SungJun Cho

This paper introduces bi-cross-validation to evaluate and compare dynamic functional connectivity models in fMRI, showing it balances model fit and complexity effectively.

This paper introduces DyNeStE, a model that combines recurrent neural networks with Bayesian inference to capture long-term dynamics in brain networks while maintaining interpretability, outperforming HMMs in simulations and resting-state MEG data.

This study compares static and dynamic resting-state networks identified by medium-density EEG and high-density MEG, finding comparable network descriptions with MEG showing higher sensitivity and reproducibility.