A Tensor-Based Framework for rs-fMRI Classification and Functional Connectivity Construction

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.

[1]  Daniele Marinazzo,et al.  Advancing functional connectivity research from association to causation , 2019, Nature Neuroscience.

[2]  Gang Li,et al.  High‐order resting‐state functional connectivity network for MCI classification , 2016, Human brain mapping.

[3]  C. Sorg,et al.  Progressively Disrupted Intrinsic Functional Connectivity of Basolateral Amygdala in Very Early Alzheimer’s Disease , 2016, Front. Neurol..

[4]  Joos Vandewalle,et al.  A Multilinear Singular Value Decomposition , 2000, SIAM J. Matrix Anal. Appl..

[5]  Jubin Abutalebi,et al.  The impact of bilingualism on brain reserve and metabolic connectivity in Alzheimer's dementia , 2017, Proceedings of the National Academy of Sciences.

[6]  Marie Chupin,et al.  Automatic classi fi cation of patients with Alzheimer ' s disease from structural MRI : A comparison of ten methods using the ADNI database , 2010 .

[7]  Eswar Damaraju,et al.  Tracking whole-brain connectivity dynamics in the resting state. , 2014, Cerebral cortex.

[8]  Emily B Fox,et al.  Statistical model-based approaches for functional connectivity analysis of neuroimaging data , 2019, Current Opinion in Neurobiology.

[9]  A. Brickman,et al.  Reconsidering harbingers of dementia: progression of parietal lobe white matter hyperintensities predicts Alzheimer's disease incidence , 2015, Neurobiology of Aging.

[10]  Silva Hecimovic,et al.  Phospholipids and Alzheimer’s Disease: Alterations, Mechanisms and Potential Biomarkers , 2013, International journal of molecular sciences.

[11]  Kathryn Ziegler-Graham,et al.  Forecasting the global burden of Alzheimer’s disease , 2007, Alzheimer's & Dementia.

[12]  Si Zhang,et al.  Graph convolutional networks: a comprehensive review , 2019, Computational Social Networks.

[13]  Brian Barton,et al.  Visual cortex in aging and Alzheimer's disease: changes in visual field maps and population receptive fields , 2012, Front. Psychol..

[14]  Lei Wang,et al.  Functional Brain Network Classification With Compact Representation of SICE Matrices , 2015, IEEE Transactions on Biomedical Engineering.

[15]  S. Rombouts,et al.  Loss of ‘Small-World’ Networks in Alzheimer's Disease: Graph Analysis of fMRI Resting-State Functional Connectivity , 2010, PloS one.

[16]  P. Schulz,et al.  Early Detection of Alzheimer’s Disease Using Non-invasive Near-Infrared Spectroscopy , 2018, Front. Aging Neurosci..

[17]  J. Gabrieli,et al.  Memory encoding in Alzheimer's disease: an fMRI study of explicit and implicit memory. , 2005, Brain : a journal of neurology.

[18]  A. Camins,et al.  Current Research Therapeutic Strategies for Alzheimer's Disease Treatment , 2016, Neural plasticity.

[19]  Akram Bakkour,et al.  The effects of aging and Alzheimer's disease on cerebral cortical anatomy: Specificity and differential relationships with cognition , 2013, NeuroImage.

[20]  Kathryn Roeder,et al.  Global spectral clustering in dynamic networks , 2018, Proceedings of the National Academy of Sciences.

[21]  Selin Aviyente,et al.  Multi-scale higher order singular value decomposition (MS-HoSVD) for resting-state FMRI compression and analysis , 2017, 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[22]  L. Baxter,et al.  Longitudinal changes in cognition and behavior in asymptomatic carriers of the APOE e4 allele , 2004, Neurology.

[23]  Mark W. Woolrich,et al.  Multi-subject hierarchical inverse covariance modelling improves estimation of functional brain networks , 2018, NeuroImage.

[24]  Reinhold Schmidt,et al.  A comprehensive analysis of resting state fMRI measures to classify individual patients with Alzheimer's disease , 2018, NeuroImage.

[25]  Emiliano Bruner,et al.  The cerebellum in Alzheimer's disease: evaluating its role in cognitive decline. , 2018, Brain : a journal of neurology.

[26]  Chaogan Yan,et al.  DPARSF: A MATLAB Toolbox for “Pipeline” Data Analysis of Resting-State fMRI , 2010, Front. Syst. Neurosci..

