Article

Staging Alzheimer's disease progression with multimodality neuroimaging

Department of Radiology, University of California at San Francisco, San Francisco, USA.
Progress in Neurobiology (Impact Factor: 10.3). 06/2011; 95(4):535-46. DOI: 10.1016/j.pneurobio.2011.06.004
Source: PubMed

ABSTRACT Rapid developments in medical neuroimaging have made it possible to reconstruct the trajectory of Alzheimer's disease (AD) as it spreads through the living brain. The current review focuses on the progressive signature of brain changes throughout the different stages of AD. We integrate recent findings on changes in cortical gray matter volume, white matter fiber tracts, neuropathological alterations, and brain metabolism assessed with molecular positron emission tomography (PET). Neurofibrillary tangles accumulate first in transentorhinal and cholinergic brain areas, and 4-D maps of cortical volume changes show early progressive temporo-parietal cortical thinning. Findings from diffusion tensor imaging (DTI) for assessment fiber tract integrity show cortical disconnection in corresponding brain networks. Importantly, the developmental trajectory of brain changes is not uniform and may be modulated by several factors such as onset of disease mechanisms, risk-associated and protective genes, converging comorbidity, and individual brain reserve. There is a general agreement between in vivo brain maps of cortical atrophy and amyloid pathology assessed through PET, reminiscent of post mortem histopathology studies that paved the way in the staging of AD. The association between in vivo and post mortem findings will clarify the temporal dynamics of pathophysiological alterations in the development of preclinical AD. This will be important in designing effective treatments that target specific underlying disease AD mechanisms.

0 Bookmarks
 · 
334 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: Despite a massive research effort to elucidate Alzheimer's disease (AD) in recent decades, effective treatment remains elusive. This failure may relate to an oversimplification of the pathogenic processes underlying AD and also lack of understanding of AD progression during its long latent stages. Although evidence shows that the two specific neuropathological hallmarks in AD (neuronal loss and protein accumulation), which are opposite in nature, do not progress in parallel, the great majority of studies have focused on only one of these aspects. Furthermore, research focusing on single structures is likely to render an incomplete picture of AD pathogenesis because as AD involves complete brain networks, potential compensatory mechanisms within the network may ameliorate impairment of the system to a certain extent. Here, we describe an approach for enabling integrative analysis of the dual-nature lesions, simultaneously, in all components of one of the brain networks most vulnerable to AD. This approach is based on significant development of methods previously described mainly by our group that were optimized and complemented for this study. It combines unbiased stereology with immunohistochemistry and immunofluorescence, making use of advanced graphics computing for three-dimensional (3D) volume reconstructions. Although this study was performed in human brainstem and focused in AD, it may be applied to the study of any neurological disease characterized by dual-nature lesions, in humans and animal models. This approach does not require a high level of investment in new equipment and a significant number of specimens can be processed and analyzed within a funding cycle
    Journal of neuroscience methods 04/2014; 226. DOI:10.1016/j.jneumeth.2014.01.030 · 2.30 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Alzheimer's disease (AD) is regarded a human-specific condition, and it has been suggested that brain regions highly expanded in humans compared with other primates are selectively targeted. We calculated shared and unique variance in the distribution of AD atrophy accounted for by cortical expansion between macaque and human, affiliation to the default mode network (DMN), ontogenetic development and normal aging. Cortical expansion was moderately related to atrophy, but a critical discrepancy was seen in the medial temporo-parietal episodic memory network. Identification of "hotspots" and "coldspots" of expansion across several primate species did not yield compelling evidence for the hypothesis that highly expanded regions are specifically targeted. Controlling for distribution of atrophy in aging substantially attenuated the expansion-AD relationship. A path model showed that all variables explained unique variance in AD atrophy but were generally mediated through aging. This supports a systems-vulnerability model, where critical networks are subject to various negative impacts, aging in particular, rather than being selectively targeted in AD. An alternative approach is suggested, focused on the interplay of the phylogenetically old and preserved medial temporal lobe areas with more highly expanded association cortices governed by different principles of plasticity and stability.
    Cerebral Cortex 03/2014; DOI:10.1093/cercor/bhu055 · 8.31 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Diffusion tensor imaging (DTI) based assessment of white matter fiber tract integrity can support the diagnosis of Alzheimer's disease (AD). The use of DTI as a biomarker, however, depends on its applicability in a multicenter setting accounting for effects of different MRI scanners. We applied multivariate machine learning (ML) to a large multicenter sample from the recently created framework of the European DTI study on Dementia (EDSD). We hypothesized that ML approaches may amend effects of multicenter acquisition. We included a sample of 137 patients with clinically probable AD (MMSE 20.6+/-5.3) and 143 healthy elderly controls, scanned in nine different scanners. For diagnostic classification we used the DTI indices fractional anisotropy (FA) and mean diffusivity (MD) and, for comparison, gray matter and white matter density maps from anatomical MRI. Data were classified using a Support Vector Machine (SVM) and a Naive Bayes (NB) classifier. We used two cross-validation approaches, (i) test and training samples randomly drawn from the entire data set (pooled cross-validation) and (ii) data from each scanner as test set, and the data from the remaining scanners as training set (scanner-specific cross-validation). In the pooled cross-validation, SVM achieved an accuracy of 80% for FA and 83% for MD. Accuracies for NB were significantly lower, ranging between 68% and 75%. Removing variance components arising from scanners using principal component analysis did not significantly change the classification results for both classifiers. For the scanner-specific cross-validation, the classification accuracy was reduced for both SVM and NB. After mean correction, classification accuracy reached a level comparable to the results obtained from the pooled cross-validation. Our findings support the notion that machine learning classification allows robust classification of DTI data sets arising from multiple scanners, even if a new data set comes from a scanner that was not part of the training sample.
    PLoS ONE 05/2013; 8(5):e64925. DOI:10.1371/journal.pone.0064925 · 3.53 Impact Factor

Full-text (2 Sources)

Download
86 Downloads
Available from
May 31, 2014