied to human brain MRI registration. Neuroimage 2009;46:786–802. [11] Eickhoff SB, Stephan KE, Mohlberg H, et al. A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data. NeuroImage 2005;25(4):1325–35. [12] Liu Y, Yu C, Liang M, et al. Whole brain functional connectivity in the early blind. Brain 2007;130:2085–96. [13] Shu N, Liu Y, Li J, Li Y, Yu C, Jiang T. Altered anatomical network in early blindness revealed by diffusion tensor tractography. PLoS ONE 2009;4(9):e7228. [14] Stoodley CJ, Schmahmanna JD. Functional topography in the human cerebellum: a meta-analysis of neuroimaging studies. NeuroImage 2009;44(2):489–501.
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819 [15] van Broekhoven PCA, Schraa-Tam CKL, van der Lugt A, Smits M, Frens MA, van der Geest JN. Cerebellar contributions to the processing of saccadic errors. Cerebellum 2009, doi:10.1007/s12311-009-0116-6. [16] Jäncke L, Kleinschmidt A, Mirzazade S, Shah NJ, Freund HJ. Role of the inferior parietal cortex in linking the tactile perception and manual construction of object shapes. Cerebral Cortex 2001;11(2):114–21. [17] Naito E, Ehrsson HH. Somatic sensation of hand-object interactive movement is associated with activity in the left inferior parietal cortex. Journal of Neuroscience 2006;26(14):3783–90. [18] Ricciardi E, Pietrini P. New light from the dark: what blindness can teach us about brain function. Current Opinion in Neurology 2011;4(4):357–63. [19] Leporé N, Shi Y, Lepore F, et al. Pattern of hippocampal shape and volume differences in blind subjects. Neuroimage 2009;46(4):949–57.
2819
[20] Hebert AE, Dash PK. Extracellular signal-regulated kinase activity in the entorhinal cortex is necessary for long-term spatial memory. Learning and Memory 2002;9:156–66.
Web references [21] Ashburner J. VBM tutorial. www.fil.ion.ucl.ac.uk/∼john/misc/VBMclass10.pdf. [22] http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0807&L=SPM&P=R7075&1. [23] http://www.fz-juelich.de/inm/inm-1/DE/Forschung/ docs/SPMAnantomy Toolbox/SPMAnantomyToolbox node.html.
Contents lists available at SciVerse ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Effect of visual experience on structural organization of the human brain: A voxel based morphometric study using DARTEL Shilpi Modi 1 , Manisha Bhattacharya 1 , Namita Singh 1 , Rajendra Prasad Tripathi 1 , Subash Khushu ∗ NMR Research Centre, Institute of Nuclear Medicine and Allied Sciences (DRDO), Lucknow Road, Timarpur, Delhi, India
a r t i c l e
i n f o
Article history: Received 13 September 2011 Received in revised form 18 October 2011 Accepted 25 October 2011 Keywords: Voxel-Based Morphometry (VBM) DARTEL Blindness Atrophy Structural reorganization
a b s t r a c t Objective: To investigate structural reorganization in the brain with differential visual experience using Voxel-Based Morphometry with Diffeomorphic Anatomic Registration Through Exponentiated Lie algebra algorithm (DARTEL) approach. Materials and methods: High resolution structural MR images were taken in fifteen normal sighted healthy controls, thirteen totally blind subjects and six partial blind subjects. The analysis was carried out using SPM8 software on MATLAB 7.6.0 platform. Results: VBM study revealed gray matter volume atrophy in the cerebellum and left inferior parietal cortex in total blind subjects and in left inferior parietal cortex, right caudate nucleus, and left primary visual cortex in partial blind subjects as compared to controls. White matter volume loss was found in calcarine gyrus in total blind subjects and Thlamus-somatosensory region in partially blind subjects as compared to controls. Besides, an increase in Gray Matter volume was also found in left middle occipital and middle frontal gyrus and right entorhinal cortex, and an increase in White Matter volume was found in superior frontal gyrus, left middle temporal gyrus and right Heschl’s gyrus in totally blind subjects as compared to controls. Comparison between total and partial blind subjects revealed a greater Gray Matter volume in left cerebellum of partial blinds and left Brodmann area 18 of total blind subjects. Conclusion: Results suggest that, loss of vision at an early age can induce significant structural reorganization on account of the loss of visual input. These plastic changes are different in early onset of total blindness as compared to partial blindness. