Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 8, 23

Cross-View Neuroimage Pattern Analysis in Alzheimer's Disease Staging


Cross-View Neuroimage Pattern Analysis in Alzheimer's Disease Staging

Sidong Liu et al. Front Aging Neurosci.


The research on staging of pre-symptomatic and prodromal phase of neurological disorders, e.g., Alzheimer's disease (AD), is essential for prevention of dementia. New strategies for AD staging with a focus on early detection, are demanded to optimize potential efficacy of disease-modifying therapies that can halt or slow the disease progression. Recently, neuroimaging are increasingly used as additional research-based markers to detect AD onset and predict conversion of MCI and normal control (NC) to AD. Researchers have proposed a variety of neuroimaging biomarkers to characterize the patterns of the pathology of AD and MCI, and suggested that multi-view neuroimaging biomarkers could lead to better performance than single-view biomarkers in AD staging. However, it is still unclear what leads to such synergy and how to preserve or maximize. In an attempt to answer these questions, we proposed a cross-view pattern analysis framework for investigating the synergy between different neuroimaging biomarkers. We quantitatively analyzed nine types of biomarkers derived from FDG-PET and T1-MRI, and evaluated their performance in a task of classifying AD, MCI, and NC subjects obtained from the ADNI baseline cohort. The experiment results showed that these biomarkers could depict the pathology of AD from different perspectives, and output distinct patterns that are significantly associated with the disease progression. Most importantly, we found that these features could be separated into clusters, each depicting a particular aspect; and the inter-cluster features could always achieve better performance than the intra-cluster features in AD staging.

Keywords: Alzheimer's disease; mild cognitive impairment; multi-modal; neuroimaging; pattern recognition.


Figure 1
Figure 1
The work-flow of the cross-view pattern analysis. It is a five-step pipeline, which takes brain T1-MRI and FDG-PET images as inputs and generates the classification results as the outputs.
Figure 2
Figure 2
Back projection of the normalized weights of the ROIs for four MRI views onto the ICBM_152 template using 3D Slicer (Fedorov et al., 2012).
Figure 3
Figure 3
Back projection of the normalized weights of the ROIs for five PET views onto the ICBM_152 template using 3D Slicer.
Figure 4
Figure 4
The clustering results for the nine views in the 2D space. The structural and functional features are substantially separated when considering the first two eigenmodes.

Similar articles

See all similar articles

Cited by 2 PubMed Central articles


    1. Andreasson U., Lautner R., Schott J. M., Mattsson N., Hansson O., Herukka S. K., et al. (2014). CSF biomarker for Alzheimer's pathology and the effect size of APOE ϵ4. Mol. Psychiatry 19, 148–149. 10.1038/mp.2013.18 - DOI - PMC - PubMed
    1. Atrey P., Hossain M. A., El Saddik A., Kankanhalli M. (2010). Multimodal fusion for multimedia analysis: a survey. Multi. Syst. 16, 345–379. 10.1007/s00530-010-0182-0 - DOI
    1. Awate S. P., Yushkevich P. A., Song Z., Licht D. J., Gee J. C. (2010). Cerebral cortical folding analysis with multivariate modeling and testing: studies on gender differences and neonatal development. Neuroimage 53, 450–459. 10.1016/j.neuroimage.2010.06.072 - DOI - PMC - PubMed
    1. Batty S., Clark J., Fryer T., Gao X. (2008). Prototype system for semantic retrieval of neurological PET images, in Medical Imaging and Informatics Vol. 4987 LNCS, eds Gao X., Müller H., Loomes M., Comley R., Luo S., editors. (Berlin; Heidelberg: Springer; ), 179–188.
    1. Bergstra J., Bengio Y. (2012). Random search for hyper-parameter optimization. J. Machine Learn. Res. 13, 281–305. Available online at:

LinkOut - more resources