In the MDCA project, we provide different methods that we develop to characterize and analyize brain signals. In different project, we proposed differnet methods of brain signal analysis. We applied our methodologies on real data and parts of results are provided here. In the below, a short description foreach project is presented.

In State Space Coherence (SS-Coh) project, we provide mor robust algorithm for estimating Global Coherence (GCoh). To understand some aspects of SS-Coh modelling, you can read EMBC-2019 and bioRxiv-2020. In the RO1 proposal, we provide more complete modelling of SS-Coh. The complete description of the experimental protocol can be found in Purdon et al.

In our previous work, we demonstrated a latent dynamical modeling framework called state-space global coherence, which characterizes spectral measures to capture slow-changing dynamics in network-level coherence. In this research, we develop a more general class of the state-space coherence model, that can capture fast and switching changes in the network-level rhythmic dynamics. For this framework, we assume both continuous and discrete latent processes derive the network-level rhythmic dynamics; this modeling assumption, will help us to build a more flexible model structure that can capture sophisticated dynamics present in the neural data. The complete description of the experimental protocol can be found in Purdon et al.

In this project, we will evaluate Switching mechanism in the brain by applying Global Coherence Algorithm. We evaluate the cluster analysis in one participant. A small number of functional circuits appear at different segments of the experiment; moreover, the same functional circuit emerges when a specific cognitive task is repeated.

Large scale recordings of brain-wide activity in awake mice during spontaneous behavior are publicly available to allow studies on how multidimensional behaviors are represented in distributed neural data. Using the mice dataset, we are interested in studying how coherence analysis will reveal interactions across neural nodes and their connections to behavior. To study the data, we will use empirical and parametric point-process coherence; we will also utilize mixed-data coherence analysis to study neural and behavioral data in tandem.

Data is publicly available. Please check the Stringer et. al (2019).

In this project, we use empirical point-process coherence and global coherence analysis to study the human Amygdala dataset. Data is publicly available. Please check the OpenNeuro Repository.