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Voxel-Based Morphometry (VBM), introduced by Ashburner and Friston offers rapid unbiased assessment of the whole brain [1]. This automated technique examines differences in local composition of tissue on a voxel-by-voxel basis after global differences in anatomy have been discounted. VBM has been used to characterize structural brain differences in a variety of disease conditions including Down’s syndrome, autism, schizophrenia, epilepsy, Alzheimer’s disease, Parkinsonism, type II diabetes mellitus [2], etc. VBM studies have also investigated the impact of learning and practice on brain structure of the healthy subjects [3]. These morphological studies clearly give evidence of morphological plasticity of the brain. Moreover, the morphology and the function seem to be inter-related. Morphological studies on diseased conditions have shown that
∗ Corresponding author. Tel.: +91 11 23905313/5336; fax: +91 11 23919509. E-mail addresses: modi [email protected] (S. Modi), [email protected] (M. Bhattacharya), [email protected] (N. Singh), [email protected] (R.P. Tripathi), [email protected] (S. Khushu). 1 Tel.: +91 11 23905336; fax: +91 11 23919509. 0720-048X/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2011.10.022
altered function is often associated with brain atrophy in the related areas [2]. On the other hand, training or acquisition of new skills often results in an increase in gray matter volume in the relevant regions [3]. Neuroimaging studies have demonstrated that the visual cortex of both the early and late blind subjects exhibits significant cross-modal functional plasticity [4]. However, fewer studies have dealt with structural plasticity associated with blindness using automated techniques like Voxel or Tensor Based Morphometry [4–8]. As the disuse related and compensatory mechanisms, both govern the structural plasticity in the brain, we hypothesized that the morphological changes would be different in totally blind subjects who had no vision left in them, as compared to the morphological changes in partial blind subjects who had partial vision left. In the present study we attempted to detect gray matter changes in the brain, with the extent of sightedness a person has, using Voxel Based Morphometry technique applied to three groups consisting of totally blind subjects, partially blind subjects and controls. To improve inter-subject registration of the MRI images, we applied Diffeomorphic Anatomic Registration Through Exponentiated Lie algebra algorithm (DARTEL) [9], which has been found to optimize the sensitivity of such analyses [10] by using
2812
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
Table 1 Clinical characteristics of the blind subjects. Subject IDa
Age (years)
Onset of blindness
Cause of blindness
Light sense
T1 T2
21 23
Birth 3.0
None None
T3 T4 T5 T6 T7 T8 T9
23 25 24 21 20 23 25
5.0 years Birth Birth 4.0 years Birth Birth 6.0 years
T10 T11 T12 T13 P1 P2 P3 P4 P5 P6
21 23 20 24 21 20 23 27 20 25
6.0 years Birth Birth 4.0 years 5.0 years Birth Birth Birth 5.0 years Birth
Congenital cataract Poor medical aid after conjunctivitis Eye infection Optic nerve atrophy Glaucoma Post iridocyclitis Congenital cataract Post iridocyclitis Poor medication when affected with smallpox Ophthalmitis Congenital cataract Optic nerve atrophy Ophthalmitis Congenital cataract Optic nerve atrophy Congenital cataract Congenital cataract Congenital cataract Post iridocyclitis
a
Table 2 Normalized gray matter volume (nGM), normalized white matter volume (nWM), normalized cerebrospinal fluid volume (nCSF) and total intracranial volume (TIV) for the three groups (mean ± SD).
Controls Total blinds Partial blinds
None Bright light None None Bright light Bright light None
TIV (ml)*
nGM*
nWM*
nCSF*
1445.77 ± 112.68 1403.66 ± 84.62 1358.68 ± 0.09
0.45 ± 0.01 0.45 ± 0.01 0.46 ± 0.01
0.34 ± 0.01 0.33 ± 0.01 0.33 ± 0.01
0.21 ± 0.01 0.22 ± 0.01 0.21 ± 0.01
* TIV, nGM, nWM and nCSF showed no significant difference between the three groups (p > 0.05).
