CNS 2018 Workshop, July 17, 2018, Allen Institute, Seattle, USA

In this workshop we will explore Dynamic Functional Connectivity on different temporal and spatial scales. We aim to review recent results and put them in perspective to understand common points and discrepancies across different neuroimaging communities. In particular, we will target the difficulties faced by methodological approaches when bridging scales due to the differences in how neural dynamics are described.
As an example, similar results about the sources that contribute to dynamic connectivity patterns have been reported on different scales. On the one hand, there are changes in global coherence, sometimes described as standing or traveling waves. On the other hand, there are causal interactions between brain regions/neuronal populations which can be extracted by considering time delays. Ideally, the workshop will help to identify opportunities that have thus far remained unexplored.

Organizers
Katharina Glomb 
Joana Cabral

Confirmed Speakers
Petra Ritter, Charité Berlin, Germany 
Jérémie Lefebvre, Krembil Research Institute, Canada
Gijs Plomp, University of Fribourg, Switzerland
Amrit Kashyap, Shella Keilholz’ lab, Georgia Tech, USA
Sebastien Naze, IBM, USA
Louis-David Lord, University of Oxford, UK
Joana Cabral, University of Oxford, UK
Katharina Glomb, Hagmann Lab, Switzerland
Registration
Attendance to all CNS workshops is subject to registration at the conference website.

Location
All CNS workshops will be held at the Allen Institute (map)

Schedule
This will be a 1-day workshop, happening on the 17th of July.

The order of speakers is tentative.

09:00 - 09:10     - Welcome 

09:10 - 09:55     - Amrit Kashyap: 
Brain dynamics viewed through BOLD, electrophysiology and computational modeling
09:55 - 10:30     - Joana Cabral: Mechanistic network models of MEG and fMRI functional connectivity

10:30 - 11:00     - Coffee break

11:00 - 11:45     - Louis-David Lord: Characterization of the brain’s dynamical repertoire in the psychedelic state
11:45 - 12:30 - Sebastien Naze: Sensitivity analysis of the connectome harmonics and implications in neurodegenerative diseases

12:30 - 14:00     - Lunch break

14:00 - 14:45     - Jeremie Lefebvre: State-Dependent Entrainment of Cortical Oscillations with Periodic Stimulation
14:45 - 15:30     - Gijs Plomp: Fast directed interactions between brain areas and cortical layers

15:30 - 16:00     - Coffee break

16:00 - 16:45     - Petra Ritter:
Inferring multi-scale neural mechanisms with brain network modelling
16:45 - 17:30     - Katharina Glomb: Graph signal processing for anatomically constrained source-reconstructed EEG data


Abstracts

Amrit Kashyap

Brain dynamics viewed through BOLD, electrophysiology and computational modeling.


The BOLD signal fluctuations reflect the complex coordination of dynamic brain activity, in conjunction with hemodynamics and physiological noise. Combining invasive recordings from microelectrodes with MRI, our work shows that different neurophysiological processes have distinct spatiotemporal signatures. For example, a quasi-periodic spatiotemporal pattern is linked to infraslow electrical activity and arousal levels, while time-varying BOLD correlation tends to reflect changes in the coordination of higher frequency electrical activity. At even slower time scales, the clustering of windowed functional connectivity aggregates activity into distinct states. Computational studies have tried to characterize this complex system that exhibits structure over a wide range of spatial and temporal scales using noise-driven dynamical systems based on neural mass models (NMMs). We have shown that existing models do not recapitulate time-varying correlation well and capture only certain aspects of the quasiperiodic patterns. The discrepancies suggest that empirical characterization of brain dynamics reflecting neurophysiological processes has the potential to better constrain and improve computational models and therefore improve our understanding on how the brain coordinates the information flow between different brain regions.




Joana Cabral*

Mechanistic network models of MEG and fMRI functional connectivity


* Department of Psychiatry, University of Oxford, UK

* Life and Health Sciences Research Institute, University of Minho, Portugal


Over the last decade, a number of reduced whole-brain network models have been used to investigate the activity emerging from the interplay between brain areas when embedded in the neuroanatomical connectome. With different degrees of abstraction, these models propose different mechanisms for the emergence of slow (<0.1 Hz) spatio-temporally organized fluctuations in the BOLD signal but the link with the power modulations of faster electrophysiological rhythms detected with EEG and MEG has only been addressed in a few studies. In my talk, I will present 2 candidate scenarios for the emergence of MEG power modulations. Both scenarios rely on a mechanism of metastable synchronization but the genesis of power fluctuations in the alpha- and beta-frequency bands depends on the assumptions made for the intrinsic node dynamics and the transmission delays (Cabral et al., 2014, Deco et al., 2017).


