Identifying canonical and replicable multi-scale intrinsic connectivity networks in 100k+ resting-state fMRI datasets

A Iraji; Z Fu; A Faghiri; M Duda; J Chen; S Rachakonda; T DeRamus; P Kochunov; B M Adhikari; A Belger; J M Ford; D H Mathalon; G D Pearlson; S G Potkin; A Preda; J A Turner; T G M van Erp; J R Bustillo; K Yang; K Ishizuka; A Faria; A Sawa; K Hutchison; E A Osuch; J Theberge; C Abbott; B A Mueller; D Zhi; C Zhuo; S Liu; Y Xu; M Salman; J Liu; Y Du; J Sui; T Adali; V D Calhoun

doi:10.1002/hbm.26472

Identifying canonical and replicable multi-scale intrinsic connectivity networks in 100k+ resting-state fMRI datasets

Hum Brain Mapp. 2023 Dec 1;44(17):5729-5748. doi: 10.1002/hbm.26472. Epub 2023 Oct 3.

Authors

A Iraji^{1

2}, Z Fu¹, A Faghiri¹, M Duda¹, J Chen¹, S Rachakonda¹, T DeRamus¹, P Kochunov³, B M Adhikari³, A Belger⁴, J M Ford^{5

6}, D H Mathalon^{5

6}, G D Pearlson⁷, S G Potkin⁸, A Preda⁸, J A Turner⁹, T G M van Erp¹⁰, J R Bustillo¹¹, K Yang¹², K Ishizuka¹², A Faria¹², A Sawa^{13

14}, K Hutchison¹⁵, E A Osuch¹⁶, J Theberge¹⁶, C Abbott¹⁷, B A Mueller¹⁸, D Zhi¹⁹, C Zhuo²⁰, S Liu²¹, Y Xu²¹, M Salman^{1

22}, J Liu^{1

2}, Y Du^{1

23}, J Sui^{1

19}, T Adali²⁴, V D Calhoun^{1

2

12

22}

Affiliations

¹ Tri-Institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS), Georgia State University, Georgia Institute of Technology, and Emory University, Atlanta, Georgia, USA.
² Department of Computer Science, Georgia State University, Atlanta, Georgia, USA.
³ Maryland Psychiatric Research Center, Department of Psychiatry, School of Medicine, University of Maryland, Baltimore, Maryland, USA.
⁴ Department of Psychiatry, University of North Carolina, Chapel Hill, North Carolina, USA.
⁵ Department of Psychiatry, University of California San Francisco, San Francisco, California, USA.
⁶ San Francisco VA Medical Center, San Francisco, California, USA.
⁷ Departments of Psychiatry and Neuroscience, School of Medicine, Yale University, New Haven, Connecticut, USA.
⁸ Department of Psychiatry and Human Behavior, University of California Irvine, Irvine, California, USA.
⁹ Department of Psychiatry and Behavioral Health, Ohio State University Medical Center in Columbus, Columbus, Ohio, USA.
¹⁰ Clinical Translational Neuroscience Laboratory, Department of Psychiatry and Human Behavior, University of California Irvine, Irvine, California, USA.
¹¹ Department of Psychiatry and Behavioral Sciences, University of New Mexico, Albuquerque, New Mexico, USA.
¹² Department of Psychiatry, School of Medicine, Johns Hopkins University, Baltimore, Maryland, USA.
¹³ Departments of Psychiatry, Neuroscience, Biomedical Engineering, Pharmacology, and Genetic Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA.
¹⁴ Department of Mental Health, Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland, USA.
¹⁵ Department of Psychology, University of Colorado, Boulder, Colorado, USA.
¹⁶ Department of Psychiatry, Schulich School of Medicine and Dentistry, London Health Sciences Centre, Lawson Health Research Institute, London, Canada.
¹⁷ Department of Psychiatry (CCA), University of New Mexico, Albuquerque, New Mexico, USA.
¹⁸ Department of Psychiatry, University of Minnesota, Minneapolis, Minnesota, USA.
¹⁹ The State Key Lab of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing, China.
²⁰ Tianjin Mental Health Center, Nankai University Affiliated Anding Hospital, Tianjin, China.
²¹ The Department of Psychiatry, First Clinical Medical College/First Hospital of Shanxi Medical University, Taiyuan, China.
²² School of Electrical & Computer Engineering, Georgia Institute of Technology, Atlanta, Georgia, USA.
²³ School of Computer and Information Technology, Shanxi University, Taiyuan, China.
²⁴ Department of CSEE, University of Maryland Baltimore County, Baltimore, Maryland, USA.

Abstract

Despite the known benefits of data-driven approaches, the lack of approaches for identifying functional neuroimaging patterns that capture both individual variations and inter-subject correspondence limits the clinical utility of rsfMRI and its application to single-subject analyses. Here, using rsfMRI data from over 100k individuals across private and public datasets, we identify replicable multi-spatial-scale canonical intrinsic connectivity network (ICN) templates via the use of multi-model-order independent component analysis (ICA). We also study the feasibility of estimating subject-specific ICNs via spatially constrained ICA. The results show that the subject-level ICN estimations vary as a function of the ICN itself, the data length, and the spatial resolution. In general, large-scale ICNs require less data to achieve specific levels of (within- and between-subject) spatial similarity with their templates. Importantly, increasing data length can reduce an ICN's subject-level specificity, suggesting longer scans may not always be desirable. We also find a positive linear relationship between data length and spatial smoothness (possibly due to averaging over intrinsic dynamics), suggesting studies examining optimized data length should consider spatial smoothness. Finally, consistency in spatial similarity between ICNs estimated using the full data and subsets across different data lengths suggests lower within-subject spatial similarity in shorter data is not wholly defined by lower reliability in ICN estimates, but may be an indication of meaningful brain dynamics which average out as data length increases.

Keywords: functional connectivity (FC); functional templates; independent component analysis (ICA); intrinsic connectivity networks (ICNs).

Publication types

Research Support, N.I.H., Extramural
Research Support, U.S. Gov't, Non-P.H.S.
Research Support, Non-U.S. Gov't

MeSH terms

Brain / diagnostic imaging
Brain Mapping* / methods
Humans
Magnetic Resonance Imaging* / methods
Nerve Net / diagnostic imaging
Reproducibility of Results

Abstract

Publication types

MeSH terms

Grants and funding