Random Matrix Theory Tools for the Predictive Analysis of Functional Magnetic Resonance Imaging Examinations

Overview

Journal J Med Imaging (Bellingham)

Specialty Radiology

Date 2023 Jun 16

PMID 37323123

Authors

Derek Berger

Gurpreet S Matharoo

Jacob Levman

Affiliations

Soon will be listed here.

Abstract

Purpose: Random matrix theory (RMT) is an increasingly useful tool for understanding large, complex systems. Prior studies have examined functional magnetic resonance imaging (fMRI) scans using tools from RMT, with some success. However, RMT computations are highly sensitive to a number of analytic choices, and the robustness of findings involving RMT remains in question. We systematically investigate the usefulness of RMT on a wide variety of fMRI datasets using a rigorous predictive framework.

Approach: We develop open-source software to efficiently compute RMT features from fMRI images and examine the cross-validated predictive potential of eigenvalue and RMT-based features ("eigenfeatures") with classic machine-learning classifiers. We systematically vary pre-processing extent, normalization procedures, RMT unfolding procedures, and feature selection and compare the impact of these analytic choices on the distributions of cross-validated prediction performance for each combination of dataset binary classification task, classifier, and feature. To deal with class imbalance, we use the area under the receiver operating characteristic curve (AUROC) as the main performance metric.

Results: Across all classification tasks and analytic choices, we find RMT- and eigenvalue-based "eigenfeatures" to have predictive utility more often than not (82.4% of median ; median AUROC range across classification tasks 0.47 to 0.64). Simple baseline reductions on source timeseries, by contrast, were less useful (58.8% of median , median AUROC range across classification tasks 0.42 to 0.62). Additionally, eigenfeature AUROC distributions were overall more right-tailed than baseline features, suggesting greater predictive potential. However, performance distributions were wide and often significantly affected by analytic choices.

Conclusions: Eigenfeatures clearly have potential for understanding fMRI functional connectivity in a wide variety of scenarios. The utility of these features is strongly dependent on analytic decisions, suggesting caution when interpreting past and future studies applying RMT to fMRI. However, our study demonstrates that the inclusion of RMT statistics in fMRI investigations could improve prediction performances across a wide variety of phenomena.

References

Tetreault P, Mansour A, Vachon-Presseau E, Schnitzer T, Apkarian A, Baliki M . Brain Connectivity Predicts Placebo Response across Chronic Pain Clinical Trials. PLoS Biol. 2016; 14(10):e1002570. PMC: 5082893. DOI: 10.1371/journal.pbio.1002570. View

Zhang S, Li X, Lv J, Jiang X, Guo L, Liu T . Characterizing and differentiating task-based and resting state fMRI signals via two-stage sparse representations. Brain Imaging Behav. 2015; 10(1):21-32. PMC: 4559495. DOI: 10.1007/s11682-015-9359-7. View

Poldrack R, Gorgolewski K . OpenfMRI: Open sharing of task fMRI data. Neuroimage. 2016; 144(Pt B):259-261. PMC: 4669234. DOI: 10.1016/j.neuroimage.2015.05.073. View

Jalan S, Sarkar C, Madhusudanan A, Dwivedi S . Uncovering randomness and success in society. PLoS One. 2014; 9(2):e88249. PMC: 3922759. DOI: 10.1371/journal.pone.0088249. View

Fossion R, Torres Vargas G, Lopez Vieyra J . Random-matrix spectra as a time series. Phys Rev E Stat Nonlin Soft Matter Phys. 2014; 88(6):060902. DOI: 10.1103/PhysRevE.88.060902. View

Riaz A, Asad M, Alonso E, Slabaugh G . DeepFMRI: End-to-end deep learning for functional connectivity and classification of ADHD using fMRI. J Neurosci Methods. 2020; 335:108506. DOI: 10.1016/j.jneumeth.2019.108506. View

Matharoo G, Razul M, Poole P . Spectral statistics of the quenched normal modes of a network-forming molecular liquid. J Chem Phys. 2009; 130(12):124512. DOI: 10.1063/1.3099605. View

Gu L, Yu Z, Ma T, Wang H, Li Z, Fan H . Random matrix theory for analysing the brain functional network in lower limb motor imagery. Annu Int Conf IEEE Eng Med Biol Soc. 2020; 2020:506-509. DOI: 10.1109/EMBC44109.2020.9176442. View

Shalev L, Ben-Simon A, Mevorach C, Cohen Y, Tsal Y . Conjunctive Continuous Performance Task (CCPT)--a pure measure of sustained attention. Neuropsychologia. 2011; 49(9):2584-91. DOI: 10.1016/j.neuropsychologia.2011.05.006. View

10.

Bhadola P, Deo N . Targeting functional motifs of a protein family. Phys Rev E. 2016; 94(4-1):042409. DOI: 10.1103/PhysRevE.94.042409. View

11.

Wang R, Zhang Z, Ma J, Yang Y, Lin P, Wu Y . Spectral properties of the temporal evolution of brain network structure. Chaos. 2016; 25(12):123112. DOI: 10.1063/1.4937451. View

12.

Takagi Y, Sakai Y, Lisi G, Yahata N, Abe Y, Nishida S . A Neural Marker of Obsessive-Compulsive Disorder from Whole-Brain Functional Connectivity. Sci Rep. 2017; 7(1):7538. PMC: 5548868. DOI: 10.1038/s41598-017-07792-7. View

13.

Wang R, Wang L, Yang Y, Li J, Wu Y, Lin P . Random matrix theory for analyzing the brain functional network in attention deficit hyperactivity disorder. Phys Rev E. 2016; 94(5-1):052411. DOI: 10.1103/PhysRevE.94.052411. View

14.

Fonov V, Evans A, Botteron K, Almli C, McKinstry R, Collins D . Unbiased average age-appropriate atlases for pediatric studies. Neuroimage. 2010; 54(1):313-27. PMC: 2962759. DOI: 10.1016/j.neuroimage.2010.07.033. View

15.

Mandrekar J . Receiver operating characteristic curve in diagnostic test assessment. J Thorac Oncol. 2010; 5(9):1315-6. DOI: 10.1097/JTO.0b013e3181ec173d. View

16.

Steegen S, Tuerlinckx F, Gelman A, Vanpaemel W . Increasing Transparency Through a Multiverse Analysis. Perspect Psychol Sci. 2016; 11(5):702-712. DOI: 10.1177/1745691616658637. View

17.

Martin C, Peng T, Mahoney M . Predicting trends in the quality of state-of-the-art neural networks without access to training or testing data. Nat Commun. 2021; 12(1):4122. PMC: 8257709. DOI: 10.1038/s41467-021-24025-8. View

18.

van den Heuvel M, Hulshoff Pol H . Exploring the brain network: a review on resting-state fMRI functional connectivity. Eur Neuropsychopharmacol. 2010; 20(8):519-34. DOI: 10.1016/j.euroneuro.2010.03.008. View

19.

Gorgolewski K, Lurie D, Urchs S, Kipping J, Craddock R, Milham M . A correspondence between individual differences in the brain's intrinsic functional architecture and the content and form of self-generated thoughts. PLoS One. 2014; 9(5):e97176. PMC: 4019564. DOI: 10.1371/journal.pone.0097176. View

20.

Yarkoni T, Westfall J . Choosing Prediction Over Explanation in Psychology: Lessons From Machine Learning. Perspect Psychol Sci. 2017; 12(6):1100-1122. PMC: 6603289. DOI: 10.1177/1745691617693393. View