References:
Cabral J, Luckhoo H, Woolrich M, Joensson M, Mohseni H, Baker A, Kringelbach ML, Deco G. Exploring mechanisms of spontaneous functional connectivity in MEG: how delayed network interactions lead to structured amplitude envelopes of band-pass filtered oscillations. Neuroimage. 2014 Apr 15;90:423-35.

Deco G, Cabral J, Woolrich MW, Stevner AB, Van Hartevelt TJ, Kringelbach ML. Single or multiple frequency generators in on-going brain activity: A mechanistic whole-brain model of empirical MEG data. Neuroimage. 2017 May 15;152:538-50.




Louis-David Lord*

Characterization of the brain’s dynamical repertoire in the psychedelic state


* Department of Psychiatry, University of Oxford, UK


Brain activity can be understood as the exploration of a dynamical landscape of activity configurations over both space and time. Accordingly, brain dynamics have been described in terms of spontaneous transitions within a repertoire of discrete metastable states of functional connectivity (FC), or “FC states”, which underlie different mental processes. How disruptions in the brain’s dynamical landscape might influence behavior and cognition remains unclear. We therefore investigated changes in the brain’s dynamical repertoire in a rare fMRI dataset consisting of healthy participants intravenously injected with the psychedelic compound psilocybin (the psychoactive ingredient in “magic mushrooms”).We specifically employed a novel data-driven approach to study brain dynamics in the psychedelic state, which focuses on the dominant FC pattern captured by the leading eigenvector of dynamic FC matrices, and enables the identification of recurrent FC patterns over time (“FC-states”) via unsupervised clustering of the leading eigenvectors across subjects and experimental conditions. This approach enabled us to investigate the probability of occurrence of and transition profiles between FC-states extracted from fMRI data in healthy participants under the influence of psilocybin.




Sebastien Naze

Sensitivity analysis of the connectome harmonics and implications in neurodegenerative diseases.

Sebastien Naze*, Selen Atasoy # , Timothée Proix + , James Kozloski*


* : Computational Biology Center, IBM Research, Yorktown Heights, New York, United States

# : Department of Psychiatry, University of Oxford, England, United Kingdom

+ : Department of Neuroscience, Brown University, Providence, Rhode Island, United States


Decomposition of the connectome into Laplacian eigenmodes has been suggested to relate to several neurodegenerative disorders such as dementia, Alzheimer and possibly Huntington’s disease (Raj et al., 2012). Recently, a new framework called connectome harmonics proposed that the default-mode network is also present in the low frequency modes of the Laplacian decomposition (Atasoy et al. 2016, 2017). In short, the discrete Laplacian operator is constructed by combining long-distance white-matter connectivity estimated from diffusion MRI data and local gray matter structural connectivity of the cortical surface. The resulting matrix is decomposed into Laplacian eigenmodes, which are called ‘connectome harmonics’. Through a sensitivity analysis of the connectome harmonics computation that we incorporated into an existing open-source pre-processing pipeline (Proix et al., 2016), we describe how robust harmonic patterns can be obtained from diffusion MRI data and tractography. Specifically, we investigate how the balance of local versus long-range connectivity and the selective trimming of tracks based on their distance to cortical mesh affect the spatial patterns forming on the cortical surface. The impact of several tractography algorithms (i.e. probabilistic vs deterministic) and associated parameters (fractional anisotropy cutoff at fiber bounds, minimum tract lengths) on harmonics robustness are also evaluated. We perform our sensitivity analysis using imaging data from healthy subjects made publicly available by the Human Connectome Project (HCP). We find that the number of tracks retained to compute the graph Laplacian based on the shortest distance of their extremities to the cortical surface is an influential factor of the connectome harmonics reliability. Small changes in the shape of the cortical surface that occur during the downsampling of the cortical mesh can also alter the higher frequency harmonics, but low-frequency connectome harmonics are generally robust to processing method variations. We discuss that prospective disease structural changes, such as the atrophy of inter-hemispheric callosal fibers observed in certain pathologies such as Huntington’s disease (McColgan et al., 2017), can be applied to the benchmarked healthy connectome to predict functional deficits associated with the disease conditions.


References:

Atasoy S, Donnelly I, Pearson J (2016) Human brain networks function in connectome-specific harmonic waves. Nature Communications 7:10340.


Atasoy S, Deco G, Kringelbach ML, Pearson J (2017) Harmonic Brain Modes: A Unifying Framework for Linking Space and Time in Brain Dynamics. The Neuroscientist:107385841772803.


