November 2014, Volume 3 Issue 3

Atypical Structural Brain Asymmetry as a Biomarker for Alzheimer’s Disease

Ien Li1* and Serena McCalla2
Student1, Teacher2: Jericho High School, Jericho, NY
*Corresponding author: ienalice10@gmail.com

Abstract

Alzheimer's disease (AD) pathology is hypothesized to begin decades prior to symptom exhibition, though prognosis in the 'pre-clinical' stage remains elusive due to individual differences in gender, age, cognitive function, education, and genetics. Characterized by pathological hemispheric bias, AD's multifactorial nature necessitates examination of previously stated vulnerability markers with brain asymmetry, rather than investigation on only a few variables alone. Thus, the current study aimed to identify a general trend for AD prediction by collectively analyzing hemispheric symmetry with various demographics and clinical variables. MRI data and demographics, including Mini Mental State Examinations (MMSE) and Clinical Dementia Rating (CDR), were obtained from Open Access Series of Imaging Studies (OASIS) and inputted into MATLAB R2013b for patch-based analysis, assessing brain symmetry with differences in pixel intensities for homologous regions. Females in both the healthy controls (HC) and AD groups were more symmetric than male counterparts. Symmetry decreased in HC with age progression, and was modestly positively correlated with age in AD due to cortical atrophy. Higher MMSEs were associated with increased symmetry, and less education was associated with greater symmetry. AD individuals were more globally asymmetric than HC, and brain symmetry progressively decreased with disease severity. The study demonstrated the potential for evaluation of both brain asymmetry and vulnerability markers as an opportune method for identifying the healthy from the mentally-at-risk, as well as predicting AD onset for early intervention purposes.

Introduction

Alzheimer's disease (AD) is the most common neurodegenerative disease1, afflicting 5.2 million Americans as of 20142. Currently affecting one in nine adults ages 65 and older, it is predicted that the number of people with AD will rise to 7.1 million by 2025 and a striking 13.8 million by 2050 if clinical treatments do not keep pace2.

The pathological hallmarks of AD include the accumulation of atypically folded beta amyloid and tau proteins, resulting in the formation of neuritic plaques and neurofibrillary tangles which disrupt extra- and intra-cellular signaling and induce cell death3, 4. Strikingly, recent research has discovered the described pathologic brain atrophy progresses asymmetrically5 in the human brain, implicating a left-hemisphere bias for the distribution of neuropathology6. More rapid AD-related neurodegeneration in the left cerebral hemisphere has also been observed7, reflecting a more anatomically asymmetrical brain pattern in individuals with AD compared to healthy controls8 (HC).

It is important to note, however, that structural brain asymmetry exists in cognitively-normal individuals as well. The Yakovlevian anticlockwise torque is prevalent throughout the human population, as well as in non-human primates9, 10, characterized by petalia in the right frontal and left occipital lobes11. Functional brain asymmetry, or lateralization, also mirrors interhemispheric structural asymmetry, which is under the influence of various genetic and environmental factors12, 13. Structural brain asymmetry is a multifactorial characteristic influenced by or related to factors such as sexual preference14, genetics10, handedness11, 15, and predictively, neurological disease and disorder16. Illnesses ranging from schizophrenia17, 18 and bipolar disorder19, to dyslexia11, 20, autism21, 22, and notably, AD5, 11, experience asymmetrical anomalies which may influence cognitive capabilities.

Findings of robust hemispheric asymmetry evident in AD have given rise to speculation that abnormalities in laterality may elevate vulnerability to AD pathology6. Compared to healthy controls (HC), AD patients in a recent study exhibited significantly increased asymmetry measures in the temporal, parietal and occipital lobes5. The atypically exaggerated degree of structural asymmetry in the brains of individuals with AD may, hence, serve as a clinical marker for AD. Hemispheric symmetry has previously been incorporated in the identification of various stages of AD progression23, but has not been assessed in conjunction with a panel of other potential vulnerability markers for prediction of AD onset, post-extensive literature review by the author. Studies have indicated a preclinical phase of AD development, ranging from one to two decades, in which pathology may be manifesting latently24. It is during the preclinical period that predictive methods may be conducted to identify AD risk before the onset of cognitive impairment4.

