New InBrain publication at Human Brain Mapping
We are very please to announce that the publication titled "A deep learning based approach identifies regions more relevant than resting-state networks to the prediction of general intelligence from resting-state fMRI" has been accepted for publication in Human Brain Mapping. This paper stems from Bruno H. Vieira's PhD Thesis, and was written in collaboration with professors Julien Dubois, Vince D. Calhoun and Carlos E. G. Salmon.
In this paper, authors propose a new methodology to obtain predictions of intelligence differences from resting-state fMRI. Instead of extracting features such as functional connectivity, raw timeseries are fed to an ensemble of deep recurrent neural networks. This methodology results in increased accuracy compared to the best method of the literature and opens up the possibility of extracting new insights about brain-cognition relationships.
The paper is Open Access and can be accessed online ahead of print here.
Hebling Vieira, B., Dubois, J., Calhoun, V. D., & Garrido Salmon, C. E. (2021). A deep learning based approach identifies regions more relevant than resting-state networks to the prediction of general intelligence from resting-state fMRI. Human Brain Mapping, 1– 15. https://doi.org/10.1002/hbm.25656
