Detection of focal changes in human cortical thickness: Spherical wavelets versus Gaussian smoothing

Human in vivo evidence of associations between herpes simplex virus and cerebral amyloid-beta load in normal aging

BACKGROUND

Mounting data suggests that herpes simplex virus type 1 (HSV-1) is involved in the pathogenesis of AD, possibly instigating amyloid-beta (Aβ) accumulation decades before the onset of clinical symptoms. However, human in vivo evidence linking HSV-1 infection to AD pathology is lacking in normal aging, which may contribute to the elucidation of the role of HSV-1 infection as a potential AD risk factor.

METHODS

To shed light into this question, serum anti-HSV IgG levels were correlated with 18F-Florbetaben-PET binding to Aβ deposits and blood markers of neurodegeneration (pTau181 and neurofilament light chain) in cognitively normal older adults. Additionally, we investigated whether associations between anti-HSV IgG and AD markers were more evident in APOE4 carriers.

RESULTS

We showed that increased anti-HSV IgG levels are associated with higher Aβ load in fronto-temporal regions of cognitively normal older adults. Remarkably, these cortical regions exhibited abnormal patterns of resting state-functional connectivity (rs-FC) only in those individuals showing the highest levels of anti-HSV IgG. We further found that positive relationships between anti-HSV IgG levels and Aβ load, particularly in the anterior cingulate cortex, are moderated by the APOE4 genotype, the strongest genetic risk factor for AD. Importantly, anti-HSV IgG levels were unrelated to either subclinical cognitive deficits or to blood markers of neurodegeneration.

CONCLUSIONS

All together, these results suggest that HSV infection is selectively related to cortical Aβ deposition in normal aging, supporting the inclusion of cognitively normal older adults in prospective trials of antimicrobial therapy aimed at decreasing the AD risk in the aging population.

Cortical amyloid-beta burden is associated with changes in intracortical myelin in cognitively normal older adults

Amyloid-beta (Aβ) aggregates and myelin breakdown are among the earliest detrimental effects of Alzheimer's disease (AD), likely inducing abnormal patterns of neuronal communication within cortical networks. However, human in vivo evidence linking Aβ burden, intracortical myelin, and cortical synchronization is lacking in cognitively normal older individuals. Here, we addressed this question combining ¹⁸F-Florbetaben-PET imaging, cortical T1-weigthed/T2-weighted (T1w/T2w) ratio maps, and resting-state functional connectivity (rs-FC) in cognitively unimpaired older adults. Results showed that global Aβ burden was both positively and negatively associated with the T1w/T2w ratio in different cortical territories. Affected cortical regions were further associated with abnormal patterns of rs-FC and with subclinical cognitive deficits. Finally, causal mediation analysis revealed that the negative impact of T1w/T2w ratio in left posterior cingulate cortex on processing speed was driven by Aβ burden. Collectively, these findings provide novel insights into the relationship between initial Aβ plaques and intracortical myelin before the onset of cognitive decline, which may contribute to monitor the efficacy of novel disease-modifying strategies in normal elderly individuals at risk for cognitive impairment.

Effects of non-modifiable risk factors of Alzheimer's disease on intracortical myelin content

BACKGROUND

Non-modifiable risk factors of Alzheimer's disease (AD) have lifelong effects on cortical integrity that could be mitigated if identified at early stages. However, it remains unknown whether cortical microstructure is affected in older individuals with non-modifiable AD risk factors and whether altered cortical tissue integrity produces abnormalities in brain functional networks in this AD-risk population.

METHODS

Using relative T1w/T2w (rT1w/T2w) ratio maps, we have compared tissue integrity of normal-appearing cortical GM between controls and cognitively normal older adults with either APOE4 (N = 50), with a first-degree family history (FH) of AD (N = 52), or with the co-occurrence of both AD risk factors (APOE4+FH) (N = 35). Additionally, individuals with only one risk factor (APOE4 or FH) were combined into one group (N = 102) and compared with controls. The same number of controls matched in age, sex, and years of education was employed for each of these comparisons. Group differences in resting state functional connectivity (rs-FC) patterns were also investigated, using as FC seeds those cortical regions showing significant changes in rT1w/T2w ratios.

RESULTS

Overall, individuals with non-modifiable AD risk factors exhibited significant variations in rT1w/T2w ratios compared to controls, being APOE4 and APOE4+FH at opposite ends of a continuum. The co-occurrence of APOE4 and FH was further accompanied by altered patterns of rs-FC.