There is currently no single test available for assessing AD risk, as existing biomarkers independently cannot yet sufficiently predict cognitive decline. Current analyses for beta amyloid and tau levels in cerebrospinal fluid remain complicated by high inter- and intra-laboratory variability in measurements, rendering it challenging to determine thresholds for biomarker positivity4. Fluorodeoxyglucose-PET imaging measuring neuronal activity can only indicate AD-related hypometabolism after the emergence of symptoms. Blood biomarker proteins remain lowly expressed in the periphery, where greater levels of overall protein make biomarker detection difficult4. The e4 allele of the apolipoprotein E gene is a critical risk factor for AD and does contribute to pathologic asymmetry in early disease development6, but is only present in 50% of late-onset AD individuals and thus, cannot be used alone for prognosis and diagnosis25. Magnetic resonance imaging (MRI) can detect structural anomalies such as whole brain volume and hippocampal atrophy26, but both the lack of specificity to AD and the high inter-individual variability in brain structure11 prevent its independency in clinical applications. Predictively, it has been proposed that a combination panel of diverse biomarkers may better serve to identify cognitive-normal individuals from the individuals at risk. Integrating information from multiple approaches, in addition to MRI, may better predict cognitive impairment26 and show improved specificity and sensitivity4.    

Despite extensive research, there has yet to be a study to incorporate hemispheric brain asymmetry as a factor in biomarker panel assessments for AD. Thus, the present study sought to identify a general trend for AD risk by utilizing a unique combination panel of vulnerability markers to collectively analyze the association between brain asymmetry and gender, age, cognitive ability, education, and dementia severity in both HC and AD populations. Since female brains are more symmetric than male brains27, speculations were also made regarding an influence of symmetry on differential disease prevalence between genders. The current study furthermore aimed to help clarify the many contradictory findings among symmetry, cognition28, 29, 30, 31, and age32, 33. Elucidation of the relationships among the aforementioned variables may potentially enhance current understandings of the influence of healthy aging or disease pathology on neuroanatomy and cognitive capabilities.  

Materials and Methods

MRI Data and Demographics Acquisition

Preprocessed (i.e. atlas-registered gain field-corrected) T1-weighted magnetization prepared rapid gradient-echo magnetic resonance imaging (MRI) brain scans (n = 416) and demographic measures were obtained with permission from the Open Access Series of Imaging Studies34 (OASIS). One coronal slice prior to skull stripping was used for each subject, with the following imaging parameters: slice thickness of 1.25 mm, 1 x 1 x 1.25 mm isotropic voxel size, flip angle of 10º, matrix size 176 x 176 voxels. Subjects were issued random identification numbers, and full data anonymization was accounted for before OASIS public release34. Scans with poor image quality, exhibiting extensive motion artifacts and noise, were excluded from the released sample. 

Participants

Right-handed subjects (n = 416), ages 18-96 years, were obtained from OASIS (Figure 1). Healthy controls (n = 316) were categorized after clinical assessment, and the remaining patients (n = 100) were clinically diagnosed with very mild, mild, or moderate Alzheimer's disease (AD). Patients with comorbid diseases other than AD potentially contributing to the development of dementia, such as major depression, stroke, or lesions, were excluded from the sample34. Demographics and clinical measurements available included gender, age, Clinical Dementia Ratings (CDR; 0 = no dementia; 0.5 = very mild dementia; 1 = mild dementia; 2 = moderate dementia), education (1 = less than high school graduate; 2 = high school graduate; 3 = college; 4 = college graduate; 5 = beyond college), and Mini Mental State Examination (MMSE) scores. ID numbers were renumbered into a consecutive sequence for computer analysis. Individuals (n = 235) with data available for MMSEs and years of education were used for cognition and education assessment. 

 

Table 1. Subject demographics (gender, age, dementia status) obtained from OASIS. All individuals (n=416) used for the present study were right-handed. Adapted from the Open Access Series of Imaging Studies, 2007.

 

Patch-Based Data Analysis

MR images were inputted into MATLAB R2013b (MathWorks, Natick, MA) to test spatial correlation, rspatial (Equation 1), between pixels of homologous regions between different brain hemispheres for each subject. Assessment of anatomical asymmetry was based on differences in the intensities of homologous pixels in opposite hemispheres. Specifically, pixel intensity indicates differences in proton density. "Patches" or masks with radii of 8 pixels, forming 17 x 17 matrices, were overlaid on top of the brain scans to calculate median pixel intensities of the overlaid regions (Figure 1A). Image patches evaluated pixel intensities one hemisphere at a time before the data from both hemispheres were spatially correlated (Equation 1). The intensities of pixels from one hemisphere were calculated, before the image was flipped to calculate intensities of the homologous pixels in the opposite hemisphere. Based on preliminary investigations, patches with radii of 8 pixels were more accurate than smaller patch sizes, which are greatly skewed by noise, and larger patch sizes, which took averages of larger regions, thus over-generalizing and decreasing in accuracy.

Though other methodologies for brain symmetry assessment exist, such as the use of an asymmetry or laterality index35, previous studies utilizing the index have produced highly inconsistent findings with regards to regions of interest in cognitive disorders35,36. A patch-based approach for the present investigation was thus developed in an attempt to evaluate symmetry using a different method, as well as to identify potentially influential clinical variables and demographics which may be contributing to the heterogeneity of neuroanatomical symmetry outcomes in AD.

