Predicting the fMRI signal fluctuation with echo-state neural networks trained on vascular network dynamics

Resting-state functional MRI (rs-fMRI) studies have revealed specific low-frequency hemodynamic signal fluctuations (<0.1 Hz) in the brain, which could be related to oscillations in neural activity through several mechanisms. Although the vascular origin of the fMRI signal is well established, the neural correlates of global rs-fMRI signal fluctuations are difficult to separate from other confounding sources. Recently, we reported that single-vessel fMRI slow oscillations are directly coupled to brain state changes. Here, we used an echo-state network (ESN) to predict the future temporal evolution of the rs-fMRI slow oscillatory feature from both rodent and human brains. rs-fMRI signals from individual blood vessels that were strongly correlated with neural calcium oscillations were used to train an ESN to predict brain state-specific rs-fMRI signal fluctuations. The ESN-based prediction model was also applied to recordings from the Human Connectome Project (HCP), which classified variance-independent brain states based on global fluctuations of rs-fMRI features. The ESN revealed brain states with global synchrony and decoupled internal correlations within the default-mode network.

[1]  Ha Hong,et al.  Performance-optimized hierarchical models predict neural responses in higher visual cortex , 2014, Proceedings of the National Academy of Sciences.

[2]  Jesper Andersson,et al.  A multi-modal parcellation of human cerebral cortex , 2016, Nature.

[3]  R. S. Hinks,et al.  Time course EPI of human brain function during task activation , 1992, Magnetic resonance in medicine.

[4]  Peter Whittle,et al.  Hypothesis Testing in Time Series Analysis. , 1951 .

[5]  B. Biswal,et al.  Hypercapnia Reversibly Suppresses Low-Frequency Fluctuations in the Human Motor Cortex during Rest Using Echo–Planar MRI , 1997, Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism.

[6]  Steen Moeller,et al.  ICA-based artefact removal and accelerated fMRI acquisition for improved resting state network imaging , 2014, NeuroImage.

[7]  Terrence J. Sejnowski,et al.  An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[8]  Steen Moeller,et al.  The Human Connectome Project: A data acquisition perspective , 2012, NeuroImage.

[9]  Thomas T. Liu,et al.  Differences in the resting-state fMRI global signal amplitude between the eyes open and eyes closed states are related to changes in EEG vigilance , 2016, NeuroImage.

[10]  Timothy O. Laumann,et al.  Sources and implications of whole-brain fMRI signals in humans , 2017, NeuroImage.

[11]  SchmidhuberJürgen,et al.  Learning precise timing with lstm recurrent networks , 2003 .

[12]  Laura Leuchs,et al.  Spontaneous pupil dilations during the resting state are associated with activation of the salience network , 2016, NeuroImage.

[13]  Bruce R. Rosen,et al.  Ultra-Slow Single-Vessel BOLD and CBV-Based fMRI Spatiotemporal Dynamics and Their Correlation with Neuronal Intracellular Calcium Signals , 2018, Neuron.

[14]  M. Fukunaga,et al.  Low frequency BOLD fluctuations during resting wakefulness and light sleep: A simultaneous EEG‐fMRI study , 2008, Human brain mapping.

[15]  D. Kleinfeld,et al.  Entrainment of Arteriole Vasomotor Fluctuations by Neural Activity Is a Basis of Blood-Oxygenation-Level-Dependent “Resting-State” Connectivity , 2017, Neuron.

[16]  Mantas Lukosevicius,et al.  A Practical Guide to Applying Echo State Networks , 2012, Neural Networks: Tricks of the Trade.

[17]  Herbert Jaeger,et al.  Reservoir computing approaches to recurrent neural network training , 2009, Comput. Sci. Rev..

[18]  Kevin Murphy,et al.  Towards a consensus regarding global signal regression for resting state functional connectivity MRI , 2017, NeuroImage.

[19]  Herbert Jaeger,et al.  Optimization and applications of echo state networks with leaky- integrator neurons , 2007, Neural Networks.

[20]  Ming Li,et al.  Impact of global signal regression on characterizing dynamic functional connectivity and brain states , 2018, NeuroImage.

[21]  Manuel Schabus,et al.  Hemodynamic cerebral correlates of sleep spindles during human non-rapid eye movement sleep , 2007, Proceedings of the National Academy of Sciences.

