Regularized Hypothesis Testing in Random Fields with Applications to Neuroimaging

Authors

  • Oscar S. Dalmau-Cedeño Centro de Investigación en Matemáticas, CIMAT A. C., Mexico
  • Dora E. Alvarado-Carrillo Centro de Investigación en Matemáticas, CIMAT A. C., Mexico https://orcid.org/0000-0003-1984-7546
  • José Luis Marroquín Centro de Investigación en Matemáticas, CIMAT A. C. México

DOI:

https://doi.org/10.17488/RMIB.41.2.2

Keywords:

Regularized hypothesis test, Markovian random fields, Bayesian estimation, functional Magnetic resonance imaging

Abstract

The task of determining for which elements of a random field (e.g., pixels in an image) a certain null hypothesis may be rejected is a relevant problem in several scientific areas. In the current contribution, we introduce a new method for performing this task, the regularized hypothesis testing (RHT) method, focusing on its use in neuroimaging re- search. RHT is based on the formulation of the hypothesis testing task as a Bayesian estimation problem, with the previous application of a Markovian random field. The latter allows for the incorporation of local spatial informa- tion and considers different noise models, including spatially correlated noise. In tests on synthetic data showing regular activation levels on uncorrelated noise fields, RHT furnished a true positive rate (TPR) of 0.97, overcoming the state-of-the-art morphology-based hypothesis testing (MBHT) method and the traditional family-wise error rate (FWER) method, which afforded 0.93 and 0.58, respectively. For fields with highly correlated noise, the TPR provided by RHT was 0.65, and by MBHT and FWER was 0.35 and 0.29, respectively. For tests utilizing real func- tional magnetic resonance imaging (fMRI) data, RHT managed to locate the activation regions when 60% of the original signal were removed, while MBHT located only one region and FWER located none.

Downloads

Download data is not yet available.

References

Cole MW, Ito T, Bassett DS, Schultz DH. Activity flow over resting-state networks shapes cognitive task activations. Nature Neuroscience. 2016 October; 19(0): p. 1718–1726, doi: 10.1038/nn.4406.

Poldrack RA, Baker CI, Durnez J, Gorgolewski KJ, Matthews PM, Munafo MR, et al. Scanning the horizon: towards transparent and reproducible neuroimaging research. Nature Reviews Neuroscience. 2017 January; 18(0): p. 115–126, doi: 10.1038/nrn.2016.167.

Mejia A, Yue YR, Bolin D, Lindren F, Lindquist MA. A Bayesian General Linear Modeling Approach to Cortical Surface fMRI Data Analysis. Journal of the American Statistical Association. 2019 April; 0(0): p. 1-20, doi: 10.1080/01621459.2019.1611582.

Flores-Leal M, Sacristán-Rock E, Jiménez-Ángeles L, Leehan JA. Primed low frequency transcranial magnetic stimulation effects on smoking cue-induced craving. Revista mexicana de ingeniería biomédica. 2016 April; 37(1): p. 39-48, doi: 10.17488/RMIB.37.1.3.

Acosta-Franco JA, Saiffe-Farías AF, Gómez-Velázquez FR, González-Garrido AA, Romo-Vázquez RC. Activación Hemisférica Cerebral en Adultos Jóvenes Mientras Ejecutan Tareas Ortográficas Utilizando fMRI. Revista Mexicana de Ingeniería Biomédica. 2019 August; 40(2): p. 1-10, doi: 10.17488/RMIB.40.2.5.

Lindquist MA, Mejia A. Zen and the Art of Multiple Comparisons. Psychosomatic medicine. 2015 February; 77(2): p. 114-125, doi: 10.1097/PSY.0000000000000148.

Woo CW, Krishnan A, Wage TD. Woo, C.-W., Krishnan, A., & Wager, T. D. (2014). Cluster-extent based thresholding in fMRI analyses: Pitfalls and recommendations. NeuroImage. 2013 December; 91(0): p. 412–419, doi: 10.1016/j.neuroimage.2013.12.058.

Abdi H. The Bonferonni and Sidak corrections for multiple comparisons. In Salkind NJ, editor. Encyclopedia of Measurement and Statistics. Thousand Oaks, CA, USA: Sage Publications; 2007. p. 103-107, doi: 10.4135/9781412952644.

Hayasakaa S, Nichols TE. Validating cluster size inference: random field and permutation methods. NeuroImage. 2003 December; 20(4): p. 2343-2356, doi: 10.1016/j.neuroimage.2003.08.003.

Hagler DJ, Saygin AP, Sereno MI. Smoothing and cluster thresholding for cortical surface-based group analysis of fMRI data. NeuroImage. 2006 December; 33(4): p. 1093-1103, doi: 10.1016/j.neuroimage.2006.07.036.

Eklund A, Nichols TE, Knutsson H. Cluster failure: Why fMRI inferences for spatial extent have inflated false-positive rates. Proceedings of the National Academy of Sciences. 2016 July; 113(28): p. 7900-7905, doi: 10.1073/pnas.1602413113.

Marroquin JL, Biscay RJ, Ruiz-Correa S, Alba A, Ramirez R, Armony JL. Morphology-based hypothesis testing in discrete random fields: A non-parametric method to address the multiple-comparison problem in neuroimaging. NeuroImage. 2011 February; 56(4): p. 1954-1967, doi: 10.1016/j.neuroimage.2011.09.051.

Swets JA. Signal detection theory and ROC analysis in psychology and diagnostics: Collected papers. 1st ed. Press P, editor. New York, USA.: Psychology Press; 2014, doi: 10.4324/9781315806167.