 

equation 1

Equation 1

 

Outputted rspatial images were Fisher Transformed (Equation 2, Figure 1B), and then reshaped into individual vectors collated on top of one another in a 416 x 30976 matrix, where each row indicated data of a single subject and each column indicated outputted values for a particular brain region (Figure 1C). The matrix allows for separation of subjects based on different testing parameters. In spatial correlation, x and y are intensities of homologous pixels in opposite hemispheres, and s indicates standard deviation. 

 

equation 2

Equation 2

 

Testing Analyses

Two-sample t-tests compared gender on brain symmetry in HC and in AD populations, separating the total 416 x 30976 matrix by gender. Regions with significantly greater symmetry in each gender were outputted with values of 1, colored white.

Pearson’s correlations correlated all subjects with respective ages, all males with respective ages, and all females with respective ages. Two outputted statistical r-maps depicted regions of positive and negative correlation as values of 1, colored white (Figure 1F).

For each gender, symmetry values were correlated with their respective ages for the age by gender tests. After Fisher transformation (Equation 1), which stabilizes the variance to compare correlations (r) across populations, both outputs for each gender were inputted into the z-score formula and reshaped back into 176 x 176 matrices. The z-score formula (Equation 2), investigating age by gender effects, compares correlations (r') of two variables between populations, where n denotes number of observations in a population.

 

equation 3

Equation 3

 

Pearson’s correlations conducted compared MMSE scores with symmetry values of subjects (n = 235) with available MMSE data. A similar analysis was employed for comparing education with respective symmetry values (n = 235; Table 2). Two maps were outputted for each test, signifying regions of positive and regions of negative correlations for comparison of both trends. CDR scores for all individuals were also tested with Pearson’s correlation to investigate any association with symmetry values that may identify regions of the brain significantly affected by dementia severity (Table 2).

Two-sample t-tests compared symmetry scores between subjects with very mild AD (CDR = 0.5; n = 70) and mild AD (CDR = 1; n = 28). Patients with moderate AD (CDR = 2) were excluded from analyses due to lack of sufficient sample size (n = 2). T-tests subsequently compared brain symmetry between HC and AD groups overall (Table 2).

 

Table 2. Experimentation tests with parameters. Two-sample t-tests, Pearson’s correlations, and z-score tests were used for HC, AD, and HC & AD. Male, female, and combined image significance thresholds (mIST; fIST; cIST) are shown.

table 2

 

All statistical maps outputted were subject to appropriate thresholds of p-value significance (p < .05) based upon specific testing parameters and sample sizes (p < .05; Table 3). As intensity values for black pixels and white pixels are 0 and 1 respectively, pixel values were summed in each outputted image and graphed. Graphed data for the 2-sample t-tests indicated which group was more symmetric and the bar graphs for the Pearson’s correlations indicated which correlation was stronger as determined by the magnitude of the outputted r values.

 

Figure 1. Flow chart of data analysis.

 

Results

Gender and Age Differences

In HC subjects, two sample t-tests revealed that females exhibit more brain symmetry than males (Figure 2), with summed pixel values for the male and female statistical maps indicated in Table 3. Symmetry was also under higher age control in males than females (Table 3), indicating that the effect of age on structural brain symmetry was greater in males (Figure 3).

 

Table 3. Comparisons of brain symmetry between gender and age in HC and AD. Sum of white (intensity value = 1) pixels in each resulting statistical map post-threshold (p < .05) are indicated to assess gender and age differences in brain symmetry. ‘Pos’ and ‘Neg’ denote the sum of intensity values for positive and negative correlation analyses.

 

Figure 2. Comparison between HC and AD brain symmetry and gender. Females show more regions of symmetry than males in HC and AD, indicating similar regional gender differences in brain structure (p < .05).

 

Figure 3. Z-scores of correlation between symmetry and age. Regions where correlation is greater in males than in females is shown in ‘Male,’ and vice versa in ‘Female’ (p < .05). For both HC and AD groups, male brain symmetry was more sensitive to age than female brain symmetry.

 

When age was investigated across male, female, and combined male and female HC groups, there was a greater inverse than positive correlation between age and brain symmetry in all but the males-only group (Figure 4; Table 3). The AD groups exhibited fewer regions of significant correlation with aging, but showed a more positive than negative relationship (Figure 5; Table 3).

 

Figure 4. Pearson’s correlations between brain symmetry and age in HC group. Regions of negative correlation are more prevalent when testing all (‘All’) HC subjects and HC females (‘F’), except in HC males (‘M’). Colored images denote pixel significance (p < .05; dark red = positive correlation; dark blue = negative correlation).

