Resting state fMRI (R-fMRI) and intrinsic functional connectivity (iFC) mapping have energized developmental and psychiatric functional neuroimaging, but concerns about the impact of participant motion and preprocessing strategies on R-fMRI findings have begun to stall progress in the field. Faced with a plethora of options for retrospective motion correction and nuisance signal regression, researchers are unsure of how to proceed. A recently published method, multi-echo independent components regression (ME-ICR), offers a principled and objective approach to identifying and removing motion-related artifact and other signals of no- interest from R-fMRI data. This novel method capitalizes on the principle that multiecho (ME) acquisitions can be used to differentiate non-BOLD changes in fMRI signal intensity (related to nuisance sources such as motion, physiological rhythms, and scanner-related noise) from changes in BOLD signals of interest, which exhibit a dependence on echo time (TE). This principle is applied in ME-ICR to sort signal components identified using independent components analysis (ICA) according to their TE dependence, thus separating TE-dependent BOLD signals of interest from TE-independent non-BOLD signals of no interest. When validated in a sample of 35 healthy adults, ME-ICR exhibited increased temporal signal-to-noise, successful removal of motion artifacts, increased specificity of iFC maps, and improved statistical inference, relative to conventional analyses. Although compelling, the approach requires not only a change in analytic procedures but also in data acquisition (from single-shot to ME). Labs will be reluctant to adopt ME-ICR method in the absence of conclusive evidence of its applicability to clinical populations. Efforts to acquire new ME datasets in clinical populations could take years. Providentially, our lab has used an ME sequence since 2009. We thus propose to apply ME-ICR to our uniquely large database of ME R-fMRI scans collected from over 600 children and adults with ADHD, autism, as well as typical comparisons. We will examine the impact of ME-ICR on estimates of age- and diagnosis-related differences in R-fMRI measures, and on their test-retest reliability. By assessing the validity and reliability of ME-ICR in a large sample of existing data, the proposed study will accelerate the demonstration of its utility, and ultimately, adoption, for developmental and psychiatric R-fMRI studies. This proposal is timely and potentially transformational, since the methodological advances offered by ME-ICR would permit the field to transcend present challenges, thus enabling R-fMRI to fulfill its promise as a translational approach that faithfully reveals macro-scale brain function in health and disease. In addition, by permitting valid statistical inference at the individual subject level, ME-ICR can move neuroimaging toward the goals of personalized medicine. Finally, by proposing sharing of the ME data, the scientific impact of this project will be greatly enhanced at no additional cost.

Public Health Relevance

Resting state approaches have opened up new frontiers in developmental and psychiatric functional neuroimaging, permitting the study of increasingly younger and lower functioning participants. However, analytical concerns have begun to stall progress in the field. We propose to establish the utility of a novel approach that overcomes these concerns by re-analyzing our uniquely large database of over 1000 scans collected from developing and clinical populations, thus permitting the field to transcend present challenges.

Agency
National Institute of Health (NIH)
Type
Small Research Grants (R03)
Project #
1R03MH104334-01
Application #
8748494
Study Section
Child Psychopathology and Developmental Disabilities Study Section (CPDD)
Program Officer
Freund, Michelle
Project Start
Project End
Budget Start
Budget End
Support Year
1
Fiscal Year
2014
Total Cost
Indirect Cost
Name
New York University
Department
Psychiatry
Type
Schools of Medicine
DUNS #
City
New York
State
NY
Country
United States
Zip Code
10016
Di Martino, Adriana; Fair, Damien A; Kelly, Clare et al. (2014) Unraveling the miswired connectome: a developmental perspective. Neuron 83:1335-53