Methodological evaluation of individual cognitive prediction based on the brain white matter structural connectome

Abstract

An emerging trend is to use regression‐based machine learning approaches to predict cognitive functions at the individual level from neuroimaging data. However, individual prediction models are inherently influenced by the vast options for network construction and model selection in machine learning pipelines. In particular, the brain white matter (WM) structural connectome lacks a systematic evaluation of the effects of different options in the pipeline on predictive performance. Here, we focused on the methodological evaluation of brain structural connectome‐based predictions. For network construction, we considered two parcellation schemes for defining nodes and seven strategies for defining edges. For the regression algorithms, we used eight regression models. Four cognitive domains and brain age were targeted as predictive tasks based on two independent datasets (Beijing Aging Brain Rejuvenation Initiative [BABRI]: 633 healthy older adults; Human Connectome Projects in Aging [HCP‐A]: 560 healthy older adults). Based on the results, the WM structural connectome provided a satisfying predictive ability for individual age and cognitive functions, especially for executive function and attention. Second, different parcellation schemes induce a significant difference in predictive performance. Third, prediction results from different data sets showed that dMRI with distinct acquisition parameters may plausibly result in a preference for proper fiber reconstruction algorithms and different weighting options. Finally, deep learning and Elastic‐Net models are more accurate and robust in connectome‐based predictions. Together, significant effects of different options in WM network construction and regression algorithms on the predictive performances are identified in this study, which may provide important references and guidelines to select suitable options for future studies in this field.

Spatial and temporal pattern of structure-function coupling of human brain connectome with development

“…Regarding fMRI data, the preprocessing pipeline included spatial distortion correction, motion correction, EPI distortion correction, registration to MNI space, intensity normalization, mapping volume time series to 32k_fs_LR mesh, and smoothing using a 2 mm average surface vertex. Following our previous methodological evaluation study [11], the dMRI procedures consisted of intensity normalization of the mean b0 image, correction of EPI distortion and eddy current, motion correction, gradient nonlinearity correction, and linear registration to T1w space.…”

“…Following our previous methodological evaluation study [11], the ball-and-stick model estimated from the bedpostx command-line in the FDT toolbox of FSL (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FDT) was used to estimate fibre orientations (three fibres modelled per voxel) [76][77][78][79]. The BNA atlas was applied to individual volume space by inverse transformation derived from preprocessed steps.…”

“…The BNA atlas was applied to individual volume space by inverse transformation derived from preprocessed steps. Next, probabilistic tractography (probtrackx) [77,79] was implemented in the FDT toolbox to estimate the probability of connectivity between two regions by sampling 5,000 fibres for each voxel within each region, correcting for distance, dividing by the total fibres number in source region, and calculating the average bidirectional probability [11]. Notably, the connections in subcortical areas were removed.…”

“…SC-FC coupling predicts individual differences in cognitive functions. As we found that SC-FC coupling can encode brain maturation, we next evaluated the implications of coupling for individual cognition using Elastic-Net algorithm [11]. After controlling for sex, intracranial volume and in-scanner head motion, we found the SC-FC coupling significantly predicted individual differences in fluid intelligence, crystal intelligence and general intelligence (Pearson's r=0.3~0.4, p<0.001, FDR corrected, Figure 4A), suggesting that increased SC-FC coupling during development is associated with higher intelligence.…”

“…Network modelling of the brain enables the characterization of complex information interactions at a system level and provides natural correspondences between structure and function in the cortex 7,10 . Advances in diffusion MRI (dMRI) and tractography techniques have allowed the in vivo mapping of the white matter (WM) connectome (WMC), which depicts extracortical excitatory projections between regions 11 . The T1-to T2-weighted (T1w/T2w) ratio of MRI has been proposed as a means of quantifying microstructure profile covariance (MPC), which reflects intracortical microcircuit differentiation in molecular, cellular, and laminar differentiation 6,[12][13][14][15] .…”