 

Figure 5. Pearson’s correlations between brain symmetry and age in AD group. Lack of decreased symmetry, and more modest correlations, in AD ‘All’, females (‘F’) and males (‘M’) groups may have resulted from AD pathology skewing age-symmetry relationships. Colored images denote pixel significance (p < .05; dark red = positive correlation; dark blue = negative correlation).

 

MMSE and Education Differences
Pearson’s correlations between brain symmetry and MMSE scores revealed a greater positive than negative correlation between HC global cognitive function and anatomical symmetry (Figure 6; Table 4). Symmetry and education in the HC sample were inversely correlated (Figure 7), with summed pixel values for positive and negative correlation denoted in Table 4. 

 

Table 4. Comparisons of brain symmetry between MMSE and education in HC and AD. Sum of white (intensity value = 1) pixels in each resulting statistical map post-threshold (p < .05) are indicated to assess cognitive and education differences in brain symmetry. ‘Positive’ and ‘Negative’ refer to correlation direction.

table 4

 

Figure 6. Pearson’s correlations between symmetry and MMSE. Regions of positive correlation are more prevalent than those of negative correlation in HC and AD (p < .05).

 

Figure 7. Pearson’s correlations between symmetry and education level. Regions of negative correlation are more prevalent than regions of positive correlation (p < .05).

 

HC and AD Differences
T-tests comparing brain symmetry between HC (CDR = 0) and AD individuals overall revealed significantly greater hemispheric symmetry in HC brains (Table 5; Figure 8). Comparisons of brain symmetry between patients with very mild AD (CDR = 0.5) and patients with mild AD (CDR = 1) furthermore indicated greater symmetry with very mild AD relative to mild AD, implicating a continuous brain symmetry reduction with disease progression (Table 5; Figure 9).

 

Table 5. Comparisons of brain symmetry between CDR, and HC and AD. Sum of white (intensity value = 1) pixels in each resulting statistical map post-threshold (p < .05) are indicated to assess differences in brain symmetry based on dementia severity, if present.

 

Figure 8. Two-sample t-test between symmetry in HC and AD. Regions of hemispheric symmetry in brain are more prevalent in HC (n = 316) than in AD (n = 100; p < .05).

 

Figure 9. Two-sample t-test between symmetry and CDR. Regions where AD subjects with very mild AD (CDR = 0.5; n = 70), and AD subjects with mild AD (CDR = 1; n = 28) differ in symmetry (p < .05).

 

Discussion

Gender and Age Differences

Paralleling the findings of various previous investigations, the current study found, in both HC and AD groups, greater brain symmetry in female than in male brains. The greater female than male symmetry observed in AD may be due to a more symmetric baseline. As gross hemispheric cerebral asymmetry is considered a sexually dimorphic trait11, 27, a plausible explanation attributes increased levels of fetal testosterone to the greater male brain asymmetry observed; fetal testosterone may prevent cell apoptosis, and modulate both neural connectivity and development, rendering differential testosterone exposure between genders to be a potential contributor to sexual dimorphism in morphology and lateralization37, 38. Though the underlying mechanisms of the laterality effects of AD are not well characterized, it may be that abnormally increased degrees of brain asymmetry in females—the more neuroanatomically and functionally symmetric gender—are important indicators of interhemispheric imbalance which may contribute to vulnerability to cognitive decline.   

The greater age sensitivity of male cerebral symmetry exhibited confirms that gender modulates the effects of age on the brain39. The finding that symmetry fluctuates more easily with male aging supports that age-associated brain atrophy may often be more pronounced in males than in females32.

The modestly positive correlations detected between age and symmetry in AD may indicate a lack of global decreased symmetry due to the asymmetrical pathological effects of the neurodegenerative disease, which is implicated to increase brain asymmetry over time with asymmetric atrophy5. A decreased symmetrical pattern with healthy age progression has been previously implicated in recent studies with similar sample age distributions40, though future investigations should seek to assess symmetry of individual brain structures, as the magnitude and direction of asymmetry in one region may differ from another27. Because contradictory findings exist with regards to the relationship between age and asymmetry, with some studies observing increased symmetry32 or no significant symmetry change with age41, both global and regional asymmetries should be delved into further to explore their functional and clinical significance.  