[22]  M. Fox,et al.  The global signal and observed anticorrelated resting state brain networks. , 2009, Journal of neurophysiology.

[23]  Jeff H. Duyn,et al.  Temporal dynamics of the BOLD fMRI impulse response , 2005, NeuroImage.

[24]  Peter Ford Dominey,et al.  Reservoir Computing Properties of Neural Dynamics in Prefrontal Cortex , 2016, PLoS Comput. Biol..

[25]  Biyu J. He,et al.  Electrophysiological correlates of the brain's intrinsic large-scale functional architecture , 2008, Proceedings of the National Academy of Sciences.

[26]  Marcel van Gerven,et al.  Modeling the Dynamics of Human Brain Activity with Recurrent Neural Networks , 2016, Front. Comput. Neurosci..

[27]  Harald Haas,et al.  Harnessing Nonlinearity: Predicting Chaotic Systems and Saving Energy in Wireless Communication , 2004, Science.

[28]  E. Golanov,et al.  Spontaneous waves of cerebral blood flow associated with a pattern of electrocortical activity. , 1994, The American journal of physiology.

[29]  Vince D. Calhoun,et al.  Spatio-Temporal Dynamics of Intrinsic Networks in Functional Magnetic Imaging Data Using Recurrent Neural Networks , 2016, Front. Neurosci..

[30]  Shella Keilholz,et al.  The Not-So-Global Blood Oxygen Level-Dependent Signal , 2018, Brain Connect..

[31]  S. H. Curry,et al.  Slow Potential Changes in the Human Brain , 1993, NATO ASI Series.

[32]  Dieter Jaeger,et al.  Infraslow LFP correlates to resting-state fMRI BOLD signals , 2013, NeuroImage.

[33]  B. Biswal,et al.  Functional connectivity in the motor cortex of resting human brain using echo‐planar mri , 1995, Magnetic resonance in medicine.

[34]  Mark S. Cohen,et al.  Simultaneous EEG and fMRI of the alpha rhythm , 2002, Neuroreport.

[35]  Kevin Murphy,et al.  Resting-state fMRI confounds and cleanup , 2013, NeuroImage.

[36]  Mariel G Kozberg,et al.  Resting-state hemodynamics are spatiotemporally coupled to synchronized and symmetric neural activity in excitatory neurons , 2016, Proceedings of the National Academy of Sciences.

[37]  Henry Markram,et al.  Real-Time Computing Without Stable States: A New Framework for Neural Computation Based on Perturbations , 2002, Neural Computation.

[38]  K. Scheffler,et al.  Principles and applications of balanced SSFP techniques , 2003, European Radiology.

[39]  Vinod Menon,et al.  Functional connectivity in the resting brain: A network analysis of the default mode hypothesis , 2002, Proceedings of the National Academy of Sciences of the United States of America.

[40]  Emery N Brown,et al.  Neural oscillations demonstrate that general anesthesia and sedative states are neurophysiologically distinct from sleep , 2017, Current Opinion in Neurobiology.

[41]  M. Berger,et al.  High Gamma Power Is Phase-Locked to Theta Oscillations in Human Neocortex , 2006, Science.

[42]  N. Logothetis,et al.  Neurophysiological investigation of the basis of the fMRI signal , 2001, Nature.

[43]  J. Fell,et al.  The role of phase synchronization in memory processes , 2011, Nature Reviews Neuroscience.

[44]  Stephen M. Smith,et al.  Probabilistic independent component analysis for functional magnetic resonance imaging , 2004, IEEE Transactions on Medical Imaging.

[45]  Kevin Murphy,et al.  The impact of global signal regression on resting state correlations: Are anti-correlated networks introduced? , 2009, NeuroImage.

[46]  Thomas T. Liu,et al.  A component based noise correction method (CompCor) for BOLD and perfusion based fMRI , 2007, NeuroImage.

[47]  Marcel A. J. van Gerven,et al.  Deep Neural Networks Reveal a Gradient in the Complexity of Neural Representations across the Ventral Stream , 2014, The Journal of Neuroscience.

[48]  Catie Chang,et al.  Effects of model-based physiological noise correction on default mode network anti-correlations and correlations , 2009, NeuroImage.

[49]  Jeff H. Duyn,et al.  Modulation of spontaneous fMRI activity in human visual cortex by behavioral state , 2009, NeuroImage.