age ± SD = 22.42 ± 1.78 years, group 2) and six right handed early partial blind males (20–25 years old, mean age ± SD = 22.86 ± 2.67 years, group 3) participated in the study. No statistically significant age difference was found between the three groups (p < 0.134 for groups 1 and 2; p < 0.222 for groups 1 and 3; p < 0.974 for groups 2 and 3). All the subjects chosen for the study were Indian natives and none of them had any clinical evidences of stroke, head injury, cardiovascular diseases, history of drug dependence, psychiatric disorder or cognitive impairment nor did they have any cortical infarctions on the T2 -weighted MR images. The reported onset of total or partial blindness for all the subjects ranged from birth to 6 years of age (Table 1). In ophthalmologic examinations, five of the totally blind subjects were found to be sensitive only to strong sunlight that too only from one of their eyes. The maximum distance to which partial blind subjects could perform finger counting ranged from ‘close to face’ to 2 m. Both the partially blind and totally blind subjects were students, undergoing either graduate or post graduate courses in various subjects of Arts (Political Science, History, Hindi, Programming, Sanskrit, Education, English or Buddhist Study). Further, all subjects gave their consent to participate in the study.
Bright light None Bright light None Yes Yes Yes Yes Yes Yes
T – total blind subjects; P – partial blind subjects.
the Levenberg–Marquardt strategy as compared to standard VBM. The current study on one hand explores the morphological changes using DARTEL, which to the best of our knowledge has never been applied in blind subjects, as compared to previous studies using other methods viz. optimized VBM [5], fast fluid registration [6], analysis of cortical thickness and surface area [8], etc. Secondly, it also compares the morphological changes in early total and partial blinds.
2.2. Scanning protocol 2. Materials and methods The MRI scans were acquired using 1.5 Tesla whole-body MRI system (Siemens Magnetom Vision, Erlangen, Germany) with a circularly polarized head coil and 25 mT/m actively shielded gradient system. T1 weighted 3D Magnetization Prepared Rapid Acquisition Gradient Echo (MPRAGE) sequence with 160 slices (1 mm slice thickness) covering the whole brain was performed in
2.1. Subjects Fifteen normal sighted right handed healthy males (21–27 years old, mean age ± SD = 23.61 ± 2.10 years, group 1), thirteen right handed early totally blind males (20–25 years old, mean Table 3 Regions of GM volume changes. Hemisphere
Localization of peak voxels
Controls minus Total blinds* L Lobule VIIIa (Hem), Cerebellum Lobule VIIa Crus II (Hem), Cerebellum R IPC (PF), Inferior Parietal Lobule L Controls minus Partial blinds# L IPC (PFm), Middle Temporal Gyrus Caudate Nucleus R BA 17, Cuneus L BA 17 L Partial blinds minus Total blinds@ L Lobule VIIIa (Hem) L Lobule VIIa Crus II (Hem) Total blinds minus Controls$ L Middle Occipital Gyrus (hOC5 (V5)) BA 6, Middle Frontal Gyrus L Hippocampus (EC) R Total blinds minus Partial blinds! L BA 18 MNI: Montreal Neurological Institute; BA: Brodmann area. * Small Volume Correction (FWE corrected), p < 0.012. # Small Volume Correction (FWE corrected), p < 0.011. @ Small Volume Correction (FWE corrected), p < 0.002. $ Small Volume Correction (FWE corrected), p < 0.01. ! Small Volume Correction (FWE corrected), p < 0.01.
MNI coordinates
Z value (peak voxel)
Cluster size (no. of voxels)
−27 47 −57
−58 −65 −39
−63 −57 43
4.26 3.98 3.37
4836 2898 369
−62 12 −12 −21
−54 15 −79 −64
19 16 16 0
3.58 3.67 3.75 3.95
473 319 302 235
−24 −42
−66 −49
−60 −51
3.49 3.48
460 301
−42 −50 20
−79 51 3
1 6 −35
4.36 4.32 3.28
1034 451 157
−5
−81
28
3.66
354
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
2813
Fig. 1. (a) Three dimensional rendered view and (b) overlay of the peak cluster on the gray matter template provided in SPM8, of regions of GM volume loss in (i) total blind subjects as compared to controls; (ii) partial blind subjects as compared to controls; (iii) total blind subjects as compared to partial blind subjects and regions of GM volume increase in (iv) total blind subjects as compared to controls; (v) total blind subjects as compared to partial blind subjects.
the sagittal plane with TR/TE/TI of 9.7 ms/4 ms/300 ms, flip angle of 12◦ , FOV = 256 mm × 256 mm, number of acquisitions = 1 and an in-plane resolution of 1 mm × 1 mm. All subjects also underwent an axial double spin echo sequence (TE1 = 20 ms, TE2 = 90 ms), which was used to exclude subjects with possible cortical infarctions.