MMSE and Education Differences

The positive correlation between MMSE scores and symmetry in both HC and AD groups is supportive of previous findings of less lateralization associated with better cognitive ability42. According to one cognitive model, older adults may exhibit more symmetric functional activity due to the recruitment of additional neural units from both hemispheres, as opposed to only one hemisphere, to compensate for age-related neural deterioration and augment cognitive performance32. As functional brain asymmetry mirrors structural asymmetry12, mechanisms responsible for more anatomically symmetric female brains may underlie the observation from a recent study of better female language performance than that of males42. Nevertheless, it is well known that there may be an innate female superiority in left-hemisphere linguistic skills and male dominance in right-hemisphere visuospatial abilities, which must also be accounted for11, 38. Extensive research has even demonstrated faster language development in females37, though complex language processes tend to require both hemispheres in females as opposed to just one hemisphere in males. As MMSEs test global cognitive function, and language is a cognitive domain characteristically impaired in AD43, the asymmetric trajectory of AD pathology may impact relatively symmetric female brains more severely than male brains, and contribute to the disproportionately more severe symptoms in females3.

According to a comprehensive recent review, roughly two-thirds of AD-afflicted Americans are women3. Multiple factors have been suggested which may help explain the increased female prevalence because higher female longevity alone cannot account for such disproportions44; as sex hormones provide neuroprotective effects on the brain, the sharp loss of estrogen post-menopause may render females more susceptible to neurodegeneration and cognitive decline3, 44. Combined with higher prevalence of autoimmune diseases and depression3, many variables in play may be responsible for the increased severity of cognitive deficits in females. Future research should confirm whether women typically score lower on MMSEs and whether cognitive performance may be associated with abnormal laterality effects.

The observed inverse correlation between education and cerebral symmetry may seem contradictory to the aforementioned positive relationship between general cognitive ability and symmetry, but it may not be so. It is true that in the same dataset used for the current study, older adults with AD experienced fewer years of education relative to their HC counterparts34. Yet years or level of education do not necessarily indicate one’s cognitive abilities, as it is proposed that only when education reflects cognitive capacity and is assessed in conjunction with other vulnerability factors can the influence of education on dementia be evaluated43. Furthermore, increasing gyral pattern asymmetry has been discovered from healthy childhood through adulthood46, while disrupted learning capabilities for language in particular have been associated with weak functional lateralization which, as mentioned, indicates abnormally increased structural symmetry20.  

HC and AD Differences

Comparing total HC and AD individuals revealed decreased hemispheric symmetry in AD patients, an observation in keeping with recent neuroimaging findings of asymmetric AD pathology distribution between hemispheres6, 8. It remains unclear why the left hemisphere is more vulnerable to AD-related neurodegeneration, although it is proposed that the e4 allele of the apolipoprotein E gene (APOE-4) may play a role in the asymmetry of pathological effects on the brain during early AD development6. Though genetic data was not available for analysis in the current study, the APOE-4 allele, one notable risk factor for late-life AD onset, also putatively affects early brain growth and future neurodegenerative processes, which may be an important variable for investigation6. Hence, combining neuroimaging data with genotyping may reveal associations with functional specialization and AD symptomology.

The decreased brain symmetry in mild AD compared to in very mild AD indicates a continuous reduction in anatomical symmetry with disease progression to greater severity. Though it is known that AD-related brain asymmetry can mark different phases of disease23, it remains to be confirmed whether symmetry can indicate latent AD pathology during the ten to twenty year preclinical period before the onset of symptoms4, 24. Therefore, incorporating brain asymmetry assessment in current combination panels of vulnerability markers may assist in early detection prior to cognitive decline. At present, clinical diagnoses may be conducted through neurological and neuropsychological assessment24, though definitive AD diagnosis may only be determined postmortem with the observation of neuritic plaques and neurofibrillary tangles4. Recognizing the dynamics of AD biomarkers, such as anatomical hemispheric asymmetry, with illness progression may be vital for detection, prevention, and treatment26.

Although the present study exhibits several strengths, including a large sample size, some limitations must also be addressed. It was not possible to investigate the effects of measures such as handedness, genotype, and family history of AD, due to lack of available data in the latter and lack of left-handed individuals in the OASIS dataset. Structural symmetry measures in specific brain regions, in addition to global measures, should also be evaluated, such as regional cortical thicknesses33, and volumes of the hippocampus47 and ventricles11.

The implications of bilateral asymmetry in neuroimaging are extensive, with the potential to uncover how abnormalities in cerebral structures affect the neurocognitive skills which depend on them. In AD, one of many diseases that modulate brain symmetry, it was found that female brains were still more symmetric than male brains, and that AD-related brain atrophy disrupts the normal reduction of symmetry with aging. The exaggeration of healthy brain asymmetry due to AD pathology may hinder global cognitive ability, and greater years of education may induce functional lateralization of cognitive function33, thus contributing to increased anatomical asymmetry. The progressive reduction of symmetry with disease severity may be indicative of more extensive atrophy in the left hemisphere, and serve as a clinical marker for disease stage and trajectory. As it is implicated that models combining data from multiple modalities increases both sensitivity and specificity of detection and prognosis26, the collective analysis of gender, age, cognition, education, AD severity, and structural brain asymmetry may help facilitate the identification of the healthy from the mentally-at-risk. The presented new panel of vulnerability markers may, perhaps, be combined with genotyping and additional aforementioned variables to develop a more effective predictive method for AD onset.