Hochberg Y, Tamhane AC. Distribution-Free and Robust Procedures. 1st ed. Shube B, editor. New York, USA.: Wiley Online Library; 1987, doi: 10.1002/9780470316672.

Pantazis D. General Linear Modeling of Magnetoencephalography Data. In Bronzino JD, Peterson DR, editors. Biomedical Signals, Imaging, and Informatics. Boca Raton, Florida, USA.: CRC Press; 2014. p. 33-350, doi: 10.1201/b15468.

Benjamini Y. Discovering the false discovery rate. Journal of the Royal Statistical Society: Series B (Statistical Methodology). 2010 August; 72(4): p. 405-416, doi:10.1111/j.1467-9868.2010.00746.x.

Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I. Controlling the false discovery rate in behavior genetics research. Behavioural brain research. 2001 November; 125(1-2): p. 279-284, doi: 10.1016/S0166-4328(01)00297-2.

Friston KJ, Ashburner JT, Kiebel SJ, Nichols TE, Penny WD. Statistical parametric mapping: the analysis of functional brain images. 1st ed. Penny W, editor. London, UK.: Academic press; 2011, doi :10.1016/B978-0-12-372560-8.X5000-1.

Nichols TE, Holmes AP. Nonparametric permutation tests for functional neuroimaging: a primer with examples. Human brain mapping. 2002 January; 15(1): p. 1-25, doi: 10.1002/hbm.1058.

Welvaert M, Rosseel Y. On the definition of signal-to-noise ratio and contrast-to-noise ratio for fMRI data. PloS one. 2013 November; 8(11): p. 1-10, doi: 10.1371/journal.pone.0077089.

Zhang H, Nichols TE, Johnson TD. Cluster mass inference via random field theory. Neuroimage. 2009 January; 44(1): p. 51-61, doi: 10.1016/j.neuroimage.2008.08.017.

Rivera M, Ocegueda O, Marroquin JL. Entropy-controlled quadratic Markov measure field models for efficient image segmentation. IEEE Transactions on Image Processing. 2007 December; 16(12): p. 3047-3057, doi: 10.1109/TIP.2007.909384.

Dalmau O, Rivera M. Beta-Measure for Probabilistic Segmentation. In Advances in Artificial Intelligence: 9th Mexican International Conference on Artificial Intelligence, MICAI 2010, Pachuca, Mexico, November 8-13, 2010, Proceedings, Part I; 2010; Berlin, Heidelberg: Springer Berlin Heidelberg. p. 312-324, doi: 10.1007/978-3-642-16761-4_28.

Besag J. On the statistical analysis of dirty pictures. Journal of the Royal Statistical Society B. 1986 January; 48(3): p. 48-259, doi: 10.1080/02664769300000059.

Geman S, Geman D. Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1984 November; PAMI-6(6): p. 721-741, doi: 10.1109/TPAMI.1984.4767596.

Boykov Y, Veksler O, Zabih R. Fast Approximate Energy Minimization via Graph Cuts. IEEE Trans. Pattern Anal. Mach. Intell. 2001 November; 23(11): p. 1222-1239, doi: 10.1109/34.969114.

Wu YL, Agrawal D, Abbadi AE. Applying the golden rule of sampling for query estimation. ACM SIGMOD Record. 2001 May; 10(2): p. 449-460, doi: 10.1145/376284.375724.

Devroye L. General Principles in Random Variate Generation. In Springer-Verlag , editor. Nonuniform random variate generation. New York, New York 10010, U.S.A: Elsevier; 2006. p. 27-36, doi: 10.1016/S0927-0507(06)13004-2.

Drew JH, Evans DL, Glen AG, Leemis LM. Transformations of random variables. In Grassmann WK, editor. Computational Probability. Boston, MA, USA: Springer; 2017. p. 47-46, doi: 10.1007/978-0-387-74676-0.

Oldham KB, Myland JC, Spanier J. The Error Function erf (x) and Its Complement erfc (x). In Science S, editor. An Atlas of Functions. New York, USA: Springer; 2008. p. 405-415, doi: 10.1007/978-0-387-48807-3_41.

Besag J. Statistical Analysis of Non-Lattice Data. Journal of the Royal Statistical Society. Series D (The Statistician). 1975 January; 24(3): p. 179-195, doi: 10.2307/2987782.

Friston GRaK. Single subject epoch (block) auditory fMRI activation data. 1999. URL: http://www. fil. ion. ucl. ac. uk/spm/data/auditory.

Raz J, Zheng H, Ombao H, Turetsky B. Statistical tests for fMRI based on experimental randomization. NeuroImage. 2003 June; 19(2): p. 226-232, doi: 10.1016/s1053-8119(03)00115-0.

Amunts K, Morosan P, Hilbig H, Zilles K. Chapter 36 - Auditory System. In Mai JK, Paxinos G, editors. The Human Nervous System (Third Edition). New York, USA.: Academic Press; 2012. p. 1270-1300, doi: 10.1016/C2009-0-02721-4.

Bizley J. Audition. In Conn PM, editor. Conn's Translational Neuroscience. New York, USA: Elsevier; 2017. p. 579-598, doi: 10.1016/B978-0-12-802381-5.00042-7.

Downloads

Published

2020-06-13

How to Cite

Dalmau-Cedeño, O. S. ., Alvarado-Carrillo, D. E. ., & Marroquín, J. L. (2020). Regularized Hypothesis Testing in Random Fields with Applications to Neuroimaging. Revista Mexicana De Ingenieria Biomedica, 41(2), 22–39. https://doi.org/10.17488/RMIB.41.2.2

Issue

Section

Research Articles

Dimensions Citation

Similar Articles

You may also start an advanced similarity search for this article.