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

[51]  Yizhen Zhang,et al.  Neural Encoding and Decoding with Deep Learning for Dynamic Natural Vision , 2016, Cerebral cortex.

[52]  Surya Ganguli,et al.  Memory traces in dynamical systems , 2008, Proceedings of the National Academy of Sciences.

[53]  Peter A. Bandettini,et al.  The respiration response function: The temporal dynamics of fMRI signal fluctuations related to changes in respiration , 2008, NeuroImage.

[54]  Martin V. Butz,et al.  Balanced echo state networks , 2012, Neural Networks.

[55]  Y. Benjamini,et al.  Controlling the false discovery rate: a practical and powerful approach to multiple testing , 1995 .

[56]  G. Buzsáki,et al.  Neuronal Oscillations in Cortical Networks , 2004, Science.

[57]  Vince D. Calhoun,et al.  Deep learning for neuroimaging: a validation study , 2013, Front. Neurosci..

[58]  Timothy O. Laumann,et al.  Informatics and Data Mining Tools and Strategies for the Human Connectome Project , 2011, Front. Neuroinform..

[59]  N. Logothetis,et al.  Neurophysiology of the BOLD fMRI Signal in Awake Monkeys , 2008, Current Biology.

[60]  S. Linnainmaa Taylor expansion of the accumulated rounding error , 1976 .

[61]  Hellmut Merkle,et al.  Sensory and optogenetically driven single-vessel fMRI , 2016, Nature Methods.

[62]  Vince D. Calhoun,et al.  Disambiguating the role of blood flow and global signal with partial information decomposition , 2019, NeuroImage.

[63]  Jingyuan E. Chen,et al.  Dissociated patterns of anti‐correlations with dorsal and ventral default‐mode networks at rest , 2017, Human brain mapping.

[64]  Mark D. McDonnell,et al.  Methods for Generating Complex Networks with Selected Structural Properties for Simulations: A Review and Tutorial for Neuroscientists , 2011, Front. Comput. Neurosci..

[65]  James Theiler,et al.  Testing for nonlinearity in time series: the method of surrogate data , 1992 .

[66]  Mark Jenkinson,et al.  MSM: A new flexible framework for Multimodal Surface Matching , 2014, NeuroImage.

[67]  Vince D. Calhoun,et al.  Save the Global: Global Signal Connectivity as a Tool for Studying Clinical Populations with Functional Magnetic Resonance Imaging , 2014, Brain Connect..

[68]  Yoshua Bengio,et al.  Random Search for Hyper-Parameter Optimization , 2012, J. Mach. Learn. Res..

[69]  Ludovica Griffanti,et al.  Automatic denoising of functional MRI data: Combining independent component analysis and hierarchical fusion of classifiers , 2014, NeuroImage.

[70]  Aapo Hyvärinen,et al.  Fast and robust fixed-point algorithms for independent component analysis , 1999, IEEE Trans. Neural Networks.

[71]  V. Haughton,et al.  Frequencies contributing to functional connectivity in the cerebral cortex in "resting-state" data. , 2001, AJNR. American journal of neuroradiology.

[72]  Terrence J. Sejnowski,et al.  Brain-state dependent astrocytic Ca2+ signals are coupled to both positive and negative BOLD-fMRI signals , 2018, Proceedings of the National Academy of Sciences.

[73]  P. Skudlarski,et al.  Brain Connectivity Related to Working Memory Performance , 2006, The Journal of Neuroscience.

[74]  Ilya E. Monosov,et al.  The Basal Forebrain Regulates Global Resting-State fMRI Fluctuations , 2018, Neuron.

[75]  J. Palva,et al.  Infraslow oscillations modulate excitability and interictal epileptic activity in the human cortex during sleep. , 2004, Proceedings of the National Academy of Sciences of the United States of America.

[76]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[77]  Marisa O. Hollinshead,et al.  The organization of the human cerebral cortex estimated by intrinsic functional connectivity. , 2011, Journal of neurophysiology.

[78]  Biyu J. He,et al.  The fMRI signal, slow cortical potential and consciousness , 2009, Trends in Cognitive Sciences.

[79]  Dinggang Shen,et al.  State-space model with deep learning for functional dynamics estimation in resting-state fMRI , 2016, NeuroImage.

