Disentangling mono-synaptic connectivity from cortico-cortical evoked potentials
Poster No:
1191
Submission Type:
Abstract Submission
Authors:
Chun Hei Michael Chan1, Alexandre Cionca2, Maciej Jedynak3, Yasser Alemán-Gómez4, Arthur Spencer5, Saina Asadi6, Ileana Jelescu7, Olivier David3, Serge Vulliémoz8, Patric Hagmann9, Dimitri Van De Ville10
Institutions:
1École polytechnique fédérale de Lausanne (EPFL), Ecublens, Vaud, 2EPFL, Lausanne, Vaud, 3Univ. Aix Marseille, Marseille, Bouches-du-Rhône, 4Centre hospitalier universitaire vaudois (CHUV), Lausanne, VT, 5Lausanne University Hospital (CHUV), Lausanne, Switzerland, 6University of Lausanne, Lausanne University Hospital (CHUV), Lausanne, Vaud, 7Department of Radiology, Lausanne University Hospital (CHUV) and University of Lausanne (UNIL), Lausanne, Vaud, 8University of Geneva, Geneva, Geneva, 9Lausanne University Hospital and University of Lausanne (CHUV-UNIL), Lausanne, Vaud, 10École polytechnique fédérale de Lausanne (EPFL), Geneva, Geneva
First Author:
Co-Author(s):
Ileana Jelescu
Department of Radiology, Lausanne University Hospital (CHUV) and University of Lausanne (UNIL)
Lausanne, Vaud
Department of Radiology, Lausanne University Hospital (CHUV) and University of Lausanne (UNIL)
Lausanne, Vaud
Introduction:
Brain network neuroimaging is an emerging field of neuroscience that considers one's brain as a set of interconnected regions; i.e. a graph. Current approaches often consider symmetric relationships between regions (Hagmann, 2007) but not directed ones. Additionally, recent techniques such as stereo-encephalography can now provide directional information of brain activity through electrically induced cortico-cortical evoked potentials (Lemaréchal, 2022). However such information isn't solely representative of the immediate influence from a region to another i.e the mono-synaptic connectivity. Indeed it also includes all other paths connecting two regions, i.e poly-synaptic connectivity. Leveraging this information, we present a regression model that aims at extracting mono-synaptic communication between brain regions by disentangling the contribution from poly-synaptic ones to obtain a directed mono-synaptic brain connectome.
Methods:
We build a ridge regression model on the assumption that conduction delay measured between two regions is a combination of mono and poly-synaptic communications. This translates to a model where the measured conduction delay m(AB) between node A and B is a weighted sum of: one graph hop interaction between A and B defined as the effective delay d(AB), and the sum of effective delays sampled along all multi-hops paths (e.g d(AC) and d(CB)). Two parameters are introduced in the model, α that controls the contribution of poly-synaptic connections and δsyn that represents the added synaptic delay at each graph node. Formally, the relation between m(AB) and d(AB) is expressed by following equation with notations in (Fig.1a)
$$m(AB)=d(AB)+\delta_{syn}+\alpha\left[d(AC)+d(CB)+2\delta_{syn}\right]$$
Conduction delays ranging up to 400 ms, are taken from the F-TRACT project (https://f-tract.eu, Lemaréchal, 2022). We integrate structural information from population-level white matter bundle atlas (Alemán-Gómez, 2022), by computing effective delay only between regions connected by bundle tracts. By design of the regression model, existing conduction delays between regions that do not present a white matter connection will be distributed to estimate the effective delays on multi-hops paths between these regions. Furthermore, the effective delay between two regions with no measurement of conduction delay can be estimated when it is part of at least one multi-hop path between two other regions. The regression model is solved through a simple gradient descent while a grid search was performed on α and δsyn (Fig.1d).
$$m(AB)=d(AB)+\delta_{syn}+\alpha\left[d(AC)+d(CB)+2\delta_{syn}\right]$$
Conduction delays ranging up to 400 ms, are taken from the F-TRACT project (https://f-tract.eu, Lemaréchal, 2022). We integrate structural information from population-level white matter bundle atlas (Alemán-Gómez, 2022), by computing effective delay only between regions connected by bundle tracts. By design of the regression model, existing conduction delays between regions that do not present a white matter connection will be distributed to estimate the effective delays on multi-hops paths between these regions. Furthermore, the effective delay between two regions with no measurement of conduction delay can be estimated when it is part of at least one multi-hop path between two other regions. The regression model is solved through a simple gradient descent while a grid search was performed on α and δsyn (Fig.1d).
