Paul Yates

Cerebral microhemorrhage and brain β-amyloid in aging and Alzheimer disease.

Objectives: Incidental cerebral microhemorrhage (MH) is frequently found in older individuals scanned with susceptibility-weighted MRI (SWI) or gradient-recalled echo MRI. MH have been linked with β-amyloid (Aβ) deposition using 11C-Pittsburgh compound B (PiB) PET in Alzheimer disease (AD) and cerebral amyloid angiopathy (CAA). We hypothesized that Aβ deposition in asymptomatic elderly individuals is associated with lobar MH (LMH). Methods: This was a cross-sectional study of 84 elderly healthy controls (HC), 28 subjects with mild cognitive impairment (MCI), and 26 subjects with probable AD who underwent 3-T SWI and 11C-PiB PET. 11C-PiB cortical binding was quantified normalized to cerebellar cortex (standardized uptake value ratio [SUVR]) and scans classified as positive (PiB+) or negative (PiB−) by visual inspection. MH were manually counted and categorized by region and as lobar or nonlobar. Results: LMH were present in 30.8% of AD, 35.7% of MCI, and 19.1% of HC. The prevalence of LMH among PiB+ subjects was similar, regardless of clinical classification (AD 30.8%, MCI 38.9%, HC 41.4%, p > 0.7). HC with LMH had significantly higher mean neocortical SUVR (1.7 ± 0.5) than HC without LMH (1.3 ± 0.3, p ± 0.01). In HC, there was a positive correlation between number of LMH and SUVR, and between LMH and age. In HC, PiB+ (odds ratio [OR] 7.3, 95% confidence interval [CI] 1.6–33.7, p = 0.01) and age (OR 1.2, 95% CI 1.03–1.3, p = 0.02) both independently predicted the occurrence of LMH using logistic regression. Conclusion: Asymptomatic Aβ deposition in older adults is strongly associated with LMH.

Incidence of cerebral microbleeds in preclinical Alzheimer disease

Objective: We sought to determine the incidence and associations of lobar microbleeds (LMBs) in a longitudinal cohort with 11C–Pittsburgh compound B (PiB) PET imaging. Methods: One hundred seventy-four participants from the observational Australian Imaging, Biomarkers and Lifestyle Study of Ageing (97 with normal cognition [NC], 37 with mild cognitive impairment [MCI], and 40 with Alzheimer disease [AD] dementia) were assessed at 3 time points over 3 years with 3-tesla susceptibility-weighted MRI and 11C-PiB PET. MRIs were inspected for microbleeds, siderosis, infarction, and white matter hyperintensity severity, blind to clinical and PiB findings. Neocortical PiB standardized uptake value ratio, normalized to cerebellar cortex, was dichotomized as positive or negative (PiB+/−, standardized uptake value ratio >1.5). Annualized LMB incidence was calculated, and logistic regression was used to determine the association of incident LMBs with PiB, APOE ε4+ status, and cerebrovascular disease. Results: LMBs were present in 18.6% of NC, 24.3% of MCI, and 40% of AD participants (p < 0.05 vs NC). LMB incidence was 0.2 ± 0.6 per year in NC participants, 0.2 ± 0.5 in MCI, and 0.7 ± 1.4 in AD (p < 0.03 vs NC) and was 6-fold higher in PiB+ than PiB-NC. Incident LMBs were associated with age, APOE ε4+, PiB+, and baseline LMBs. Incidence of multiple LMBs was also associated with lacunar infarction and white matter hyperintensity severity. Conclusions: Older age, baseline LMBs, higher β-amyloid burden, and concomitant cerebrovascular disease may all confer higher risk of incident LMBs. This should be considered when designing protocols for amyloid-modifying clinical trials.

Automatic detection of small spherical lesions using multiscale approach in 3D medical images

Automated detection of small, low level shapes such as circular/spherical objects in images is a challenging computer vision problem. For many applications, especially microbleed detection in Alzheimers disease, an automatic pre-screening scheme is required to identify potential seeds with high sensitivity and reasonable specificity. A new method is proposed to detect spherical objects in 3D medical images within the multi-scale Laplacian of Gaussian framework. The major contributions are(1)breaking down 3D sphere detection into 1D line profile detection along each coordinate dimension, (2) identifying center of structures bynormalizing the line response profile and (3) employing eigenvalues of the Hessian matrix at optimum scale for the center points to determine spherical objects. The method is validated both on simulated data and susceptibility weighted MRI images with ground truth provided by a medical expert. Validation results demonstrate that the current approach has higher performance in terms of sensitivity and specificity and is effective in detecting adjacent microbleeds, with invariance to intensity, orientation, translation and object scale.

