Skip to content

Assessing consciousness by perturbing a dynamic mean-field whole-brain model fitted to empirical neuroimaging data

License

Notifications You must be signed in to change notification settings

tomasberjaga/Master-Thesis

 
 

Repository files navigation

Master-Thesis

Title: Assessing consciousness by perturbing a dynamic mean-field whole-brain model fitted to empirical neuroimaging data

Author: Tomas Berjaga Buisan

Supervisors: Dr. Yonatan Sanz Pearl and Prof. Dr. Gustavo Deco

Affiliation: Computational Neuroscience Research Group, Center for Brain and Cognition, Universitat Pompeu Fabra.

Abstract: Consciousness assessment in the clinics still relies mainly on the patient's ability to interact with the environment and motor functioning. However, as an absence of behavioral signs cannot be considered proof of the absence of consciousness, clinical consciousness evaluation represents a fundamental shortcoming. The search for reliable and objective methods that do not depend on the lack of responsiveness has been increasing. Perturbational Complexity Index (PCI) is the most promising method developed to date, but, despite its promising and accurate results, it is associated with many technical and clinical problems. In fact, it needs transcranial magnetic stimulation (TMS) for its calculation. Hence, there are notable ethical concerns as it interferes with brain excitability and could even lead to seizure induction or syncope. This thesis aims to develop an in-silico framework to calculate PCI without interfering with brain excitability. Furthermore, the thesis also aims to gain insight into the intrinsic mechanism of the metric. For that, a whole-brain dynamic mean-field model has been fitted to empirical neuroimaging data from different consciousness states. The fitted whole-brain model is perturbed to calculate the PCI and assess the metric's intrinsic mechanisms. Results demonstrate that PCI can be calculated with an in-silico framework, thus erasing ethical concerns and possible side effects of TMS. Moreover, the thesis has demonstrated an optimal amplitude window for differentiating pairs of states and a mechanistic dependency of PCI value on information integration and network segregation. All in all, this thesis represents a further step into the formidable challenge of validating and calibrating an index of consciousness for its transfer into clinical practice.

In this folder, you can find all the codes used to reproduce Tomas Berjaga Master's Thesis results. Empirical data is available upon request due to ethical concerns.

About

Assessing consciousness by perturbing a dynamic mean-field whole-brain model fitted to empirical neuroimaging data

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 83.4%
  • MATLAB 7.8%
  • Shell 7.7%
  • Python 1.1%