Learning-based structurally-guided construction of resting-state functional correlation tensors

Abstract

Functional magnetic resonance imaging (fMRI) measures changes in blood-oxygenation-level-dependent (BOLD) signals to detect brain activities. It has been recently reported that the spatial correlation patterns of resting-state BOLD signals in the white matter (WM) also give WM information often measured by diffusion tensor imaging (DTI). These correlation patterns can be captured using functional correlation tensor (FCT), which is analogous to the diffusion tensor (DT) obtained from DTI. In this paper, we propose a noise-robust FCT method aiming at further improving its quality, and making it eligible for further neuroscience study. The novel FCT estimation method consists of three major steps: First, we estimate the initial FCT using a patch-based approach for BOLD signal correlation to improve the noise robustness. Second, by utilizing the relationship between functional and diffusion data, we employ a regression forest model to learn the mapping between the initial FCTs and the corresponding DTs using the training data. The learned forest can then be applied to predict the DTI-like tensors given the initial FCTs from the testing fMRI data. Third, we re-estimate the enhanced FCT by utilizing the DTI-like tensors as a feedback guidance to further improve FCT computation. We have demonstrated the utility of our enhanced FCTs in Alzheimer’s disease (AD) diagnosis by identifying mild cognitive impairment (MCI) patients from normal subjects.

Multi-Channel 3D Deep Feature Learning for Survival Time Prediction of Brain Tumor Patients Using Multi-Modal Neuroimages

“…Note that these 5 maps mainly focus on grey matter. To also extract functional features in white matter, we propose to use the functional connectivity tensor (fTensor-fMRI) 11,47 to provide functional information in white matter. The fTensor-fMRI was originally proposed in the work 47 to measure the structured spatiotemporal relationship among the BOLD signals of neighboring voxels in white matter, which shows the anisotropic pattern that is generally consistent with the diffusion anisotropy derived from DTI in major fiber bundles.…”

Functional tractography of white matter by high angular resolution functional‐correlation imaging (HARFI)

“…There is also potential to harmonize diffusion MRI structural data and fMRI functional data. For example, Zhang et al used machine learning to predict or inform functional anisotropy mapping from DTI data. We also envision the use of functional correlations to inform DTI tractography, possibly by weighting the streamline propagation step with the functional orientation, or filtering diffusion tractography streamlines using functional information extracted from the HARFI FODs.…”

“…First, the use of nearest-neighbor voxels (in combination with a limited SNR of WM BOLD effects) makes FCTs sensitive to noise. 16 Second, adjacent neighboring voxels tend to have higher correlations than diagonal neighbors, possibly caused by the decreased distance from the voxel-of-interest (i.e., 1 voxel versus √2 voxel distances), which can result in orientation-related biases. Third, the use of a 2nd-order tensor model restricts functional correlations to an ellipsoidal shape.…”

Automatic brain labeling via multi-atlas guided fully convolutional networks

“…Anatomical brain labeling is highly desired for region-based analysis of MR brain images, which is important for many research studies and clinical applications, such as facilitating diagnosis [1,2] and investigating early brain development [3]. Also, brain labeling is a fundamental step in brain network analysis pipelines, where regions-of-interest (ROIs) need to be identified prior to exploring any connectivity traits [4][5][6][7]. But it is labor-intensive and impractical to manually label a large set of 3D MR images, thus recent developments focused on automatic labeling of brain anatomy.…”