Results:
Results of our regression show a realistic range of values for effective delays, mainly ranging from 5 to 30 ms (Fig.1b), as found in literature (Silverstein, 2020).
Furthermore, our model yields preliminary values for both α and δsyn. At maximum correlation strength between regressed delay and fiber length (Fig.1c, Fig.1d) we obtain α=1, i.e contributions of mono-synaptic and poly-synaptic delay are equally weighted and we obtain synaptic delays of δsyn=30 ms.
Finally, we compute a first estimate of the general axonal bundle conduction speed, yielding 1.6 m.s-1 (Fig.1c) in the range of dendritic conduction velocities (Stuart, 1998) however slow for myelinated axons (Mancini, 2021).
Furthermore, our model yields preliminary values for both α and δsyn. At maximum correlation strength between regressed delay and fiber length (Fig.1c, Fig.1d) we obtain α=1, i.e contributions of mono-synaptic and poly-synaptic delay are equally weighted and we obtain synaptic delays of δsyn=30 ms.
Finally, we compute a first estimate of the general axonal bundle conduction speed, yielding 1.6 m.s-1 (Fig.1c) in the range of dendritic conduction velocities (Stuart, 1998) however slow for myelinated axons (Mancini, 2021).
Conclusions:
This work provides a first dive into a population-level asymmetric connectome of the human brain extracting monosynaptic connections. Our preliminary results show estimated effective delays of realistic ranges and integrate both conduction speed and synaptic constant. We believe this sets the ground for highlighting the direction of brain information flow and to consider brain connectivity with a sense of causality.
Further work includes validating our model by relating our effective delay with existing microstructure maps in addition to investigating the obtained directionality edges and network configurations using known causal patterns.
Further work includes validating our model by relating our effective delay with existing microstructure maps in addition to investigating the obtained directionality edges and network configurations using known causal patterns.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 1
Exploratory Modeling and Artifact Removal
Methods Development
Other Methods
Neuroinformatics and Data Sharing:
Brain Atlases 2
Keywords:
Atlasing
Computational Neuroscience
Modeling
Tractography
White Matter
Other - Brain Connectome ; Stereoelectroencephalography
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
Provide references using APA citation style.
Hagmann, P., Kurant, M., Gigandet, X., Thiran, P., Wedeen, V. J., Meuli, R., & Thiran, J. P. (2007). Mapping human whole-brain structural networks with diffusion MRI. PloS one, 2(7), e597.
Lemaréchal, J. D., Jedynak, M., Trebaul, L., Boyer, A., Tadel, F., Bhattacharjee, M., ... & David, O. (2022). A brain atlas of axonal and synaptic delays based on modelling of cortico-cortical evoked potentials. Brain, 145(5), 1653-1667.
Mancini, M., Tian, Q., Fan, Q., Cercignani, M., & Huang, S. Y. (2021). Dissecting whole-brain conduction delays through MRI microstructural measures. Brain Structure and Function, 226(8), 2651-2663.
Silverstein, B. H., Asano, E., Sugiura, A., Sonoda, M., Lee, M. H., & Jeong, J. W. (2020). Dynamic tractography: Integrating cortico-cortical evoked potentials and diffusion imaging. Neuroimage, 215, 116763
Stuart, G., & Spruston, N. (1998). Determinants of voltage attenuation in neocortical pyramidal neuron dendrites. Journal of neuroscience, 18(10), 3501-3510.