Front Neuroinform. 2020 Nov 30;14:581897. doi: 10.3389/fninf.2020.581897. eCollection 2020.
A Tensor-Based Framework for rs-fMRI Classification and Functional Connectivity Construction.
Frontiers in neuroinformatics
Ali Noroozi, Mansoor Rezghi
Affiliations
Affiliations
- Department of Computer Science, Tarbiat Modares University, Tehran, Iran.
PMID: 33328948
PMCID: PMC7734298 DOI: 10.3389/fninf.2020.581897
Abstract
Recently, machine learning methods have gained lots of attention from researchers seeking to analyze brain images such as Resting-State Functional Magnetic Resonance Imaging (rs-fMRI) to obtain a deeper understanding of the brain and such related diseases, for example, Alzheimer's disease. Finding the common patterns caused by a brain disorder through analysis of the functional connectivity (FC) network along with discriminating brain diseases from normal controls have long been the two principal goals in studying rs-fMRI data. The majority of FC extraction methods calculate the FC matrix for each subject and then use simple techniques to combine them and obtain a general FC matrix. In addition, the state-of-the-art classification techniques for finding subjects with brain disorders also rely on calculating an FC for each subject, vectorizing, and feeding them to the classifier. Considering these problems and based on multi-dimensional nature of the data, we have come up with a novel tensor framework in which a general FC matrix is obtained without the need to construct an FC matrix for each sample. This framework also allows us to reduce the dimensionality and create a novel discriminant function that rather than using FCs works directly with each sample, avoids vectorization in any step, and uses the test data in the training process without forcing any prior knowledge of its label into the classifier. Extensive experiments using the ADNI dataset demonstrate that our proposed framework effectively boosts the fMRI classification performance and reveals novel connectivity patterns in Alzheimer's disease at its early stages.
Copyright © 2020 Noroozi and Rezghi.
Keywords: Alzheimer's disease (AD) classification; dimension reduction; functional connectivity; high order singular value decomposition; tensor
Conflict of interest statement
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
References
- Neuroimage. 2018 Sep;178:370-384 - PubMed
- Neuroimage. 2013 Aug 1;76:332-44 - PubMed
- IEEE Trans Image Process. 2013 Jul;22(7):2723-36 - PubMed
- Alzheimers Dement. 2007 Jul;3(3):186-91 - PubMed
- Front Comput Neurosci. 2013 Nov 22;7:169 - PubMed
- Brain. 2015 Aug;138(Pt 8):2438-50 - PubMed
- Front Neurosci. 2018 Aug 06;12:525 - PubMed
- Proc Natl Acad Sci U S A. 2018 Jan 30;115(5):927-932 - PubMed
- Neuroimage. 2020 May 1;211:116604 - PubMed
- Neuroimage. 2013 Dec;83:937-50 - PubMed
- Proc Natl Acad Sci U S A. 2017 Feb 14;114(7):1690-1695 - PubMed
- Hum Brain Mapp. 2020 Jul;41(10):2808-2826 - PubMed
- Aging Dis. 2015 Nov 17;6(6):437-43 - PubMed
- PLoS One. 2015 Feb 13;10(2):e0115573 - PubMed
- Neural Netw. 2019 Jun;114:119-135 - PubMed
- Cancer. 1950 Jan;3(1):32-5 - PubMed
- Neural Plast. 2016;2016:8501693 - PubMed
- IEEE Trans Biomed Eng. 2019 Mar;66(3):695-709 - PubMed
- Neuroimage. 2007 Apr 1;35(2):488-500 - PubMed
- IEEE Trans Med Imaging. 2020 May;39(5):1419-1429 - PubMed
- IEEE Trans Biomed Eng. 2013 Aug;60(8):2332-8 - PubMed
- Front Neurol. 2016 Sep 19;7:132 - PubMed
- Brain. 2005 Apr;128(Pt 4):773-87 - PubMed
- IEEE Trans Med Imaging. 2020 Feb;39(2):478-487 - PubMed
- Int J Mol Sci. 2013 Jan 10;14(1):1310-22 - PubMed
- Neuroimage. 2002 Jan;15(1):273-89 - PubMed
- Neuroimage. 2011 May 15;56(2):766-81 - PubMed
- Front Neurosci. 2015 Sep 01;9:307 - PubMed
- Neuropsychol Rev. 2014 Mar;24(1):49-62 - PubMed
- Curr Opin Neurobiol. 2019 Apr;55:48-54 - PubMed
- IEEE Trans Med Imaging. 2020 Apr;39(4):985-996 - PubMed
- Inf Process Med Imaging. 2013;23:256-67 - PubMed
- Nat Neurosci. 2019 Nov;22(11):1751-1760 - PubMed
- Neuroimage. 2010 Mar;50(1):81-98 - PubMed
- IEEE Trans Biomed Eng. 2014 Feb;61(2):576-89 - PubMed
- Magn Reson Med. 2006 Aug;56(2):411-21 - PubMed
- Neuroimage. 2012 Nov 15;63(3):1712-9 - PubMed
- Neuroimage. 2011 Apr 1;55(3):856-67 - PubMed
- Front Neurosci. 2019 Jan 10;12:1018 - PubMed
- Front Psychol. 2014 Feb 07;5:74 - PubMed
- Neurology. 2004 Jun 8;62(11):1990-5 - PubMed
- Neurobiol Aging. 2015 Jan;36(1):27-32 - PubMed
- Neuroimage. 2016 Feb 15;127:242-256 - PubMed
- J Neurosci Methods. 2018 Oct 1;308:21-33 - PubMed
- Front Syst Neurosci. 2010 May 14;4:13 - PubMed
- Front Neuroinform. 2018 Sep 07;12:58 - PubMed
- IEEE Trans Med Imaging. 2020 May;39(5):1524-1534 - PubMed
- Brain Connect. 2011;1(1):13-36 - PubMed
- Hum Brain Mapp. 2019 Dec 15;40(18):5424-5442 - PubMed
- PLoS One. 2010 Nov 01;5(11):e13788 - PubMed
- Neuroimage. 2015 Jan 1;104:430-6 - PubMed
- Neuroimage. 2010 Apr 15;50(3):935-49 - PubMed
- Cereb Cortex. 2012 Apr;22(4):854-64 - PubMed
- Hum Brain Mapp. 2016 Sep;37(9):3282-96 - PubMed
- Proc Natl Acad Sci U S A. 2014 Jul 15;111(28):10341-6 - PubMed
- Eur Radiol. 2016 Jan;26(1):244-53 - PubMed
- Neuroimage. 2012 Aug 15;62(2):1257-66 - PubMed
- Front Aging Neurosci. 2018 Nov 09;10:366 - PubMed
- PLoS One. 2019 Feb 22;14(2):e0212582 - PubMed
- Brain Imaging Behav. 2016 Jun;10(2):342-56 - PubMed
- Int Psychogeriatr. 2016 Apr;28(4):529-36 - PubMed
- IEEE Trans Biomed Eng. 2015 Jun;62(6):1623-34 - PubMed
- Brain. 2018 Jan 1;141(1):37-47 - PubMed
- Neuroimage. 2018 Feb 15;167:62-72 - PubMed
- Cereb Cortex. 2014 Mar;24(3):663-76 - PubMed
- IEEE Trans Image Process. 2017 Sep;26(9):4499-4508 - PubMed
Publication Types