CONCLUSIONS

These findings may have practical implications for early detection of cortical abnormalities in older populations with APOE4 and/or FH of AD and open new avenues to monitor changes in cortical tissue integrity associated with non-modifiable AD risk factors.

Linking Plasma Amyloid Beta and Neurofilament Light Chain to Intracortical Myelin Content in Cognitively Normal Older Adults

Evidence suggests that lightly myelinated cortical regions are vulnerable to aging and Alzheimer’s disease (AD). However, it remains unknown whether plasma markers of amyloid and neurodegeneration are related to deficits in intracortical myelin content, and whether this relationship, in turn, is associated with altered patterns of resting-state functional connectivity (rs-FC). To shed light into these questions, plasma levels of amyloid-β fragment 1–42 (Aβ_1–42) and neurofilament light chain (NfL) were measured using ultra-sensitive single-molecule array (Simoa) assays, and the intracortical myelin content was estimated with the ratio T1-weigthed/T2-weighted (T1w/T2w) in 133 cognitively normal older adults. We assessed: (i) whether plasma Aβ_1–42 and/or NfL levels were associated with intracortical myelin content at different cortical depths and (ii) whether cortical regions showing myelin reductions also exhibited altered rs-FC patterns. Surface-based multiple regression analyses revealed that lower plasma Aβ_1–42 and higher plasma NfL were associated with lower myelin content in temporo-parietal-occipital regions and the insular cortex, respectively. Whereas the association with Aβ_1–42 decreased with depth, the NfL-myelin relationship was most evident in the innermost layer. Older individuals with higher plasma NfL levels also exhibited altered rs-FC between the insula and medial orbitofrontal cortex. Together, these findings establish a link between plasma markers of amyloid/neurodegeneration and intracortical myelin content in cognitively normal older adults, and support the role of plasma NfL in boosting aberrant FC patterns of the insular cortex, a central brain hub highly vulnerable to aging and neurodegeneration.

Impaired glucose metabolism reduces the neuroprotective action of adipocytokines in cognitively normal older adults with insulin resistance

Evidence suggests that aging-related dysfunctions of adipose tissue and metabolic disturbances increase the risk of diabetes and metabolic syndrome (MtbS), eventually leading to cognitive impairment and dementia. However, the neuroprotective role of adipocytokines in this process has not been specifically investigated. The present study aims to identify metabolic alterations that may prevent adipocytokines from exerting their neuroprotective action in normal ageing. We hypothesize that neuroprotection may occur under insulin resistance (IR) conditions as long as there are no other metabolic alterations that indirectly impair the action of adipocytokines, such as hyperglycemia. This hypothesis was tested in 239 cognitively normal older adults (149 females) aged 52 to 87 years (67.4 ± 5.9 yr). We assessed whether the homeostasis model assessment-estimated insulin resistance (HOMA-IR) and the presence of different components of MtbS moderated the association of plasma adipocytokines (i.e., adiponectin, leptin and the adiponectin to leptin [Ad/L] ratio) with cognitive functioning and cortical thickness. The results showed that HOMA-IR, circulating triglyceride and glucose levels moderated the neuroprotective effect of adipocytokines. In particular, elevated triglyceride levels reduced the beneficial effect of Ad/L ratio on cognitive functioning in insulin-sensitive individuals; whereas under high IR conditions, it was elevated glucose levels that weakened the association of the Ad/L ratio with cognitive functioning and with cortical thickness of prefrontal regions. Taken together, these findings suggest that the neuroprotective action of adipocytokines is conditioned not only by whether cognitively normal older adults are insulin-sensitive or not, but also by the circulating levels of triglycerides and glucose, respectively.

Salivary lactoferrin is associated with cortical amyloid-beta load, cortical integrity, and memory in aging

BACKGROUND

Aging is associated with declining protective immunity and persistent low-grade inflammatory responses, which significantly contribute to Alzheimer's disease (AD) pathogenesis. Detecting aging-related cerebral vulnerability associated with deterioration of the immune system requires from non-invasive biomarkers able to detect failures in the brain-immunity connection. Reduced levels of salivary lactoferrin (sLF), an iron-binding protein with immunomodulatory activity, have been related to AD diagnosis. However, it remains unknown whether decreased sLF is associated with increased cortical amyloid-beta (Aβ) load and/or with loss of cortical integrity in normal aging.