References

  1. Swerdlow, R. H. (2011). Brain aging, Alzheimer's disease, and mitochondria. Biochimica et Biophysica Acta, 1812, 1630-1639. http://dx.doi.org/10.1016/j.bbadis.2011.08.012

  2. Alzheimer's Association. (2014). Alzheimer's & Dementia: 2014 Alzheimer's disease facts and figures (J. Gaugler, B. James, T. Johnson, K. Scholz, & J. Weuve, Comps.). Author.

  3. Carter, C. L., Resnick, E. M., Mallampalli, M., & Kalbarczyk, A. (2012). Sex and gender differences in Alzheimer’s Disease: Recommendations for future research. Journal of Women's Health, 21(10), 1018-1023. http://dx.doi.org/10.1089/jwh.2012.3789

  4. Sutphen, C. L., Fagan, A. M., & Holtzman, D. M. (2014). Progress update: Fluid and imaging biomarkers in Alzheimer’s disease. Biological Psychiatry, 75, 520-526. http://dx.doi.org/10.1016/j.biopsych.2013.07.031

  5. Derflinger, S., Sorg, C., Gaser, C., Myers, N., Arsic, M., Kurz, A., . . . Mühlau, M. (2011). Grey-matter atrophy in Alzheimer's Disease is asymmetric but not lateralized. Journal of Alzheimer's Disease, 25, 347-357. http://dx.doi.org/10.3233/JAD-2011-110041

  6. Donix, M., Burggren, A. S., Scharf, M., Marschner, K., Sunthana, N. A., Siddarth, P., . . . Bookheimer, S. Y. (2013). APOE associated hemispheric asymmetry of entorhinal cortical thickness in aging and Alzheimer's disease. Psychiatry Research: Neuroimaging, 214(3), 212-220. http://dx.doi.org/10.1016/j.pscychresns.2013.09.006

  7. Thompson, P. M., Hayashi, K. M., de Zubicaray, G., Janke, A. L., Rose, S. E., Semple, J., . . . Toga, A. W. (2003). Dynamics of gray matter loss in Alzheimer’s disease. The Journal of Neuroscience, 23(3), 994-1005.

  8. Heckemann, R. A., Keihaninejad, S., Aljabar, P., Gray, K. R., Nielson, C., Rueckert, D., . . . The Alzheimer's Disease Neuroimaging Initiative 2. (2011). Automatic morphometry in Alzheimer's disease and mild cognitive impairment. NeuroImage, 56(4), 2024-2037. http://dx.doi.org/10.1016/j.neuroimage.2011.03.014

  9. Balzeau, A., Gilissen, E., & Grimaud-Hervé, D. (2012). Shared pattern of endocranial shape asymmetries among great apes, anatomically modern humans, and fossil hominins. PLoS ONE, 7(1), e29581. http://dx.doi.org/10.1371/journal.pone.0029581

  10. Corballis, M. C. (2014). Left brain, right brain: Facts and fantasies. PLoS Biology, 12(1). http://dx.doi.org/10.1371/journal.pbio.1001767

  11. Toga, A. W., & Thompson, P. M. (2003). Mapping brain asymmetry. Nature Reviews: Neuroscience, 4, 37-48. http://dx.doi.org/10.1038/nrn1009

  12. Thompson, P., Cannon, T. D., Narr, K. L., Erp, T. V., Poutanen, V.-P., Huttunen, M., . . . Toga, A. W. (2001). Genetic influences on brain structure. Nature Neuroscience, 4(12), 1253-1258. http://dx.doi.org/10.1038/nn758

  13. Ocklenburg, S., & Güntürkün, O. (2012). Hemispheric asymmetries: The comparative view. Frontiers in Psychology, 3(5), 1-9. http://dx.doi.org/10.3389/fpsyg.2012.00005

  14. Savic, I., & Lindström, P. (2008). PET and MRI show differences in cerebral asymmetry and functional connectivity between homo- and heterosexual subjects. Proceedings of the National Academy of Sciences, 105(27), 9403-9408. http://dx.doi.org/10.1073pnas.0801566105

  15. Corballis, M. C. (2013). Early signs of brain asymmetry. Trends in Cognitive Sciences, 17(11), 554-555. http://dx.doi.org/10.1016/j.tics.2013.09.008

  16. Sun, T., & Walsh, C. A. (2006). Molecular approaches to brain asymmetry and handedness. Nature Reviews: Neuroscience, 7, 655-662. http://dx.doi.org/10.1038/nrn1930

