Directed fMRI-based Functional Connectivity Estimation using Physics-Informed Neural Networks | bioRxiv

biorxiv.org · 624 words · saved by 1 readers

Estimating directed functional connectivity (dFC) within the brain is crucial for comprehending neural interactions. However, conventional methodologies encounter constraints in accuracy, scalability, and interpretation. The method presented here harnesses Physics-Informed Neural Networks (PINNs) to amalgamate the governing physical principles of brain dynamics, thereby improving dFC estimation from resting-state functional magnetic resonance imaging (rsfMRI) data. In particular, during the training phase, we derive the input weights from a long-short term memory (LSTM) network, which, within our framework, represent the influence of all other brain areas on the specific region under consideration. These input weights are then integrated into the nonlinear differential equation that models the rsfMRI time series within the specific brain area. Through the training of the PINN model, we simultaneously estimate, for each brain area, the biophysical parameters of the model, including the

Directed fMRI-based Functional Connectivity Estimation using Physics-Informed Neural Networks | bioRxiv Skip to main content New Results Directed fMRI-based Functional Connectivity Estimation using Physics-Informed Neural Networks Roberto C. Sotero , Jose M. Sanchez-Bornot doi: https://doi.org/10.1101/2024.07.09.602748 Roberto C. Sotero 1 Department of Radiology and Hotchkiss Brain Institute, University of Calgary , Calgary, AB, Canada Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: roberto.soterodiaz{at}ucalgary.ca Jose M.

Directed fMRI-based Functional Connectivity Estimation using Physics-Informed Neural Networks | bioRxiv

related reading