2.3. Data processing Data were processed using Statistical Parametric Mapping (SPM8, Wellcome Department of Cognitive Neurology) implemented in MATLAB R2008a (version 7.6.0 (Mathworks, Sherborn, MA)). To improve the registration of the MRI images,
2814
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
Fig. 1.
DARTEL toolbox for SPM8 was used. All the steps were carried out as suggested by Ashburner, 2010 [www.fil.ion.ucl.ac.uk/∼john/misc/VBMclass10.pdf]. In brief, the anatomic images were first manually reoriented so that the mm coordinate of the anterior commissure matched the origin (0, 0, 0), and the orientation approximated Montreal Neurological Institute (MNI) space. Next, T1 -weighted images were classified into gray matter, white matter and cerebrospinal fluid (CSF) using the ‘new-segment’ routine implemented in SPM8, that gives both the native space versions and DARTEL imported versions
of the tissues. The DARTEL imported versions of gray and white matter were used to generate the flow fields (which encode the shapes), and a series of template images by running ‘DARTEL (create templates)’ routine. During this step, DARTEL increases the accuracy of inter-subject alignment by modeling the shape of each brain using millions of parameters (three parameters for each voxel). DARTEL works by aligning gray matter among the images, while simultaneously aligning white matter. This is achieved by generating increasingly crisp average template data, to which the data are iteratively aligned. The flow fields
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
2815
Fig. 2. (a) Three dimensional rendered view and (b) overlay of the peak cluster on the white matter template provided in SPM8, of regions WM volume loss in (i) total blind subjects as compared to controls; (ii) partial blind subjects as compared to controls and regions of WM volume increase in (iii) total blind subjects as compared to controls; (iv) partial blind subjects as compared to controls.
and the final template image created in the previous step are used to generate smoothed (10 mm FWHM), modulated, spatially normalized and Jacobian scaled gray and white matter images resliced to 1.5 mm × 1.5 mm × 1.5 mm voxel size, in MNI space. Total gray matter (GM) volume, white matter (WM) volume, CSF volume and Total Intracranial Volume were calculated using a MATLAB script downloaded from the SPM list [http://www.jiscmail.ac.uk/cgibin/wa.exe?A2=ind0807&L=SPM&P=R7075&1].
2.4. Statistical analysis The gray matter, white matter and CSF volumes of all the subjects were normalized by dividing the individual value by total intracranial volume (TIV) of the respective subjects (Table 2). Normalized GM (nGM) volume, normalized WM (nWM) volume, normalized CSF (nCSF) volume and TIV were compared among the three groups using the Student’s t test.
2816
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
Fig. 2.
The normalized, segmented, smoothed and modulated GM and WM data were analyzed using a voxel-wise statistical parametric mapping. An absolute threshold for masking of around 0.1 was used which computes the statistics in those areas that are above this threshold. ‘ANalysis Of VAriance (ANOVA)’ with gray matter volume of the individual subjects as a covariate, was used to find the regional changes in the gray matter volumes. Total Intra Cranial Volume was entered as a global variable to correct for the global brain volume of different subjects. A whole brain analysis was performed, with a significance level of p < 0.001, uncorrected for multiple comparisons. Since this threshold might have led to false positive results, in those areas which passed this threshold a small volume correction (SVC) was further applied, setting the cut off value for significance at FWE corrected p < 0.05 (at cluster level) and using a 8 mm radius. The anatomical representation of the clusters was related to cytoarchitectonic maps as implemented in the SPM Anatomy Toolbox [11].
3. Results Student’s t test did not reveal any statistically significant inter group differences in any of the volumetric parameters (Table 2).
3.1. VBM analysis 3.1.1. Changes in gray matter A significantly smaller gray matter volume was found bilaterally in cerebellum and left Inferior Parietal Cortex (PF) in the totally blind subjects relative to the controls (Fig. 1a(i), b(i) and Table 3). In partially blind subjects, gray matter (GM) volume loss was detected in left Inferior Parietal cortex (PFm), right caudate nucleus and left BA 17 as compared to the controls (Fig. 1a(ii), b(ii) and Table 3). Moreover, smaller gray matter volume was also observed in left cerebellum in totally blind subjects when compared to partial blinds (Fig. 1a(iii), b(iii) and Table 3). We also found GM volume increase in totally blind subjects as compared to controls in left middle occipital gyrus, left BA 6 and right hippocampus (EC) (Fig. 1a(iv), b(iv) and Table 3). Similarly a greater GM volume in totally blind subjects as compared to partial blinds was found in left BA 18 (Fig. 1a(v), b(v) and Table 3).