[27]  Gustavo Deco,et al.  Can sliding-window correlations reveal dynamic functional connectivity in resting-state fMRI? , 2016, NeuroImage.

[28]  Xiaojun Bi,et al.  Early Alzheimer's disease diagnosis based on EEG spectral images using deep learning , 2019, Neural Networks.

[29]  L. Waller,et al.  GraphVar 2.0: A user-friendly toolbox for machine learning on functional connectivity measures , 2018, Journal of Neuroscience Methods.

[30]  Yang Li,et al.  Learning Brain Connectivity Sub-networks by Group- Constrained Sparse Inverse Covariance Estimation for Alzheimer's Disease Classification , 2018, Front. Neuroinform..

[31]  Dinggang Shen,et al.  Brain Network Construction and Classification Toolbox (BrainNetClass) , 2019, ArXiv.

[32]  Catie Chang,et al.  Time–frequency dynamics of resting-state brain connectivity measured with fMRI , 2010, NeuroImage.

[33]  Marion Smits,et al.  Early-stage differentiation between presenile Alzheimer’s disease and frontotemporal dementia using arterial spin labeling MRI , 2015, European Radiology.

[34]  Dimitri Van De Ville,et al.  On spurious and real fluctuations of dynamic functional connectivity during rest , 2015, NeuroImage.

[35]  Dimitri Van De Ville,et al.  Principal components of functional connectivity: A new approach to study dynamic brain connectivity during rest , 2013, NeuroImage.

[36]  J. O'Brien,et al.  Subcortical volume changes in dementia with Lewy bodies and Alzheimer's disease. A comparison with healthy aging , 2015, International Psychogeriatrics.

[37]  Dan Hu,et al.  A toolbox for brain network construction and classification (BrainNetClass) , 2019, Human brain mapping.

[38]  Mansoor Rezghi A Novel Fast Tensor-Based Preconditioner for Image Restoration , 2017, IEEE Transactions on Image Processing.

[39]  Leonardo L. Gollo,et al.  Time-resolved resting-state brain networks , 2014, Proceedings of the National Academy of Sciences.

[40]  Suvrit Sra,et al.  A new metric on the manifold of kernel matrices with application to matrix geometric means , 2012, NIPS.

[41]  D. Leys,et al.  Topography of Cortical Microbleeds in Alzheimer's Disease with and without Cerebral Amyloid Angiopathy: A Post-Mortem 7.0-Tesla MRI Study. , 2015, Aging and disease.

[42]  M. Politis,et al.  Applications of amyloid, tau, and neuroinflammation PET imaging to Alzheimer's disease and mild cognitive impairment , 2019, Human brain mapping.

[43]  J. Jacobsen,et al.  Why musical memory can be preserved in advanced Alzheimer's disease. , 2015, Brain : a journal of neurology.

[44]  Mansoor Rezghi,et al.  Generalized low-rank approximation of matrices based on multiple transformation pairs , 2020, Pattern Recognit..

[45]  Dinggang Shen,et al.  Sparse temporally dynamic resting-state functional connectivity networks for early MCI identification , 2016, Brain Imaging and Behavior.

[46]  Amirali Kazeminejad,et al.  Topological Properties of Resting-State fMRI Functional Networks Improve Machine Learning-Based Autism Classification , 2019, Front. Neurosci..

[47]  Daoqiang Zhang,et al.  Multimodal classification of Alzheimer's disease and mild cognitive impairment , 2011, NeuroImage.

[48]  Philip S. Yu,et al.  Multi-way Multi-level Kernel Modeling for Neuroimaging Classification , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[49]  Lei Guo,et al.  Predicting functional cortical ROIs via DTI-derived fiber shape models. , 2012, Cerebral cortex.

[50]  Toshimitsu Musha,et al.  EEG Markers for Characterizing Anomalous Activities of Cerebral Neurons in NAT (Neuronal Activity Topography) Method , 2013, IEEE Transactions on Biomedical Engineering.

[51]  Jing Li,et al.  Learning brain connectivity of Alzheimer's disease by sparse inverse covariance estimation , 2010, NeuroImage.

[52]  Stephen M. Smith,et al.  The future of FMRI connectivity , 2012, NeuroImage.

[53]  N. Ayache,et al.  Log‐Euclidean metrics for fast and simple calculus on diffusion tensors , 2006, Magnetic resonance in medicine.

