1 / 17

Bayesian selection of dynamic causal models for fMRI

SPC. V1. V5. SPC. V1. V5. Bayesian selection of dynamic causal models for fMRI . Will Penny Olivier David, Karl Friston, Lee Harrison, Andrea Mechelli, Klaas Stephan. Wellcome Department of Imaging Neuroscience, ION, UCL, UK.

dbjorklund
Download Presentation

Bayesian selection of dynamic causal models for fMRI

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. SPC V1 V5 SPC V1 V5 Bayesian selection of dynamic causal models for fMRI Will Penny Olivier David, Karl Friston, Lee Harrison, Andrea Mechelli, Klaas Stephan Wellcome Department of Imaging Neuroscience, ION, UCL, UK. The brain as a dynamical system ? Bridging the gap between models and data, HBM Workshop, Budapest, Hungary, June 16 2004.

  2. Single region u1 c u1 a11 z1 u2 z1 z2

  3. Multiple regions u1 c u1 a11 z1 u2 z1 a21 z2 z2 a22

  4. Modulatory inputs u1 u2 c u1 a11 z1 u2 b21 z1 a21 z2 z2 a22

  5. u1 u2 c u1 a11 z1 u2 b21 a12 z1 a21 z2 z2 a22 Reciprocal connections

  6. Neuronal Activity Inputs Change in Neuronal Activity Input Connectivity Matrix Intrinsic Connectivity Matrix Modulatory Connectivity Matrices SPC V1 V5 DCM for fMRI Neurodynamics:

  7. Hemodynamics For each region: Hemodynamic variables Dynamics Hemodynamic parameters Seconds

  8. SPC V1 V5 Model Comparison I Model, m Parameters: Prior Posterior Likelihood Evidence Laplace, AIC, BIC approximations Model fit + complexity

  9. Prior Posterior Likelihood SPC V1 V5 Model Comparison II Parameters: Model, m Parameter Parameter Model Model Evidence Prior Posterior

  10. SPC V1 V5 SPC V1 V5 Model Comparison III Model, m=i Model Evidences: Bayes factor: Model, m=j 1 to 3: Weak 3 to 20: Positive 20 to 100: Strong >100: Very Strong

  11. Attention to Visual Motion STIMULI 250 radially moving dots at 4.7 degrees/s PRE-SCANNING 5 x 30s trials with 5 speed changes (reducing to 1%) Task - detect change in radial velocity SCANNING(no speed changes) 6 normal subjects, 4 100 scan sessions; each session comprising 10 scans of 4 different condition Buchel et al. 1997 Experimental Factors • Photic • Motion • Attention

  12. Model 1 Photic SPC V1 V5 Motion Att Specify regions of interest GLM analysis Identify regions of Interest eg. V1, V5, SPC

  13. Model 1 Photic SPC V1 V5 Motion Att V1 V5 Estimation SPC Time (seconds)

  14. Model 1 Photic SPC Very Strong V1 V5 Motion Att SPC V1 V5 B12 > 1019 Bayes Factor Model 2 Photic Motion Att

  15. Model 1 Model 3 Photic SPC Photic SPC Positive V1 V1 Att V5 V5 Motion Motion Att Bayes Factor B13=3.6

  16. Model 1 Photic SPC SPC Weak V1 V1 V5 V5 Motion Att B14=2.8 Bayes Factor Model 4 Photic Att Motion Att

  17. LD|LVF FG left FG right LD LD LG left LG right LVF RVF Further Applications 1. Klaas Stephan et al. HBM 04 – Poster TH154, Thurs 1pm LD: Letter decision LVF: left visual field Fusiform gyrus FG: LG: Lingual gyrus Dominant right->left modulation during letter tasks LG and FG important for hemispheric integration (not MOG) 2. Olivier David et al. BIOMAG 04 – DCM for ERPs

More Related