3.1.2. Changes in white matter A significantly smaller white matter volume was found bilaterally in calcarine gyrus in the totally blind subjects relative to the controls (Fig. 2a(i), b(i) and Table 4). In partially blind subjects, white matter volume loss was detected in left Th-somatosensory (region connecting thalamus with somatosensory cortex) as compared to the controls (Fig. 2a(ii), b(ii) and Table 4). No significant
Table 4 Regions of WM volume changes. Hemisphere
Localization of peak voxels
MNI Coordinates
Z value (peak voxel)
Cluster size (no. of voxels)
*
Controls minus Total blinds L Area 18, calcarine Gyrus L Area 17, calcarine Gyrus Area 17, calcarine Gyrus R Controls minus Partial blinds# L Th-somatosensory Total blinds minus Controls$ R Superior frontal gyrus L Superior frontal gyrus L Middle temporal gyrus TE 1.0, Heschl’s Gyrus R Partial blinds minus Controls@ R Superior frontal gyrus MNI: Montreal Neurological Institute; BA: Brodmann area. * Small Volume Correction (FWE corrected), p < 0.005. # Small Volume Correction (FWE corrected), p < 0.003. $ Small Volume Correction (FWE corrected), p < 0.007. @ Small Volume Correction (FWE corrected), p < 0.005.
0 −14 23
−76 −60 −64
7 4 7
4.71 4.52 3.82
2515
−17
−19
−8
3.33
285
17 −15 −44 47
38 54 −52 −22
43 15 −2 12
4.79 4.59 4.01 3.78
479 431 380 308
21
53
25
237
3.95
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
difference was found in the white matter volume between totally blind subjects when compared to partial blinds. We also found an increase in WM volume in totally blind subjects as compared to controls bilaterally in superior frontal gyrus, left middle temporal gyrus and right Heschl’s gyrus (Fig. 2a(iii), b(iii) and Table 4). Similarly a greater WM volume in partial blind subjects as compared to controls was found in right superior frontal gyrus (Fig. 2a(iv), b(iv) and Table 4).
4. Discussion Many earlier neuroimaging studies have focused on the functional reorganization in blind subjects. Activations of the primary and associative visual cortices are frequently observed in the blinds for non-visual tasks like auditory/tactile perception, language processing [4], etc. It has been suggested that blinds often employ different cognitive mechanisms than sighted subjects as a compensatory mechanism to overcome limitations of sight loss. A functional connectivity analysis study has shown decreased functional connectivities within the visual cortices as well as between the visual cortices and various other brain networks [12]. However, the introduction of Braille earlier in life and for longer daily practice times produced stronger functional connectivities between visual and somatosensory and language areas. Thus, the findings support both the general loss and compensatory plasticity hypothesis [12]. Significant alterations in average diffusivity and relative anisotropy in the white matter of the occipital lobe of the blind subjects has also been obtained [4]. Similarly, a diffusion tensor tractography study in early blinds by Shu et al. [13] showed decreased degree of connectivity, a reduced global efficiency, and an increased characteristic path length in their brain anatomical network, especially in the visual cortex. Morphological studies on blinds provide an opportunity to understand the effects of visual deprivation on structural brain development. Earlier morphometric studies [4–6] have shown gray-white matter changes primarily in the visual, somatosensory, and motor cortices in the blind subjects as compared to controls. Ptito et al. [7] have shown a significant atrophy of the geniculo-striate system, inferior longitudinal tract and in the posterior part of the corpus callosum indicating that the afferent projections to the visual cortex in congenitally blind subjects are largely atrophied. However, if the blind subjects included in the study are categorized into two groups as partial and total, then the results of these studies can give further insight into the effect of graded visual experience in the shaping up of human brain. Moreover, as compared to earlier morphological studies on blind subjects, the use of DARTEL could provide better information as it has been shown to outperform a number of other widely used and fully automated inter subject registration algorithms [10]. Total blinds in our study have shown a reduction in gray matter volume in cerebellum and inferior parietal cortex (IPC (PF)). Cerebellum is involved in fine motor coordination. Cerebellum has also been shown to play a role in visual attention [14]. The cerebellum is also involved in maintaining saccadic accuracy, and saccade-related activation in the human cerebellum has been shown previously using positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) [15]. Moreover, clinical, animal studies, and imaging data provide support for a role for the cerebellum in spatial functions [14]. The cerebellum receives strong inputs from the parietal lobe, and cerebellar damage leads to spatial deficits [14]. Area PF is the largest region of inferior parietal cortex. Functional imaging studies in human subjects showed involvement of the IPC in location matching and in the mental rotation of two and three-dimensional objects as well as of the body. Furthermore, it was shown that object-oriented action and object recognition activates the IPC, suggesting that some form of within-object spatial