McColgan P, Seunarine KK, Gregory S, Razi A, Papoutsi M, Long JD, Mills JA, Johnson E, Durr A, Roos RA (2017) Topological length of white matter connections predicts their rate of atrophy in premanifest Huntington’s disease. JCI insight 2.


Proix T, Spiegler A, Schirner M, Rothmeier S, Ritter P, Jirsa VK (2016) How do parcellation size and short-range connectivity affect dynamics in large-scale brain network models? NeuroImage 142:135–149.


Raj A, Kuceyeski A, Weiner M (2012) A network diffusion model of disease progression in dementia. Neuron 73:1204–1215.



Jeremie Lefebvre*

State-Dependent Entrainment of Cortical Oscillations with Periodic Stimulation


*Krembil Research Institute, University Health Network, Toronto, Ontario, Canada

Department of Mathematics and Institute for Biomaterials and Biomedical Engineering, University of Toronto, Canada


Promising experimental findings over the last two decades have sparked a strong interest in using electromagnetic brain stimulation in clinical practice to treat a variety of neurophysiological disorders, such as depression, Alzheimer’s and Parkinson. One strategy is to use periodic waveforms to engage brain oscillations - endogenous non-linear rhythms caused by the synchronous firing of neurons – to promote and hopefully upregulate neural communication. While relatively well characterized at the scale of individual cells, the effect of varying electromagnetic fields on populations and large-scale brain dynamics remains poorly understood. Here we have combined computational and mathematical approaches to understand how brain state fluctuations influence the response of the thalamo-cortical system to periodic stimulation. Specifically, we have examined how rest and task-engaged states –which correspond to different oscillatory regimes - shape the susceptibility of cortical populations to entrainment by exogenous signals. Our analysis shows that the different responses to stimulation observed experimentally in different brain states can be explained by a passage through a bifurcation combined with stochastic resonance - a mechanism by which irregular fluctuations amplify the response of a nonlinear system to weak signals. Indeed, our findings suggest that modulating brain oscillations is best achieved in states of low endogenous rhythmic activity and that irregular state-dependent fluctuations in thalamic inputs control endogenous oscillatory activity. Taken together, our results show that internal control over attractor states shapes the sensitivity of neural populations to stimulation.


Short bio: Jeremie Lefebvre is a scientist at the Krembil Research Institute and an Assistant Professor in Mathematics and Biomedical Engineering at the University of Toronto. His research focuses on non-linear dynamics, stochastic dynamical systems and oscillatory neural activity, and applies to problems of brain stimulation, white matter plasticity and neurological disorders. He collaborates closely with experimentalists and clinicians worldwide.




Gijs Plomp

Fast directed interactions between brain areas and cortical layers


Visual and cognitive functions emerge from directed interactions between distributed brain areas, but how sensory and cognitive processes dynamically interact to flexibly guide behaviour remains a challenging problem. A dynamic network approach based on multivariate time-varying Granger causality can help investigate how directed functional connections support sensory and cognitive processing. Such a network approach will be introduced here and illustrated by recent results at the inter-areal scale using EEG source-imaging, and at the inter-laminar scale using intracranially recorded LFPs.




Petra Ritter

Inferring multi-scale neural mechanisms with brain network modelling

I will talk about advances of the neuroinformatics platform The Virtual Brain (thevirtualbrain.org) to integrate experimental findings with subject-specific multi-scale whole-brain network models. The Virtual Brain models enable the integration of empirical results into a biophysically based framework that allows the systematic testing of the mutual compatibility of the identified mechanisms in the context of full-brain network interaction and the prediction of system-level processes emerging from the coalescence of the individual identified mechanisms. I will demonstrate how cross-species integration may provide evidence for - possibly optimal - computational principles.



Katharina Glomb*

Graph signal processing for anatomically constrained source-reconstructed EEG data


*CHUV Lausanne, Switzerland


Combining high-density EEG, structural MRI, and source projection techniques allows imaging of cortical sources on the microsecond scale afforded by EEG. Even so, volume conduction and the low signal to noise ratio typical for EEG limit the precision with which sources and their interaction in time can be resolved. By adding results from diffusion imaging, we constrain functional connectivity with information about anatomical connections between brain areas. This way, source time courses are conceptualized as signals on a graph. I will show how filtering and optimization techniques developed for such signals can be applied to our high-density source reconstructed EEG data from visual experiments with healthy controls.
People involved in this project are: Michaël Defferard and Pierre Vandergheynst (EPFL); David Pascucci and Gijs Plomp (Uni Fribourg), Margherita Carboni, Maria Rubega and Serge Vulliemoz (University and University Hospital Geneva); as well as Sebastien Tourbier and Patric Hagmann (CHUV Lausanne).