The effect of vascular risk factors on cognition in older adults: Data from the AIBL study

version to Alzheimer’s disease (AD). Partially ordered sets (POSETS) are classification models, which are useful for representing the inherent relationships between specific cognitive functions and neuropsychological (NP) measures. This allows for statistical linkage of response patterns to cognitive profiles of functioning that can be used to understand how different cognitive functions serve as markers for AD conversion. Methods: To identify subgroups within MCI with high likelihood of conversion to AD within 24 months by analyzing cognitive data from the Alzheimer’s disease cooperative study (ADCS) MCI trial through application of POSETmethodology. We have initially considered 513MCI patients that were followed for 24 months. Response patterns to selected NP measures are linked to cognitive profiles of functioning using POSETs, which are then clustered and stratified by APoE4 status. Results: Unsurprisingly, among cognitive functions considered, impairment with episode memory (EM) is most associated with conversion within 24 months. Functioning level is determined by classification to profiles and strength of classification, as reflected by Bayesian posterior probabilities. The conversion rate of those with lower levels of EM and APOE-ε4 is 65.2%, E.g. lower level functioning is characterized by relatively poor performance on the ADAS-Cog delayed recall, Logical Memory, and NYU Paragraph. In contrast, for those ApoE4 subjects who are not at a lower level of EM functioning (e.g. perform relatively well on ADAS-Cog delayed recall) are at a significantly lower conversion rate (23.2%; Fisher’s exact test p-value < 0.001). Moreover, for those with higher levels of EM functioning (e.g. perform relatively well on the delayed recall measures) and without APOE-ε4, the observed conversion rate is only 7.8%. In an ROC analysis, the AUC for EM functioning level is 0.789. Other cognitive functions such as cognitive flexibility, working memory, processing speed, and word fluency do not appear to be significantly related to higher conversion rates. Conclusions: EM functioning and APOE-ε4 status are strongly related to MCI conversion to AD within 24 months. This corroborates the great heterogeneity in conversion rates found from a POSET-based analysis of ADNI cognitive data.

PARTICIPANTS WHO REFUSE PET SCAN HAD SIGNIFICANTLY LOWER EPISODIC MEMORY THAN THOSE WHO CONSENTED TO SCAN: DATA FROM THE WOMEN’S HEALTHY AGEING PROJECT

risks of developing Alzheimer’s disease (AD). The objective of the present study was to identify which condition in cognitively normal (CN) individuals is the most associated with an AD-like pattern of neurodegeneration. Methods: For this purpose, we compared structural-MRI and FDG-PET data in i) a group of 15 CN individuals with SCD recruited from a memory clinic to 45 matched controls without SCD recruited from the general population, and ii) a group of 10 CN individuals with amyloidpositive (Ab+) Florbetapir-PET scan to 29 matched controls with amyloid-negative scan. SCD and Ab+ individuals were matched for age, education, sex and MMSE. Results: Regarding structural-MRI, highly significant gray matter atrophy in the hippocampal region was found in the SCD compared to the controls with no SCD, and this atrophy correlated with episodic memory decline. Note that this result was not related to amyloid deposition since both groups did not differ in the proportion of amyloid positive individuals. By contrast, the Ab+ individuals did not show any sign of brain atrophy relative to the amyloid-negative controls. Regarding FDG-PET, no significant hypometabolism was found either in the SCD or in the Ab+ groups. Conclusions: Our findings showed that only CN individuals with SCD who refer to a memory clinic, but not CN individuals with amyloid deposition in the brain, have greater episodic memory-related hippocampal atrophy. Hippocampal atrophy is a key hallmark of neurodegeneration in AD and the closest biomarker of its first clinical manifestation. Thus, even if they are still asymptomatic, the short-term prognosis in individuals with SCD could be worse than that of individuals having amyloid deposition in the brain. These results emphasize the critical interest for SCD in preclinical AD.

Cerebral microbleeds, brain beta-amyloid and clinical trajectories: results from the AIBL study of ageing

Australasian Journal on Ageing, Vol 32 Supplement 1, June 2013, 6–35

Midlife vascular risk and late-life amyloid burden: Data from the Women's Healthy Ageing Project (WHAP)

baPWV (cm/s) Both PIB+ and High WMH 2.79 1.35 5.80 0.0058 2.40 1.03 5.57 0.0418 PiB+ only 1.69 0.83 3.48 0.1505 1.27 0.56 2.90 0.5701 High WMH only 1.36 0.65 2.85 0.4218 1.04 0.45 2.40 0.9239 Neither referent referent cfPWV (cm/s) Both PIB+ and High WMH 3.83 1.46 10.06 0.0063 3.26 1.16 9.12 0.0248 PiB+ only 2.24 0.82 6.13 0.1183 1.48 0.49 4.50 0.4843 High WMH only 2.98 1.13 7.86 0.0277 2.41 0.86 6.75 0.0933 Neither referent referent