2817
analysis has to be processed by this area [16]. In a study by Jäncke et al. [16], authors found the left fronto-parietal network involvement while subjects imagined constructing the previously palpated three-dimensional object. Authors concluded that the left Inferior Parietal Lobule is involved in the somatic perception of hand-object interactive movement and suggest that the underlying mechanism is the somatic integration of internal information about the body and external information about the object. Similarly, using a fMRI study on kinesthetic illusion task, Naito and Ehrsson [17] provided direct physiological evidence for a dominant role of the left hemisphere in the somatic perception of hand-object interaction. The left parietal areas are also found to be strongly associated with the control of object manipulation and tool use in human imaging studies [17]. A gray matter volume loss in the cerebellar and IPC may reflect structural reorganization in these regions on account of disuse related mechanism arising from the absence of the visual input. Blind subjects in our study also showed an increase in GM in area V5, Brodmann area 6 and hippocampus (EC) as compared to controls (Table 3), which might be attributed to compensatory mechanisms and nature of sensory inputs on which they rely to perform various tasks. Noppeney [4] has shown the involvement of extrastriate cortices in higher level semantic retrieval processes following early visual deprivation. Left hemispheric occipito-temporal activation in blind subjects has been demonstrated for: reading Braille relative to rest; tactile discrimination relative to a non-discrimination task; haptic exploration of scenes; sound imagery relative to listening to noise stimuli and other verbal tasks such as verb generation and sentence processing [4]. Individuals with no visual experience rely on supramodal brain areas within the extrastriate visual cortex to acquire knowledge about shape, movement, and localization of objects, through non-visual sensory modalities, including touch and hearing [18]. The Supplementary Motor Area (medial Brodmann area (BA) 6) and the premotor cortex (lateral BA 6) are both thought to play a role in the planning of complex, coordinated movements [6]. As the frontal lobes control executive function, increase in those areas may be related to a reorganization of those functions that now need to be generated using information from one less sense [6]. The increased GM in hippocampus (CA) is supported by a recent study by Leporé et al. [19] who found that the anterior right hippocampus was significantly larger, and the posterior right hippocampus significantly smaller in blind individuals. However, in our study blind subjects showed a greater GM volume than controls in entorhinal cortex of hippocampal formation. Specifically, the entorhinal cortex (EC) may be necessary for long-term spatial memory storage. The EC is a major input and output structure for the hippocampus, receiving sensory information from association cortices and sending projections to all hippocampal subfields [20]. Blind subjects must memorize extensive spatial information to compensate for their lack of vision. Increased demands on spatial cognition and memory may have resulted in the obtained GM volume increase in EC. On the other hand, partial blind subjects showed a slightly greater reduction in GM volume in inferior parietal cortex as compared to totally blind subjects when both groups were separately compared with controls. However, on direct comparison between partial and total blinds, IPC did not show any significant volumetric differences. Partial blind subjects also showed gray matter volume loss in cuneus (BA 17) and caudate nucleus. Thus, partial blinds showed gray matter reduction in primary visual cortex which was preserved in total blind subjects as compared to controls. Gray matter volume reduction observed in primary visual cortex (BA17) in partially blind subjects could be a manifestation of adaptive responses evoked by disuse related mechanisms originating from loss of peripheral visual input as compared to controls [5].
