C-ICT for Discovery of Multiple Associations in Multimodal Imaging Data: Application to Fusion of fMRI and DTI Data
- 1 March 2019
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in 2019 53rd Annual Conference on Information Sciences and Systems (CISS)
Abstract
Fusing datasets from different brain signal modalities improves accuracy in finding biomarkers of neuropsychiatric diseases. Several approaches, such as joint independent component analysis (ICA) and independent vector analysis (IVA), are useful but fall short of exploring multiple associations between different modalities, especially for the case where one underlying component in one modality might have multiple associations with others in another modality. This relationship is possible since one component in a given modality might have associations such as subject covariation with multiple components in another modality. We show that the consecutive independence and correlation transform (C-ICT) model, which successively performs ICA and canonical correlation analysis, is able to discover such multiple associations. C-ICT has been demonstrated to be useful for the fusion of functional magnetic resonance imaging (fMRI) and electroencephalography data but has not been tested for other data combinations. In this study, we apply the C-ICT to fuse fMRI and MRI-based diffusion tensor imaging (DTI) datasets collected from healthy controls and patients with schizophrenia. In addition to independent components that show significant differences between the two groups in the fMRI and DTI datasets separately, we find multiple associations between these components from the two modalities, which provide a unique potential biomarker for schizophrenia.Keywords
This publication has 33 references indexed in Scilit:
- A review of multivariate methods for multimodal fusion of brain imaging dataJournal of Neuroscience Methods, 2012
- Discriminating schizophrenia and bipolar disorder by fusing fMRI and DTI in a multimodal CCA+ joint ICA modelNeuroImage, 2011
- The Cerebellum and Basal Ganglia are InterconnectedNeuropsychology Review, 2010
- Evidence for topographic organization in the cerebellum of motor control versus cognitive and affective processingCortex, 2010
- A CCA+ICA based model for multi-task brain imaging data fusion and its application to schizophreniaNeuroImage, 2010
- Multi-set canonical correlation analysis for the fusion of concurrent single trial ERP and functional MRINeuroImage, 2010
- Conditional routing of information to the cortex: A model of the basal ganglia’s role in cognitive coordination.Psychological Review, 2010
- The Role of the Cerebellum in SchizophreniaBiological Psychiatry, 2008
- Combining fMRI and SNP data to investigate connections between brain function and genetics using parallel ICAHuman Brain Mapping, 2007
- Basal ganglia and cerebellar loops: motor and cognitive circuitsBrain Research Reviews, 2000