Continuous evaluation of denoising strategies in resting-state fMRI connectivity using fMRIPrep and Nilearn.

Reducing contributions from non-neuronal sources is a crucial step in functional magnetic resonance imaging (fMRI) connectivity analyses. Many viable strategies for denoising fMRI are used in the literature, and practitioners rely on denoising benchmarks for guidance in the selection of an appropriate choice for their study. However, fMRI denoising software is an ever-evolving field, and the benchmarks can quickly become obsolete as the techniques or implementations change. In this work, we present a denoising benchmark featuring a range of denoising strategies, datasets and evaluation metrics for connectivity analyses, based on the popular fMRIprep software. The benchmark prototypes an implementation of a reproducible framework, where the provided Jupyter Book enables readers to reproduce or modify the figures on the Neurolibre reproducible preprint server (https://neurolibre.org/). We demonstrate how such a reproducible benchmark can be used for continuous evaluation of research software, by comparing two versions of the fMRIprep. Most of the benchmark results were consistent with prior literature. Scrubbing, a technique which excludes time points with excessive motion, combined with global signal regression, is generally effective at noise removal. Scrubbing was generally effective, but is incompatible with statistical analyses requiring the continuous sampling of brain signal, for which a simpler strategy, using motion parameters, average activity in select brain compartments, and global signal regression, is preferred. Importantly, we found that certain denoising strategies behave inconsistently across datasets and/or versions of fMRIPrep, or had a different behavior than in previously published benchmarks. This work will hopefully provide useful guidelines for the fMRIprep users community, and highlight the importance of continuous evaluation of research methods.

Continuous Evaluation of Denoising Strategies in Resting-State fMRI Connectivity Using fMRIPrep and Nilearn.

Reducing contributions from non-neuronal sources is a crucial step in functional magnetic resonance imaging (fMRI) connectivity analyses. Many viable strategies for denoising fMRI are used in the literature, and practitioners rely on denoising benchmarks for guidance in the selection of an appropriate choice for their study. However, fMRI denoising software is an ever-evolving field, and the benchmarks can quickly become obsolete as the techniques or implementations change. In this work, we present a denoising benchmark featuring a range of denoising strategies, datasets and evaluation metrics for connectivity analyses, based on the popular fMRIprep software. The benchmark is implemented in a fully reproducible framework, where the provided research objects enable readers to reproduce or modify core computations, as well as the figures of the article using the Jupyter Book project and the Neurolibre reproducible preprint server (https://neurolibre.org/). We demonstrate how such a reproducible benchmark can be used for continuous evaluation of research software, by comparing two versions of the fMRIprep software package. The majority of benchmark results were consistent with prior literature. Scrubbing, a technique which excludes time points with excessive motion, combined with global signal regression, is generally effective at noise removal. Scrubbing however disrupts the continuous sampling of brain images and is incompatible with some statistical analyses, e.g. auto-regressive modeling. In this case, a simple strategy using motion parameters, average activity in select brain compartments, and global signal regression should be preferred. Importantly, we found that certain denoising strategies behave inconsistently across datasets and/or versions of fMRIPrep, or had a different behavior than in previously published benchmarks. This work will hopefully provide useful guidelines for the fMRIprep users community, and highlight the importance of continuous evaluation of research methods. Our reproducible benchmark infrastructure will facilitate such continuous evaluation in the future, and may also be applied broadly to different tools or even research fields.

Brain functional connectivity mirrors genetic pleiotropy in psychiatric conditions.

Pleiotropy occurs when a genetic variant influences more than one trait. This is a key property of the genomic architecture of psychiatric disorders and has been observed for rare and common genomic variants. It is reasonable to hypothesize that the microscale genetic overlap (pleiotropy) across psychiatric conditions and cognitive traits may lead to similar overlaps at the macroscale brain level such as large-scale brain functional networks. We took advantage of brain connectivity, measured by resting-state functional MRI to measure the effects of pleiotropy on large-scale brain networks, a putative step from genes to behaviour. We processed nine resting-state functional MRI datasets including 32 726 individuals and computed connectome-wide profiles of seven neuropsychiatric copy-number-variants, five polygenic scores, neuroticism and fluid intelligence as well as four idiopathic psychiatric conditions. Nine out of 19 pairs of conditions and traits showed significant functional connectivity correlations (rFunctional connectivity), which could be explained by previously published levels of genomic (rGenetic) and transcriptomic (rTranscriptomic) correlations with moderate to high concordance: rGenetic-rFunctional connectivity = 0.71 [0.40-0.87] and rTranscriptomic-rFunctional connectivity = 0.83 [0.52; 0.94]. Extending this analysis to functional connectivity profiles associated with rare and common genetic risk showed that 30 out of 136 pairs of connectivity profiles were correlated above chance. These similarities between genetic risks and psychiatric disorders at the connectivity level were mainly driven by the overconnectivity of the thalamus and the somatomotor networks. Our findings suggest a substantial genetic component for shared connectivity profiles across conditions and traits, opening avenues to delineate general mechanisms-amenable to intervention-across psychiatric conditions and genetic risks.

Genetic Heterogeneity Shapes Brain Connectivity in Psychiatry.