METHODS

Seventy-four cognitively normal older adults (51 females) participated in the study. We applied multiple linear regression analyses to assess (i) whether sLF is associated with cortical Aβ load measured by 18F-Florbetaben (FBB)-positron emission tomography (PET), (ii) whether sLF-related variations in cortical thickness and cortical glucose metabolism depend on global Aβ burden, and (iii) whether such sLF-related cortical abnormalities moderate the relationship between sLF and cognition.

RESULTS

sLF was negatively associated with Aβ load in parieto-temporal regions. Moreover, sLF was related to thickening of the middle temporal cortex, increased FDG uptake in the posterior cingulate cortex, and poorer memory. These associations were stronger in individuals showing the highest Aβ burden.

CONCLUSIONS

sLF levels are sensitive to variations in cortical Aβ load, structural and metabolic cortical abnormalities, and subclinical memory impairment in asymptomatic older adults. These findings provide support for the use of sLF as a non-invasive biomarker of cerebral vulnerability in the general aging population.

Atrophy of Basal Forebrain Initiates with Tau Pathology in Individuals at Risk for Alzheimer's Disease

Evidence suggests that the basal forebrain (BF) cholinergic system degenerates early in the course of Alzheimer's disease (AD), likely due to the vulnerability of BF cholinergic neurons to tau pathology. However, it remains unclear whether the presence of tauopathy is the only requirement for initiating the BF degeneration in asymptomatic subjects at risk for AD (AR-AD), and how BF structural deficits evolve from normal aging to preclinical and prodromal AD. Here, we provide human in vivo magnetic resonance imaging evidence supporting that abnormal cerebrospinal fluid levels of phosphorylated tau (T+) are selectively associated with bilateral volume loss of the nucleus basalis of Meynert (nbM, Ch4) in AR-AD individuals. Spreading of atrophy to medial septum and vertical limb of diagonal band Broca (Ch1-Ch2) occurred in both preclinical and prodromal AD. With the exception of A+, all groups revealed significant correlations between volume reduction of BF cholinergic compartments and atrophy of their innervated regions. Overall, these results support the central role played by tauopathy in instigating the nbM degeneration in AR-AD individuals and the necessary coexistence of both AD proteinopathies for spreading damage to larger BF territories, thus affecting the core of the BF cholinergic projection system.

Plasma tau predicts cerebral vulnerability in aging

Identifying cerebral vulnerability in late life may help prevent or slow the progression of aging-related chronic diseases. However, non-invasive biomarkers aimed at detecting subclinical cerebral changes in the elderly are lacking. Here, we have examined the potential of plasma total tau (t-tau) for identifying cerebral and cognitive deficits in normal elderly subjects. Patterns of cortical thickness and cortical glucose metabolism were used as outcomes of cerebral vulnerability. We found that increased plasma t-tau levels were associated with widespread reductions of cortical glucose uptake, thinning of the temporal lobe, and memory deficits. Importantly, tau-related reductions of glucose consumption in the orbitofrontal cortex emerged as a determining factor of the relationship between cortical thinning and memory loss. Together, these results support the view that plasma t-tau may serve to identify subclinical cerebral and cognitive deficits in normal aging, allowing detection of individuals at risk for developing aging-related neurodegenerative conditions.

Fast Polynomial Approximation of Heat Kernel Convolution on Manifolds and Its Application to Brain Sulcal and Gyral Graph Pattern Analysis

Heat diffusion has been widely used in brain imaging for surface fairing, mesh regularization and cortical data smoothing. Motivated by diffusion wavelets and convolutional neural networks on graphs, we present a new fast and accurate numerical scheme to solve heat diffusion on surface meshes. This is achieved by approximating the heat kernel convolution using high degree orthogonal polynomials in the spectral domain. We also derive the closed-form expression of the spectral decomposition of the Laplace-Beltrami operator and use it to solve heat diffusion on a manifold for the first time. The proposed fast polynomial approximation scheme avoids solving for the eigenfunctions of the Laplace-Beltrami operator, which is computationally costly for large mesh size, and the numerical instability associated with the finite element method based diffusion solvers. The proposed method is applied in localizing the male and female differences in cortical sulcal and gyral graph patterns obtained from MRI in an innovative way. The MATLAB code is available at http://www.stat.wisc.edu/~mchung/chebyshev.