  17. Pepe, A., Zhao, L., Koikkalainen, J., Hietala, J., Ruotsalainen, U., & Tohka, J. (2013). Automatic statistical shape analysis of cerebral asymmetry in 3D T1-weighted magnetic resonance images at vertex-level: Application to neuroleptic-naïve schizophrenia. Magnetic Resonance Imaging, 31, 676-687. http://dx.doi.org/10.1016/j.mri.2012.10.021

  18. Francis, A. N., Seidman, L. J., Jabbar, G. A., Mesholam-Gately, R., Thermenos, H. W., Juelich, R., . . . LeDisi, L. E. (2012). Alterations in brain structures underlying language function in young adults at high familial risk for schizophrenia. Schizophrenia Research, 141, 65-71. http://dx.doi.org/10.1016/j.schres.2012.07.015

  19. Crow, T. J., Chance, S. A., Priddle, T. H., Radua, J., & James, A. C. (2013). Laterality interacts with sex across the schizophrenia/bipolarity continuum: An interpretation of meta-analyses of structural MRI. Psychiatry Research, 210, 1232-1244. http://dx.doi.org/10.1016/j.psychres.2013.07.043

  20. Bishop, D. V. M. (2013). Cerebral asymmetry and language development: Cause, correlate, or consequence? Science, 340(6138), 1302-1310. http://dx.doi.org/10.1126/science.1230531

  21. Wan, C. Y., Marchina, S., Norton, A., & Schlaug, G. (2012). Atypical hemispheric asymmetry in the arcuate fasciculus of completely nonverbal children with autism. Annals of the New York Academy of Sciences, 1252, 332-337. http://dx.doi.org/10.1111/j.1749-6632.2012.06446.x

  22. Gage, N. M., Juranek, J., Filipek, P. A., Osann, K., Flodman, P., Isenberg, A. L., & Spence, M. A. (2009). Rightward hemispheric asymmetries in auditory language cortex in children with autistic disorder: An MRI investigation. Journal of Neurodevelopmental Disorders, 1, 205-214. http://dx.doi.org/10.1007/s11689-009-9010-2

  23. Stefanits, H., Budka, H., & Kovacs, G. G. (2012). Asymmetry of neurodegenerative disease-related pathologies: A cautionary note. Acta Neuropathologica, 123(3), 449-452. http://dx.doi.org/10.1007/s00401-011-0936-6

  24. Fuentes, P. (2012). Clinical diagnosis in preclinical stage of Alzheimer's disease. Archives of Medical Research, 43(8), 667-670. http://dx.doi.org/10.1016/j.arcmed.2012.10.014

  25. Verghese, P. B., Castellano, J. M., & Holtzman, D. M. (2011). Apolipoprotein E in Alzheimer's disease and other neurological disorders. The Lancet Neurology, 10(3), 241-252. http://dx.doi.org/10.1016/S1474-4422(10)70325-2

  26. Braskie, M. N., & Thompson, P. M. (2014). A focus on structural brain imaging in the Alzheimer’s Disease Neuroimaging Initiative. Biological Psychiatry, 75(7), 527-533. http://dx.doi.org/10.1016/j.biopsych.2013.11.020

  27. Rentería, M. E. (2012). Cerebral asymmetry: A quantitative, multifactorial, and plastic brain phenotype. Twin Research and Human Genetics, 15(3), 401-413. http://dx.doi.org/10.1017/thg.2012.13

  28. Corballis, M. C. (2009). The evolution and genetics of cerebral asymmetry. Philosophical Transactions of the Royal Society, 364, 867-879. http://dx.doi.org/10.1098/rstb.2008.0232

  29. Hirnstein, M., Hausmann, M., & Guntürkün, O. (2008). The evolutionary origins of functional cerebral asymmetries in humans: Does lateralization enhance parallel processing? Behavioral Brain Research, 187, 297-303. http://dx.doi.org/10.1016/j.bbr.2007.09.023

  30. Pascual, A., Huang, K.-L., Neveu, J., & Preat, T. (2004). Brain asymmetry and long-term memory. Nature: Brief Communications, 427, 605-606.

  31. Witelson, S. F., Kigar, D. L., & Harvey, T. (1999). The exceptional brain of Albert Einstein. Lancet, 353, 2149-2153.