[80]  G. Glover,et al.  Respiration‐induced B0 fluctuations and their spatial distribution in the human brain at 7 Tesla , 2002, Magnetic resonance in medicine.

[81]  R W Cox,et al.  AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. , 1996, Computers and biomedical research, an international journal.

[82]  D. Leopold,et al.  Neuronal correlates of spontaneous fluctuations in fMRI signals in monkey visual cortex: Implications for functional connectivity at rest , 2008, Human brain mapping.

[83]  D. Norris,et al.  Very high‐resolution three‐dimensional functional MRI of the human visual cortex with elimination of large venous vessels , 2007, NMR in biomedicine.

[84]  Aapo Hyvärinen,et al.  Group-PCA for very large fMRI datasets , 2014, NeuroImage.

[85]  A. Villringer,et al.  Spontaneous Low Frequency Oscillations of Cerebral Hemodynamics and Metabolism in Human Adults , 2000, NeuroImage.

[86]  N. Logothetis,et al.  The Amplitude and Timing of the BOLD Signal Reflects the Relationship between Local Field Potential Power at Different Frequencies , 2012, The Journal of Neuroscience.

[87]  M. Schölvinck,et al.  Tracking brain arousal fluctuations with fMRI , 2016, Proceedings of the National Academy of Sciences.

[88]  Xin Yu,et al.  Direct imaging of macrovascular and microvascular contributions to BOLD fMRI in layers IV–V of the rat whisker–barrel cortex , 2012, NeuroImage.

[89]  M. Steriade Impact of network activities on neuronal properties in corticothalamic systems. , 2001, Journal of neurophysiology.

[90]  Lutz Jäncke,et al.  Large-scale functional brain networks in human non-rapid eye movement sleep: insights from combined electroencephalographic/functional magnetic resonance imaging studies , 2011, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[91]  Mark Jenkinson,et al.  The minimal preprocessing pipelines for the Human Connectome Project , 2013, NeuroImage.

[92]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[93]  Mark W. Woolrich,et al.  Resting-state fMRI in the Human Connectome Project , 2013, NeuroImage.

[94]  M. Schölvinck,et al.  Subcortical evidence for a contribution of arousal to fMRI studies of brain activity , 2018, Nature Communications.

[95]  G. Buzsáki,et al.  Gamma (40-100 Hz) oscillation in the hippocampus of the behaving rat , 1995, The Journal of neuroscience : the official journal of the Society for Neuroscience.

[96]  Chaozhe Zhu,et al.  Amplitude of low frequency fluctuation within visual areas revealed by resting-state functional MRI , 2007, NeuroImage.

[97]  Michelle Hampson,et al.  Functional connectivity between task-positive and task-negative brain areas and its relation to working memory performance. , 2010, Magnetic resonance imaging.

[98]  Alan P. Koretsky,et al.  3D mapping of somatotopic reorganization with small animal functional MRI , 2010, NeuroImage.

[99]  Jürgen Schmidhuber,et al.  Learning Precise Timing with LSTM Recurrent Networks , 2003, J. Mach. Learn. Res..

[100]  Steven Laureys,et al.  Modulation of the spontaneous hemodynamic response function across levels of consciousness , 2019, NeuroImage.

[101]  Maurizio Corbetta,et al.  The human brain is intrinsically organized into dynamic, anticorrelated functional networks. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[102]  David A. Leopold,et al.  Dynamic functional connectivity: Promise, issues, and interpretations , 2013, NeuroImage.

[103]  Schreiber,et al.  Improved Surrogate Data for Nonlinearity Tests. , 1996, Physical review letters.

[104]  D. Tank,et al.  4 Tesla gradient recalled echo characteristics of photic stimulation‐induced signal changes in the human primary visual cortex , 1993 .

[105]  Congwu Du,et al.  Low-frequency calcium oscillations accompany deoxyhemoglobin oscillations in rat somatosensory cortex , 2014, Proceedings of the National Academy of Sciences.

[106]  M. Steriade,et al.  Natural waking and sleep states: a view from inside neocortical neurons. , 2001, Journal of neurophysiology.

[107]  P. Whittle Hypothesis testing in time series analysis , 1954 .

[108]  Terrence J. Sejnowski,et al.  Cortical travelling waves: mechanisms and computational principles , 2018, Nature Reviews Neuroscience.