Blood total antioxidant status is associated with cortical glucose uptake and factors related to accelerated aging

Identifying cerebral vulnerability in late life is of paramount importance to prevent pathological trajectories of aging before the onset of symptoms. Considerable evidence suggests that impaired antioxidant mechanisms are a fingerprint of aging-related conditions, but there is a lack of human research linking total antioxidant capacity (TAC) measured in peripheral blood to in vivo brain changes and other factors featuring accelerated aging. To address this issue, we have assessed in cognitively normal elderly subjects (N = 100) correlations between serum TAC, using the oxygen radical absorbance capacity assay, surface-based cortical thickness, surface-based 18F-fluorodeoxyglucose positron emission tomography cortical uptake, and different factors associated with accelerated aging [i.e., serum homocysteine (HCY), self-reported memory problems, and self-reported patterns of physical activity]. While no relationship was observed between serum TAC and variations in cortical thickness, decreased TAC level was significantly associated with lower FDG uptake in temporal lobes bilaterally. Remarkably, decreased TAC level was linked to increased HCY concentrations, more subjective memory complaints, and lower frequency of physical activity. Overall, our results suggest that decreased serum TAC level may be helpful to detect vulnerable trajectories of aging.

Lower serum expression of miR-181c-5p is associated with increased plasma levels of amyloid-beta 1-40 and cerebral vulnerability in normal aging

Background

Previous studies have shown that expression levels of miR-181c are downregulated by amyloid-β (Aβ) deposition and chronic cerebral hypoperfusion, both factors largely associated with the development of AD. Moreover, reduced 2-[18F]fluoro-2-deoxy-D-glucose (FDG)-PET brain metabolism and volume loss of regions of the medial temporal lobe have been generally recognized as hallmarks of AD. Based on this evidence, we have here investigated potential associations between serum levels of miR-181c-5p and these AD signatures in asymptomatic elderly subjects.

Methods

Ninety-five normal elderly subjects underwent clinical, cognitive, structural MRI, and FDG-PET explorations. Serum expression levels of miR-181c-5p and plasma Aβ concentrations were further analyzed in this cohort. Regression analyses were performed to assess associations between serum miR-181c-5p levels and cognitive functioning, plasma Aβ, structural and metabolic brain changes.

Results

Decreased serum expression of miR-181c-5p was associated with increased plasma levels of Aβ_1–40, deficits in cortical glucose metabolism, and volume reduction of the entorhinal cortex. No significant associations were found between lower miR-181c-5p levels and cognitive deficits or cortical thinning.

Conclusions

These findings suggest that deregulation of serum miR-181c-5p may indicate cerebral vulnerability in late life.

Aging-Related Changes in Cognition and Cortical Integrity are Associated With Serum Expression of Candidate MicroRNAs for Alzheimer Disease

Evidence has shown that microRNAs (miRNAs) are involved in molecular pathways responsible for aging and prevalent aging-related chronic diseases. However, the lack of research linking circulating levels of miRNAs to changes in the aging brain hampers clinical translation. Here, we have investigated if serum expression of brain-enriched miRNAs that have been proposed as potential biomarkers in Alzheimer’s disease (AD) (miR-9, miR-29b, miR-34a, miR-125b, and miR-146a) are also associated with cognitive functioning and changes of the cerebral cortex in normal elderly subjects. Results revealed that candidate miRNAs were linked to changes in cortical thickness (miR-9, miR-29b, miR-34a, and miR-125b), cortical glucose metabolism (miR-29b, miR-125b, and miR-146a), and cognitive performance (miR-9, miR-34a, and miR-125b). While both miR-29b and miR-125b were related to aging-related structural and metabolic cortical changes, only expression levels of miR-125b were associated with patterns of glucose consumption shown by cortical regions that correlated with executive function. Together, these findings suggest that serum expression of AD-related miRNAs are biologically meaningful in aging and may play a role as biomarkers of cerebral vulnerability in late life.

Effects of PER3 clock gene polymorphisms on aging-related changes of the cerebral cortex

Considerable evidence suggests that circadian rhythmicity is progressively disrupted in senescence. Among clock genes, Period3 (PER3) has been associated with circadian phenotypes, homeostatic regulation of sleep, and cognitive performance in young adults. However, the effects of PER3 genotype on aging-related changes in both cognitive function and cortical integrity remain largely unknown. To shed light into this issue, we have investigated differences in cognitive performance, patterns of cortical thickness, and cortical glucose consumption in normal elderly subjects homozygous carriers of the short (PER3^4/4, n = 32) and long repeat alleles (PER3^5/5, n = 32). Relationships between cognitive performance and cortical thickness/metabolism were further explored for each PER3 genotype. We found that PER3^5/5 carriers had poorer cognitive performance (attention, executive function, semantic memory, and verbal fluency) and lower cortical integrity (structural and functional) than PER3^4/4. PER3^5/5 further showed thinning of temporo-parietal areas, and reductions of glucose consumption in fronto-temporo-parietal regions bilaterally. Moreover, PER3^5/5 subjects exhibited significant correlations between decreased glucose metabolism in fronto-parietal regions and poorer cognitive flexibility, though only correlations with lower glucose consumption of the supramarginal gyrus distinguished PER3^5/5 from PER3^4/4 groups. Overall, these findings enhance our understanding on the gene-brain interaction in aging, and may have further implications for the detection of subclinical cognitive decline associated with PER3 genotypes in late life.