  32. Cabeza, R. (2002). Hemispheric asymmetry reduction in older adults: The HAROLD model. Psychology and Aging, 17(1), 85-100. http://dx.doi.org/10.1037/0882-7974.17.1.85

  33. Zhou, D., Lebel, C., Evans, A., & Beaulieu, C. (2013). Cortical thickness asymmetry from childhood to older adulthood. NeuroImage, 83, 66-74. http://dx.doi.org/10.1016/j.neuroimage.2013.06.073

  34. Marcus, D. S., Wang, T. H., Parker, J., Csernansky, J. G., Morris, J. C., & Buckner, R. L. (2007). Open Access Series of Imaging Studies (OASIS): Cross-sectional MRI data in young, middle aged, nondemented, and demented older adults. Journal of Cognitive Neuroscience, 19(9), 1498-1507.

  35. Cherbuin, N., Réglade-Meslin, C., Kumar, R., Sachdev, P., & Anstey, K. J. (2010). Mild cognitive disorders are associated with different patterns of brain asymmetry than normal aging: The PATH through life study. Frontiers in Psychiatry, 1. http://dx.doi.org/10.3389/fpsyt.2010.00011

  36. Kim, J. H., Lee, J. W., Kim, G. H., Roh, J. H., Kim, M.-J., Seo, S. W., . . . Na, D. L. (2012). Cortical asymmetries in normal, mild cognitive impairment, and Alzheimer's disease. Neurobiology of Aging, 33, 1959-1966. http://dx.doi.org/10.1016/j.neurobiolaging.2011.06.026

  37. Chura, L. R., Lombardo, M. V., Ashwin, E., Auyeung, B., Chakrabarti, B., Bullmore, E. T., & Baron-Cohen, S. (2010). Organizational effects of fetal testosterone on human corpus callosum size and asymmetry. Psychoneuroendocrinology, 35, 122-132. http://dx.doi.org/10.1016/j.psyneuen.2009.09.009

  38. Tomasi, D., & Volkow, N. D. (2012). Laterality patterns of brain functional connectivity: Gender effects. Cerebral Cortex, 22, 1455-1462. http://dx.doi.org/10.1093/cercor/bhr230

  39. Ingalhalikar, M., Smith, A., Parker, D., Satterthwaite, T. D., Elliott, M. A., Ruparel, K., . . . Verma, R. (2014). Sex differences in the structural connectome of the human brain. Proceedings of the National Academy of Sciences, 111(2), 823-828. http://dx.doi.org/10.1073/pnas.1316909110

  40. Kovalev, V. A., Kruggel, F., & von Cramon, D. Y. (2003). Gender and age effects in structural brain asymmetry as measured by MRI texture analysis. NeuroImage, 19, 895-905. http://dx.doi.org/10.1016/S1053-8119(03)00140-X

  41. Brain Development Cooperative Group. (2012). Total and regional brain volumes in a population-based normative sample from 4 to 18 years: The NIH MRI study of normal brain development. Cerebral Cortex, 22(1), 1-12. http://dx.doi.org/10.1093/cercor/bhr018

  42. Catani, M., Allin, M. P. G., Husain, M., Pugliese, L., Mesulam, M. M., Murray, R. M., & Jones, D. K. (2007). Symmetries in human brain language pathways correlate with verbal recall. Proceedings of the National Academy of Sciences, 104(43), 17163-17168. http://dx.doi.org/10.1073/pnas.0702116104

  43. Albert, M. S., DeKosky, S. T., Dickson, D., Dubois, B., Feldman, H. H., Fox, N. C., . . . Phelps, C. H. (2011). The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimer's & Dementia, 7(3), 270-279. http://dx.doi.org/10.1016/j.jalz.2011.03.008

  44. Viña, J., & Lloret, A. (2010). Why women have more Alzheimer's Disease than men: Gender and mitochondrial toxicity of amyloid-β peptide. Journal of Alzheimer's Disease, 20, S527-S533. http://dx.doi.org/10.3233/JAD-2010-100501

  45. Sharp, E. S., & Gatz, M. (2011). The relationship between education and dementia an updated systematic review. Alzheimer Disease and Associated Disorders, 25(4), 289-304. http://dx.doi.org/10.1097/WAD.0b013e318211c83c

  46. Toga, A. W., Thompson, P. M., & Sowell, E. R. (2006). Mapping brain maturation. Trends in Neuroscience, 29(3), 148-159. http://dx.doi.org/10.1016/j.tins.2006.01.007

  47. Woolard, A. A., & Heckers, S. (2012). Anatomical and functional correlates of human hippocampal volume asymmetry. Psychiatry Research: Neuroimaging, 201, 48-53. http://dx.doi.org/10.1016/j.psychresns.2011.07.016

Acknowledgements

The authors would like to thank Mr. Brian D. Cummings for the availability of MATLAB R2013b for project experimentation, as well as Ms. Leighanne Dunckley for providing technical support. The authors would also like to acknowledge Dr. Krista L. Moulder for her assistance with data acquirement from the Open Access Series of Imaging Studies.