BACKGROUND: Polygenicity and genetic heterogeneity pose great challenges for studying psychiatric conditions. Genetically informed approaches have been implemented in neuroimaging studies to address this issue. However, the effects on functional connectivity of rare and common genetic risks for psychiatric disorders are largely unknown. Our objectives were to estimate and compare the effect sizes on brain connectivity of psychiatric genomic risk factors with various levels of complexity: oligogenic copy number variants (CNVs), multigenic CNVs, and polygenic risk scores (PRSs) as well as idiopathic psychiatric conditions and traits. METHODS: Resting-state functional magnetic resonance imaging data were processed using the same pipeline across 9 datasets. Twenty-nine connectome-wide association studies were performed to characterize the effects of 15 CNVs (1003 carriers), 7 PRSs, 4 idiopathic psychiatric conditions (1022 individuals with autism, schizophrenia, bipolar conditions, or attention-deficit/hyperactivity disorder), and 2 traits (31,424 unaffected control subjects). RESULTS: Effect sizes on connectivity were largest for psychiatric CNVs (estimates: 0.2-0.65 z score), followed by psychiatric conditions (0.15-0.42), neuroticism and fluid intelligence (0.02-0.03), and PRSs (0.01-0.02). Effect sizes of CNVs on connectivity were correlated to their effects on cognition and risk for disease (r = 0.9, p = 5.93 × 10-6). However, effect sizes of CNVs adjusted for the number of genes significantly decreased from small oligogenic to large multigenic CNVs (r = -0.88, p = 8.78 × 10-6). PRSs had disproportionately low effect sizes on connectivity compared with CNVs conferring similar risk for disease. CONCLUSIONS: Heterogeneity and polygenicity affect our ability to detect brain connectivity alterations underlying psychiatric manifestations.

A dataset of long-term consistency values of resting-state fMRI connectivity maps in a single individual derived at multiple sites and vendors using the Canadian Dementia Imaging Protocol.

The impact of multisite acquisition on resting-state functional MRI (rsfMRI) connectivity has recently gained attention. We provide consistency values (Pearson's correlation) between rsfMRI connectivity maps of an adult volunteer (Csub) scanned 25 times over 3.5 years at 13 sites using the Canadian Dementia Imaging Protocol (CDIP, www.cdip-pcid.ca). This dataset was generated as part of the following article: Multivariate consistency of resting-state fMRI connectivity maps acquired on a single individual over 2.5 years, 13 sites and 3 vendors [1]. Acquired on three 3T scanner vendors (GE, Siemens and Philips), the Csub dataset is part of an ongoing effort to monitor the quality and comparability of MRI data collected across the Canadian Consortium on Neurodegeneration in Aging (CCNA) imaging network. The participant was scanned 25 times in the above-mentioned article: multiple times at six sites over a period of 2.5 years, and once at the remaining seven sites. Since then the participant was scanned an additional 45 times, allowing us to extend the dataset to 70 rsfMRI scans over a period of >4 years. In addition, we provide intra- and inter-subject consistency values of rsfMRI connectivity maps derived from 26 adult participants belonging to the publicly released Hangzhou Normal University dataset (HNU1). All HNU1 participants underwent 10 rsfMRI scans over one month on a single 3T scanner (GE). Connectivity maps of seven canonical networks were generated for each scan in the two datasets (Csub and HNU1). All consistency values, along with the scripts used to preprocess the rsfMRI data and generate connectivity maps and pairwise consistency values, have been made available on two public repositories, Github and Zenodo. We have also made available four Jupyter notebooks that use the provided consistency values to (a) generate interactive graphical summaries - 1 notebook, (b) perform statistical analyses - 2 notebooks, and (c) perform data-driven cluster analysis for the recovery of subject identity (i.e. rsfMRI fingerprinting) - 1 notebook. In addition, we provide two interactive dashboards that allow visualization of individual connectivity maps from the two datasets. Finally, we also provide minimally preprocessed rsfMRI data in Brain Imaging Data Standard (BIDS) format on all 70 scans in the extended dataset.

A highly predictive signature of cognition and brain atrophy for progression to Alzheimer's dementia.

BACKGROUND: Clinical trials in Alzheimer's disease need to enroll patients whose cognition will decline over time, if left untreated, in order to demonstrate the efficacy of an intervention. Machine learning models used to screen for patients at risk of progression to dementia should therefore favor specificity (detecting only progressors) over sensitivity (detecting all progressors), especially when the prevalence of progressors is low. Here, we explore whether such high-risk patients can be identified using cognitive assessments and structural neuroimaging by training machine learning tools in a high-specificity regime. RESULTS: A multimodal signature of Alzheimer's dementia was first extracted from the ADNI1 dataset. We then validated the predictive value of this signature on ADNI1 patients with mild cognitive impairment (N = 235). The signature was optimized to predict progression to dementia over 3 years with low sensitivity (55.1%) but high specificity (95.6%), resulting in only moderate accuracy (69.3%) but high positive predictive value (80.4%, adjusted for a "typical" 33% prevalence rate of true progressors). These results were replicated in ADNI2 (N = 235), with 87.8% adjusted positive predictive value (96.7% specificity, 47.3% sensitivity, 85.1% accuracy). CONCLUSIONS: We found that cognitive measures alone could identify high-risk individuals, with structural measurements providing a slight improvement. The signature had comparable receiver operating characteristics to standard machine learning tools, yet a marked improvement in positive predictive value was achieved over the literature by selecting a high-specificity operating point. The multimodal signature can be readily applied for the enrichment of clinical trials.