Volume Loss of the Nucleus Basalis of Meynert is Associated with Atrophy of Innervated Regions in Mild Cognitive Impairment

Extensive research suggests that basal forebrain (BF) cholinergic neurons are selectively vulnerable to Alzheimer's disease (AD). However, it remains unknown whether volume loss of BF cholinergic compartments parallels structural changes of their innervated regions in prodromal AD. To this aim, we have correlated volume of each BF compartment with cortical thickness and hippocampus/amygdala volume in 106 healthy older (HO) adults and 106 amnestic mild cognitive impairment (aMCI) patients. Correlations were limited to regions affected by atrophy in aMCI. The volume of the nucleus basalis of Meynert (NBM/Ch4) was positively correlated with thickness of the temporal cortex in aMCI, and with volume of amygdala in HO and aMCI, separately. Volume of the medial septum/diagonal band of Broca (Ch1-Ch3) was also positively correlated with volume of the hippocampus within the 2 groups. Only correlations between the NBM and their innervated regions showed diagnostic value. Unlike men, aMCI women showed a stronger association between volume of the NBM and thickness of the temporal lobe when compared with HO women. Altogether, these results reveal, for the first time in humans, that atrophy of NBM is associated with structural changes of their innervated regions in prodromal AD, being this relationship more evident in women.

Increased levels of plasma amyloid-beta are related to cortical thinning and cognitive decline in cognitively normal elderly subjects

Plasma levels of circulating amyloid-beta (Aβ) peptides are of particular interest in Alzheimer' disease, but little is known about cognitive and cortical correlates of peripheral Aβ levels in normal aging. Here, we compared cognitive functioning, vascular risk factors, and patterns of cortical thickness between cognitively intact elderly subjects with low (N = 60) and high (N = 60) plasma Aβ levels (cutoffs: 225 pg/mL and 23 pg/mL for Aβ1-40 and Aβ1-42, respectively). Overall, subjects with high Aβ levels showed lower cognitive performance and thinner cortex than those with low Aβ levels. More specifically, subjects with high Aβ1-40 showed bilateral thinning of the prefrontal cortex, poorer objective memory, slower processing speed, and lower nonverbal reasoning skills, whereas subjects with high Aβ1-42 had thinner temporal lobe, poorer everyday memory, and increased levels of homocysteine. Overall, these results suggest that high plasma Aβ levels in normal elderly subjects are associated with subclinical markers of vulnerable aging, which may be helpful at predicting different trajectories of aging in cognitively intact older adults.

Unified heat kernel regression for diffusion, kernel smoothing and wavelets on manifolds and its application to mandible growth modeling in CT images

We present a novel kernel regression framework for smoothing scalar surface data using the Laplace-Beltrami eigenfunctions. Starting with the heat kernel constructed from the eigenfunctions, we formulate a new bivariate kernel regression framework as a weighted eigenfunction expansion with the heat kernel as the weights. The new kernel method is mathematically equivalent to isotropic heat diffusion, kernel smoothing and recently popular diffusion wavelets. The numerical implementation is validated on a unit sphere using spherical harmonics. As an illustration, the method is applied to characterize the localized growth pattern of mandible surfaces obtained in CT images between ages 0 and 20 by regressing the length of displacement vectors with respect to a surface template.

Sleep deficits in mild cognitive impairment are related to increased levels of plasma amyloid-β and cortical thinning

