Resting state FDG-PET functional connectivity as an early biomarker of Alzheimer's disease using conjoint univariate and independent component analyses

Laboratoire d'Imagerie Fonctionnelle, UMR-S 678, INSERM-UPMC, Paris, France
NeuroImage (Impact Factor: 6.36). 04/2012; 63(2):936-46. DOI: 10.1016/j.neuroimage.2012.03.091
Source: PubMed


Imaging cerebral glucose metabolism with positron emission tomography (PET) in Alzheimer's disease (AD) has allowed for improved characterisation of this pathology. Such patterns are typically analysed using either univariate or multivariate statistical techniques. In this work we combined voxel-based group analysis and independent component analysis to extract differential characteristic patterns from PET data of glucose metabolism in a large cohort of normal elderly controls and patients with AD. The patterns were used in conjunction with a support vector machine to discriminate between subjects with mild cognitive impairment (MCI) at risk or not of converting to AD. The method was applied to baseline fluoro-deoxyglucose (FDG)-PET images of subjects from the ADNI database. Our approach achieved improved early detection and differentiation of typical versus pathological metabolic patterns in the MCI population, reaching 80% accuracy (85% sensitivity and 75% specificity) when using selected regions. The method has the potential to assist in the advance diagnosis of Alzheimer's disease, and to identify early in the development of the disease those individuals at high risk of rapid cognitive decline who could be candidates for new therapeutic approaches.

100 Reads
  • Source
    • "An ICA was performed to yield regional contributing components of metabolic networks. A total of 13 metabolic independent components (ICs) were identified in rats similar to components found in previous reports in humans on FDG PET (Di et al., 2012; Toussaint et al., 2012; Yakushev et al., 2013). All ICs were displayed in Figure 2—figure supplement 1, and a threshold z > 1.5 was applied for the voxels for display purposes. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Neuroimaging has been used to examine developmental changes of the brain. While PET studies revealed maturation related changes, maturation of metabolic connectivity of the brain is not yet understood. Here, we show that rat brain metabolism is reconfigured to achieve long-distance connections with higher energy efficiency during maturation. Metabolism increased in anterior cerebrum and decreased in thalamus and cerebellum during maturation. When functional covariance patterns of PET images were examined, metabolic networks including default mode network (DMN) were extracted. Connectivity increased between the anterior and posterior parts of DMN and sensory-motor cortices during maturation. Energy efficiency, a ratio of connectivity strength to metabolism of a region, increased in medial prefrontal and retrosplenial cortices. Our data revealed that metabolic networks mature to increase metabolic connections and establish its efficiency between large-scale spatial components from childhood to early adulthood. Neurodevelopmental diseases might be understood by abnormal reconfiguration of metabolic connectivity and efficiency.
    Full-text · Article · Nov 2015 · eLife Sciences
  • Source
    • "In contrast, [18F]2-fluoro-2-deoxyglucose 18FDG-PET provides a signal with direct correlates to neuronal activations. Not by chance, due to its clear connection with the metabolic dynamics of the brain regions of interest, 18FDG-PET has been profitably used to develop diagnostic tools for AD [2]. However, the fMRI and PET imaging systems both suffer from a poor temporal resolution, and studies on task-related investigations are strongly limited. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Alzheimer's Disease (AD) is an invalidating neurodegenerative disorders frequently affecting the aging population. In view of the increase of elderlies, not only in western countries, the related growing societal problems urge for identifying clinical biomarkers in view of potential treatments interfering or blocking the disease course. Among the plenty of anatomo-functional in vivo imaging techniques to inspect brain circuits and physiology, the Magnetic Resonance Imaging (MRI), the functional MRI (fMRI), the Electroencephalography (EEG) and Magnetoencephalography (MEG), have been extensively used for the study of AD, with different achievements and limitations. Eventually, the methodologies summoned by brain connectomics further strengthen the expectations in this field, as shown by recent results obtained with [18F]2-fluoro-2-deoxyglucose 18FDG-PET and fMRI in the prediction of the AD in early stages. However, the inherent complexity of the pathophysiology of the AD suggests that only integrative approaches combining different techniques and methodologies of brain scanning could produce significant breakthroughs in the study of AD. This review proposes a formal framework able to combine brain connectomic data from multimodal acquisitions by means of different in vivo neuroimaging techniques, briefly reporting their different advantages and drawbacks. Indeed, a specialized complex multiplex network, where nodes interact in layers linking the same pair of nodes and each layer reflects a distinct type of brain acquisition, can model the plurality of connectomes recommended in this framework.
    Full-text · Article · Nov 2015 · Current Alzheimer research
  • Source
    • "Most of these studies were conducted in large cohorts of patients from the Alzheimer3s Disease Neuromaging Initiative (ADNI) database and reached a statistical accuracy in discriminating AD patients from controls of 93% when using three biomarkers (MRI, FDG-PET and CSF) (Zhang et al., 2011) and higher than 90% with FDG-PET only (Arbizu et al., 2013; Gray et al., 2012; Illan et al., 2011). However accuracies were substantially lower in studies in which MCI (later converted to AD) was considered instead of patients with full-blown AD dementia (Cuingnet et al., 2011; Toussaint et al., 2012; Zhang et al., 2011). Though FDG-PET has been shown to correlate with the severity of cognitive impairment (Chen et al., 2010), subtle abnormalities at the earliest (prodromal ) stages of the disease may be more difficult to be detected. "
    [Show abstract] [Hide abstract]
    ABSTRACT: An emerging issue in neuroimaging is to assess the diagnostic reliability of PET and its application in clinical practice. We aimed at assessing the accuracy of brain FDG-PET in discriminating patients with MCI due to Alzheimer's disease and healthy controls. Sixty-two patients with amnestic MCI and 109 healthy subjects recruited in five centers of the European AD Consortium were enrolled. Group analysis was performed by SPM8 to confirm metabolic differences. Discriminant analyses were then carried out using the mean FDG uptake values normalized to the cerebellum computed in 45 anatomical volumes of interest (VOIs) in each hemisphere (90 VOIs) as defined in the Automated Anatomical Labeling (AAL) Atlas and on 12 meta-VOIs, bilaterally, obtained merging VOIs with similar anatomo-functional characteristics. Further, asymmetry indexes were calculated for both datasets. Accuracy of discrimination by a Support Vector Machine and the AAL VOIs was tested against a validated method (PALZ). At the voxel level SMP8 showed a relative hypometabolism in the bilateral precuneus, and posterior cingulate, temporo-parietal and frontal cortices. Discriminant analysis classified subjects with an accuracy ranging between .91 and .83 as a function of data organization. The best values were obtained from a subset of 6 meta-VOIs plus 6 asymmetry values reaching an area under the ROC curve of .947, significantly larger than the one obtained by the PALZ score. High accuracy in discriminating MCI converters from healthy controls was reached by a non-linear classifier based on SVM applied on predefined anatomo-functional regions and inter-hemispheric asymmetries. Data pre-processing was automated and simplified by an in-house created Matlab-based script encouraging its routine clinical use. Further validation toward nonconverter MCI patients with adequately long follow-up is needed.
    Full-text · Article · Nov 2014 · Clinical neuroimaging
Show more