2818
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819
The absence of GM reduction in totally blind subjects in primary visual cortex and a greater GM volume in middle occipital gyrus and BA18 (as compared to partial blind subjects) may be because of cross-modal functional plasticity in totally blind subjects. As in the absence of visual input, occipital lobe in blind subjects is involved in processing of other higher order cognitive tasks viz. semantic processing, auditory processing, Braille reading, tactile exploration of objects, etc., disuse related mechanism might not play a role in these regions in totally blind subjects. Moreover partial blind subjects had a greater GM volume when compared to total blinds in cerebellum. Thus, activity dependent mechanisms might play an important role in causing the structural changes. Our results suggest that since partial blinds have greater extent of visual experience and they have better motor coordination as compared to totally blind individuals, the structural reorganization is different in partial blind subjects as compared to total blind subjects. Along with GM changes, our study also revealed WM changes between the study groups. In totally blind subjects, a WM volume loss was observed bilaterally in calcarine gyrus (BA 17, 18). Thus, even though our study showed GM preservation in primary visual cortex and a greater GM volume in area V5 in totally blind subjects, WM atrophy was observed in these regions. In partial blind subjects, WM volume loss was found in Th-somatosensory. It is that portion of the thalamus that sends touch and proprioceptive information to the primary somatosensory cortex [http://www.fzjuelich.de/inm/inm-1/DE/Forschung/ docs/SPMAnantomy Toolbox/SPMAnantomyToolbox node.html]. White matter hypertrophy was also found in few regions both in total and partial blind subjects as compared to controls (Table 4). An increase in WM in superior frontal gyrus and primary auditory cortex (TE 1.0) may be on account of compensatory mechanisms as blind individuals may rely on other senses in the absence of visual input. It has been shown by a recent diffusion tensor tractography study that the precentral gyrus had more or strengthened connections with other brain regions, therefore predicting better performance in information transfer and interaction than in sighted subjects [13]. This finding along with ours supports experience-dependent compensatory plasticity. In the absence of visual experience, early blind subjects need more practice to perform the same routine activities of the sighted subjects, and they engage in much finer finger movements, such as tactile exploration of objects and Braille reading. The enhanced motor activity in early blind subjects may increase the number, diameter and in particular myelination of the relevant axons, especially during the critical early period of neurodevelopment [13]. Previous studies have suggested that early blind individuals attend to shifting textural patterns formed by the different Braille letters, and possibly process an overall orthography similar to the holistic interpretations of letters in reading print. Such holistic analyses would require the integrative mechanisms more likely in higher visual/language areas in posterior/middle temporal cortex [4]. Compensatory plasticity in the auditory cortex of the blind subjects has also been shown by a study on the analysis of regional cortical thickness and surface area [8]. We did not find any region with significant difference in the WM volumes between total and partial blind subjects. Almost all the earlier morphometric studies [4–6,8] on early total blind individuals have reported GM atrophy in visual areas viz. BA 17/18, which is contradictory to our findings. In our study, we did not find any GM changes in BA 17/18 in totally blind individuals as compared to controls. Moreover, we found an increase in GM in area V5 of the occipital lobe in them. We attribute the GM preservation in the occipital lobe in total blind subjects to cross modal functional and compensatory plasticity as shown in earlier studies [4,8,13]. A Transcranial Magnetic Stimulation (TMS) study concluded that blindness from an early age can cause the visual cortex to play a role in somatosensory processing. Authors proposed