Evidence suggests that amyloid-beta (Aβ) depositions parallel sleep deficits in Alzheimer's disease (AD). However, it remains unknown whether impaired sleep and changes in plasma Aβ levels are related in amnestic mild cognitive impairment (aMCI) subjects, and whether both markers are further associated with cortical thinning in canonical AD regions. To jointly address this issue, we investigated relationships between changes in physiological sleep and plasma Aβ concentrations in 21 healthy old (HO) adults and 21 aMCI subjects, and further assessed whether these two factors were associated with cortical loss in each group. aMCI, but not HO subjects, showed significant relationships between disrupted slow-wave sleep (SWS) and increased plasma levels of Aβ42. We also found that shortened rapid-eye movement (REM) sleep in aMCI correlated with thinning of the posterior cingulate, precuneus, and postcentral gyrus; whereas higher levels of Aβ40 and Aβ42 accounted for grey matter (GM) loss of posterior cingulate and entorhinal cortex, respectively. These results support preliminary relationships between Aβ burden and altered sleep physiology observed in animal models of AD amyloidosis, and provide precise cortical correlates of these changes in older adults with aMCI. Taken together, these findings open new research avenues on the combined role of sleep, peripheral Aβ levels and cortical integrity in tracking the progression from normal aging to early neurodegeneration.

Improved Statistical Power with a Sparse Shape Model in Detecting an Aging Effect in the Hippocampus and Amygdala

The sparse regression framework has been widely used in medical image processing and analysis. However, it has been rarely used in anatomical studies. We present a sparse shape modeling framework using the Laplace-Beltrami (LB) eigenfunctions of the underlying shape and show its improvement of statistical power. Traditionally, the LB-eigenfunctions are used as a basis for intrinsically representing surface shapes as a form of Fourier descriptors. To reduce high frequency noise, only the first few terms are used in the expansion and higher frequency terms are simply thrown away. However, some lower frequency terms may not necessarily contribute significantly in reconstructing the surfaces. Motivated by this idea, we present a LB-based method to filter out only the significant eigenfunctions by imposing a sparse penalty. For dense anatomical data such as deformation fields on a surface mesh, the sparse regression behaves like a smoothing process, which will reduce the error of incorrectly detecting false negatives. Hence the statistical power improves. The sparse shape model is then applied in investigating the influence of age on amygdala and hippocampus shapes in the normal population. The advantage of the LB sparse framework is demonstrated by showing the increased statistical power.

Focal cortical damage parallels cognitive impairment in minimal hepatic encephalopathy

Little attention has been paid to cortical integrity in patients with minimal hepatic encephalopathy (MHE), although cognitive functions affected in early stages of liver disease are mainly allocated in different neocortical structures. Here we used cortical surface-based analysis techniques to investigate if patterns of cortical thinning accompany the mildest form of HE. To aim this goal, cortical thickness obtained from high-resolution 3T magnetic resonance imaging (MRI) was measured in patients with no MHE (NMHE), MHE, and healthy controls. Further correlation analyses were performed to examine whether scores in the critical flicker frequency (CFF) test, and blood ammonia levels accounted for the loss of cortical integrity in different stages of liver disease. Finally, we assessed group differences in volume of different subcortical regions and their potential relationships with CFF scores/blood ammonia levels. Results showed a focal thinning of the superior temporal cortex and precuneus in MHE patients when compared with NMHE and controls. Relationships between blood ammonia levels and cortical thickness of the calcarine sulcus accounted for impaired visual judgment in patients with MHE when compared to NMHE. Regression analyses between cortical thickness and CFF predicted differences between controls and the two groups of HE patients, but failed to discriminate between patients with NMHE and MHE. Taking together, these findings provide the first report of cortical thinning in MHE patients, and they yield novel insights into the neurobiological basis of cognitive impairment associated with early stages of liver diseases.

Impact of scale space search on age- and gender-related changes in MRI-based cortical morphometry

In magnetic resonance imaging based brain morphometry, Gaussian smoothing is often applied to increase the signal-to-noise ratio and to increase the detection power of statistical parametric maps. However, most existing studies used a single smoothing filter without adequately justifying their choices. In this article, we want to determine the extent for which performing a morphometry analysis using multiple smoothing filters, namely conducting a scale space search, improves or decreases the detection power. We first compared scale space search with single-filter analysis through a simulated population study. The multiple comparisons in our four-dimensional scale space searches were corrected for using a unified P-value approach. Our results illustrate that, compared with a single-filter analysis, a scale space search analysis can properly capture the variations in analysis results caused by variations in smoothing, and more importantly, it can obviously increase the sensitivity for detecting brain morphometric changes. We also show that the cost of an increased critical threshold for conducting a scale space search is very small. In the second experiment, we investigated age and gender effects on cortical volume, thickness, and surface area in 104 normal subjects using scale space search. The obtained results provide a perspective of scale space theory on the morphological changes with age and gender. These results suggest that, in exploratory studies of aging, gender, and disease, conducting a scale space search is essential, if we are to produce a complete description of the structural changes or abnormalities associated with these dimensions.