[54]  Tianzi Jiang,et al.  Regional coherence changes in the early stages of Alzheimer’s disease: A combined structural and resting-state functional MRI study , 2007, NeuroImage.

[55]  Philip S. Yu,et al.  Spatio-Temporal Tensor Analysis for Whole-Brain fMRI Classification , 2016, SDM.

[56]  Emily L. Dennis,et al.  Functional Brain Connectivity Using fMRI in Aging and Alzheimer’s Disease , 2014, Neuropsychology Review.

[57]  Mark W. Woolrich,et al.  Optimising network modelling methods for fMRI , 2019, NeuroImage.

[59]  Dinggang Shen,et al.  Deep Learning of Static and Dynamic Brain Functional Networks for Early MCI Detection , 2020, IEEE Transactions on Medical Imaging.

[60]  Vince D. Calhoun,et al.  Classification and Prediction of Brain Disorders Using Functional Connectivity: Promising but Challenging , 2018, Front. Neurosci..

[61]  W. Youden,et al.  Index for rating diagnostic tests , 1950, Cancer.

[62]  Andrew A. Berlin,et al.  A 3D Probabilistic Deep Learning System for Detection and Diagnosis of Lung Cancer Using Low-Dose CT Scans , 2019, IEEE Transactions on Medical Imaging.

[63]  Selin Aviyente,et al.  Tensor Based Temporal and Multilayer Community Detection for Studying Brain Dynamics During Resting State fMRI , 2019, IEEE Transactions on Biomedical Engineering.

[64]  Kun Ho Lee,et al.  Hybrid multivariate pattern analysis combined with extreme learning machine for Alzheimer’s dementia diagnosis using multi-measure rs-fMRI spatial patterns , 2019, PloS one.

[65]  Hongli Lv,et al.  Denoising 3D Magnetic Resonance Images Based on Low-Rank Tensor Approximation With Adaptive Multirank Estimation , 2019, IEEE Access.

[66]  Jean-Baptiste Poline,et al.  A Novel Sparse Group Gaussian Graphical Model for Functional Connectivity Estimation , 2013, IPMI.

[67]  M. Gilardi,et al.  Magnetic resonance imaging biomarkers for the early diagnosis of Alzheimer's disease: a machine learning approach , 2015, Front. Neurosci..

[68]  Daniel A. Handwerker,et al.  Periodic changes in fMRI connectivity , 2012, NeuroImage.

[69]  Bram van Ginneken,et al.  Stacked Bidirectional Convolutional LSTMs for Deriving 3D Non-Contrast CT From Spatiotemporal 4D CT , 2020, IEEE Transactions on Medical Imaging.

[70]  Paul J. Laurienti,et al.  Defining nodes in complex brain networks , 2013, Front. Comput. Neurosci..

[71]  Daoqiang Zhang,et al.  Integration of Network Topological and Connectivity Properties for Neuroimaging Classification , 2014, IEEE Transactions on Biomedical Engineering.

[72]  Sung Won Park Multifactor Analysis for fMRI Brain Image Classification by Subject and Motor Task , 2011 .

[73]  Vin de Silva,et al.  Tensor rank and the ill-posedness of the best low-rank approximation problem , 2006, math/0607647.

[74]  Xiang Ji,et al.  Representing and Retrieving Video Shots in Human-Centric Brain Imaging Space , 2013, IEEE Transactions on Image Processing.

[75]  Heye Zhang,et al.  Privileged Modality Distillation for Vessel Border Detection in Intracoronary Imaging , 2019, IEEE Transactions on Medical Imaging.

[76]  Dimitri Van De Ville,et al.  Identifying Network Correlates of Brain States Using Tensor Decompositions of Whole-Brain Dynamic Functional Connectivity , 2013, 2013 International Workshop on Pattern Recognition in Neuroimaging.

[77]  Karl J. Friston Functional and Effective Connectivity: A Review , 2011, Brain Connect..

[78]  G. Ji,et al.  Changes in Thalamic Connectivity in the Early and Late Stages of Amnestic Mild Cognitive Impairment: A Resting-State Functional Magnetic Resonance Study from ADNI , 2015, PloS one.

[79]  N. Tzourio-Mazoyer,et al.  Automated Anatomical Labeling of Activations in SPM Using a Macroscopic Anatomical Parcellation of the MNI MRI Single-Subject Brain , 2002, NeuroImage.