that this cross-modal plasticity may account in part for the superior tactile perceptual abilities of blind subjects [4]. These studies give evidence that since occipital lobe is functional in the blind individuals for processing of higher order cognitive functions; disuse related mechanism for structural reorganization might be less dominant. However, the use of a registration method (DARTEL), which has not been used earlier in morphometric studies in blind subjects, might also be the result of this discrepancy. Moreover, the inconsistency may also be associated with the age group of the subjects. The mean age in our study was 23.61 years while that in above studies was around 36 years. In conclusion, the study suggests that loss of vision at an early age can induce significant morphological changes in the cerebellar, parieto-occipital, temporal and the hippocampal regions due to either disuse-related or compensatory mechanisms originating on account of loss of peripheral visual input. Further, the morphological changes are different between totally blind and partial blind subjects because of the difference in the extent of visual experience. However, more studies using different image registration algorithms and more number of subjects are needed for generalizing the results. Conflict of interest None. Acknowledgements We thank the volunteers from ‘Seva Kutir’, Hostel for College Going Blind Students, Kingsway Camp, Delhi for participating in the study. We also thank Dr. Asha Saxena, Ophthalmologist from Aruna Asaf Ali Hospital, Delhi for performing the ophthalmologic examinations on the blind subjects and categorizing them as total and partial blinds. References [1] Ashburner J, Friston KJ. Voxel-based morphometry-the methods. Neuroimage 2000;11:805–21. [2] Kakeda S, Korogi Y. The efficacy of a voxel-based morphometry on the analysis of imaging in schizophrenia, temporal lobe epilepsy, and Alzheimer’s disease/mild cognitive impairment: a review. Neuroradiology 2010;52(8):711–21. [3] Draganski B, Gaser C, Busch V, Schuierer G, Bogdahn U, May A. Neuroplasticity: changes in gray matter induced by training. Nature 2004;427:311–2. [4] Noppeney U. The effects of visual deprivation on functional and structural organization of the human brain. Neuroscience and Biobehavioral Reviews 2007;31:1169–80. [5] Pan WJ, Wu G, Li CX, Lin F, Sun J, Lei H. Progressive atrophy in the optic pathway and visual cortex of early blind Chinese adults: a voxel-based morphometry magnetic resonance imaging study. Neuroimage 2007;37(1):212–20. [6] Leporé N, Voss P, Lepore F, et al. Brain structure changes visualized in earlyand late-onset blind subjects. Neuroimage 2010;49(1):134–40. [7] Ptito M, Schneider FC, Paulson OB, Kupers R. Alterations of the visual pathways in congenital blindness. Experimental Brain Research 2008;187(1):41–9. [8] Park H, Lee JD, Kim EY, et al. Morphological alterations in the congenital blind based on the analysis of cortical thickness and surface area. NeuroImage 2009;47:98–106. [9] John Ashburner. A fast diffeomorphic image registration algorithm. NeuroImage 2007;38(1):95–113. [10] Klein A, Andersson J, Ardekani B, et al. Evaluation of 14 nonlinear deformation algorithms applied to human brain MRI registration. Neuroimage 2009;46:786–802. [11] Eickhoff SB, Stephan KE, Mohlberg H, et al. A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data. NeuroImage 2005;25(4):1325–35. [12] Liu Y, Yu C, Liang M, et al. Whole brain functional connectivity in the early blind. Brain 2007;130:2085–96. [13] Shu N, Liu Y, Li J, Li Y, Yu C, Jiang T. Altered anatomical network in early blindness revealed by diffusion tensor tractography. PLoS ONE 2009;4(9):e7228. [14] Stoodley CJ, Schmahmanna JD. Functional topography in the human cerebellum: a meta-analysis of neuroimaging studies. NeuroImage 2009;44(2):489–501.
S. Modi et al. / European Journal of Radiology 81 (2012) 2811–2819 [15] van Broekhoven PCA, Schraa-Tam CKL, van der Lugt A, Smits M, Frens MA, van der Geest JN. Cerebellar contributions to the processing of saccadic errors. Cerebellum 2009, doi:10.1007/s12311-009-0116-6. [16] Jäncke L, Kleinschmidt A, Mirzazade S, Shah NJ, Freund HJ. Role of the inferior parietal cortex in linking the tactile perception and manual construction of object shapes. Cerebral Cortex 2001;11(2):114–21. [17] Naito E, Ehrsson HH. Somatic sensation of hand-object interactive movement is associated with activity in the left inferior parietal cortex. Journal of Neuroscience 2006;26(14):3783–90. [18] Ricciardi E, Pietrini P. New light from the dark: what blindness can teach us about brain function. Current Opinion in Neurology 2011;4(4):357–63. [19] Leporé N, Shi Y, Lepore F, et al. Pattern of hippocampal shape and volume differences in blind subjects. Neuroimage 2009;46(4):949–57.
2819
[20] Hebert AE, Dash PK. Extracellular signal-regulated kinase activity in the entorhinal cortex is necessary for long-term spatial memory. Learning and Memory 2002;9:156–66.
Web references [21] Ashburner J. VBM tutorial. www.fil.ion.ucl.ac.uk/∼john/misc/VBMclass10.pdf. [22] http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0807&L=SPM&P=R7075&1. [23] http://www.fz-juelich.de/inm/inm-1/DE/Forschung/ docs/SPMAnantomy Toolbox/SPMAnantomyToolbox node.html.