Joint generative model for fMRI/DWI and its application to population studies

Authors:
Archana Venkataraman;Yogesh Rathi;Marek Kubicki;Carl-Fredrik Westin;Polina Golland
Affiliations:
MIT Computer Science and Artificial Intelligence Laboratory, Cambridge, MA;Psychiatry Neuroimaging Laboratory, Harvard Medical School, Boston, MA;Psychiatry Neuroimaging Laboratory, Harvard Medical School, Boston, MA;Laboratory for Mathematics Imaging, Harvard Medical School, Boston, MA and MIT Computer Science and Artificial Intelligence Laboratory, Cambridge, MA;MIT Computer Science and Artificial Intelligence Laboratory, Cambridge, MA
Venue:
MICCAI'10 Proceedings of the 13th international conference on Medical image computing and computer-assisted intervention: Part I
Year:
2010

Citing 3
Cited 2

An Introduction to Variational Methods for Graphical Models

Machine Learning
Neural Tractography Using an Unscented Kalman Filter

IPMI '09 Proceedings of the 21st International Conference on Information Processing in Medical Imaging
Combined analysis of resting-state fMRI and DTI data reveals abnormal development of function-structure in early-onset schizophrenia

RSKT'08 Proceedings of the 3rd international conference on Rough sets and knowledge technology

A probabilistic framework to infer brain functional connectivity from anatomical connections

IPMI'11 Proceedings of the 22nd international conference on Information processing in medical imaging
Probabilistic graphical model of SPECT/MRI

MLMI'11 Proceedings of the Second international conference on Machine learning in medical imaging

Quantified Score

Hi-index	0.01

Visualization

Abstract

We propose a novel probabilistic framework to merge information from DWI tractography and resting-state fMRI correlations. In particular, we model the interaction of latent anatomical and functional connectivity templates between brain regions and present an intuitive extension to population studies. We employ a mean-field approximation to fit the new model to the data. The resulting algorithm identifies differences in latent connectivity between the groups. We demonstrate our method on a study of normal controls and schizophrenia patients.