[109]  M. Raichle,et al.  Resting states affect spontaneous BOLD oscillations in sensory and paralimbic cortex. , 2008, Journal of neurophysiology.

[110]  Vince D. Calhoun,et al.  A review of group ICA for fMRI data and ICA for joint inference of imaging, genetic, and ERP data , 2009, NeuroImage.

[111]  Johannes Stelzer,et al.  Multimodal assessment of recovery from coma in a rat model of diffuse brainstem tegmentum injury , 2019, NeuroImage.

[112]  Peter A. Bandettini,et al.  Separating respiratory-variation-related fluctuations from neuronal-activity-related fluctuations in fMRI , 2006, NeuroImage.

[113]  P. Matthews,et al.  Distinct patterns of brain activity in young carriers of the APOE e4 allele , 2009, NeuroImage.

[114]  Ravi S. Menon,et al.  Intrinsic signal changes accompanying sensory stimulation: functional brain mapping with magnetic resonance imaging. , 1992, Proceedings of the National Academy of Sciences of the United States of America.

[115]  Roel H. R. Deckers,et al.  Large-amplitude, spatially correlated fluctuations in BOLD fMRI signals during extended rest and early sleep stages. , 2006, Magnetic resonance imaging.

[116]  D. Kleinfeld,et al.  Fluctuations and stimulus-induced changes in blood flow observed in individual capillaries in layers 2 through 4 of rat neocortex. , 1998, Proceedings of the National Academy of Sciences of the United States of America.

[117]  R. Turner,et al.  Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. , 1992, Proceedings of the National Academy of Sciences of the United States of America.

[118]  M. Schölvinck,et al.  Neural basis of global resting-state fMRI activity , 2010, Proceedings of the National Academy of Sciences.

[119]  César Caballero-Gaudes,et al.  Methods for cleaning the BOLD fMRI signal , 2016, NeuroImage.

[120]  S Makeig,et al.  Analysis of fMRI data by blind separation into independent spatial components , 1998, Human brain mapping.

[121]  Marco Leite,et al.  Phase–amplitude coupling and the BOLD signal: A simultaneous intracranial EEG (icEEG) - fMRI study in humans performing a finger-tapping task , 2017, NeuroImage.

[122]  G L Shulman,et al.  INAUGURAL ARTICLE by a Recently Elected Academy Member:A default mode of brain function , 2001 .

[123]  A. Mechelli,et al.  Using deep learning to investigate the neuroimaging correlates of psychiatric and neurological disorders: Methods and applications , 2017, Neuroscience & Biobehavioral Reviews.

[124]  Jeff H. Duyn,et al.  Low-frequency fluctuations in the cardiac rate as a source of variance in the resting-state fMRI BOLD signal , 2007, NeuroImage.

[125]  Nicola Toschi,et al.  Echo State Network models for nonlinear Granger causality , 2019, bioRxiv.

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

[127]  Duncan J. Watts,et al.  Collective dynamics of ‘small-world’ networks , 1998, Nature.

[128]  Albert,et al.  Emergence of scaling in random networks , 1999, Science.

[129]  Emery N. Brown,et al.  Brain correlates of autonomic modulation: Combining heart rate variability with fMRI , 2008, NeuroImage.

[130]  Rafael Malach,et al.  Coupling between pupil fluctuations and resting-state fMRI uncovers a slow build-up of antagonistic responses in the human cortex , 2015, NeuroImage.

[131]  Enzo Tagliazucchi,et al.  Automatic sleep staging using fMRI functional connectivity data , 2012, NeuroImage.

[132]  Klaus Scheffler,et al.  Identifying Respiration-Related Aliasing Artifacts in the Rodent Resting-State fMRI , 2018, Front. Neurosci..

[133]  James A. Anderson,et al.  Neurocomputing: Foundations of Research , 1988 .

[134]  A. Redish,et al.  Measuring fundamental frequencies in local field potentials , 2004, Journal of Neuroscience Methods.

[135]  Xu Zhang,et al.  Deep Learning Methods to Process fMRI Data and Their Application in the Diagnosis of Cognitive Impairment: A Brief Overview and Our Opinion , 2018, Front. Neuroinform..

[136]  B. Rockstroh,et al.  Slow potentials of the cerebral cortex and behavior. , 1990, Physiological reviews.