Optimized design and analysis of Optimized design and analysis of - PowerPoint PPT Presentation

Optimized design and analysis of Optimized design and analysis of sparse-sampling fMRI experiments sparse-sampling fMRI experiments of speech and hearing of speech and hearing Tyler Perrachione John Gabrieli Satrajit Ghosh

Overview Overview • What is sparse-sampling in fMRI and why is it useful? • What can we do to optimize sparse-sampling design and analysis? – Accounting for hemodynamics in sparse data analysis • Computational modeling of sparse designs • Human neuroimaging data Tyler Perrachione -- tkp@mit.edu 2 May, 2011

Sparse-sampling: What and why? Sparse-sampling: What and why? • Acoustic noise from • Head motion during vocal switching MR scanner responses adds noise, gradients impedes registration – Obfuscates auditory • Solution : Present stimuli stimuli during silent delays – Saturates activation of between functional volume auditory cortices acquisition Tyler Perrachione -- tkp@mit.edu 3 May, 2011

What studies benefit from sparse? What studies benefit from sparse? • Hearing & Language – Children – Aging adults – Auditory impairment auditory neuropathy, central auditory processing disorders – Dyslexia • Speech Production Schmidt et al., 2008 • Neurosurgical mapping of language regions • Studies on older scanner hardware Tyler Perrachione -- tkp@mit.edu 4 May, 2011

How can we optimize sparse fMRI? How can we optimize sparse fMRI? • “Boxcar” models typically used to represent sparse timeseries – Ignores dynamic properties of hemodynamic response – Cannot distinguish the contribution of multiple event types to individual TAs • Optimal delay between TRs? – TR = 4? TR = 14? – Psychological and physiological factors • Detection of activation in non- auditory regions? Tyler Perrachione -- tkp@mit.edu 5 May, 2011

Simulating optimized sparse designs Simulating optimized sparse designs • What delay is best? ⊗ – TR = 4, 8, 12s • What model is best? – Boxcar vs. HRF • Generate a high temporal resolution timeseries • Convolve with canonical HRF • Resample at points corresponding to actual TA • Add noise – 5dB SNR – From actual resting fMRI • Compare Models: – 10,000 simulations in R Tyler Perrachione -- tkp@mit.edu 6 May, 2011

Simulating optimized sparse designs Simulating optimized sparse designs Result of 10,000 simulations Tyler Perrachione -- tkp@mit.edu 7 May, 2011

Human neuroimaging data Human neuroimaging data • Modified “Stroop” task • Acquisition Parameters: – Press button when color of – TR = {4, 8, 12} sec – TA = 2 sec text matches spoken color – 3.125 mm 3 voxels name. – 5min {92, 47, 32 volumes} • 14 control participants • Analysis – Right-handed, native- – Functional data in SPM8 English, normal hearing and NiPype • Motion correction, volume “green” smoothing, within-subject model estimation – Coregistration using “blue” red Freesurfer 5.0.0 “red”  blue – Symmetric diffeomorphic nonlinear normalization in green ANTS v1.5 Tyler Perrachione -- tkp@mit.edu 8 May, 2011

Human neuroimaging data Human neuroimaging data Tyler Perrachione -- tkp@mit.edu 9 May, 2011

Human neuroimaging data Human neuroimaging data L p < 0.001, FDR q = 0.05 Tyler Perrachione -- tkp@mit.edu 10 May, 2011

Human neuroimaging data Human neuroimaging data Classic “Boxcar” Model Convolved HRF Model TR = 4 TR = 8 TR = 12 L p < 0.001, FDR q = 0.05 Tyler Perrachione -- tkp@mit.edu 11 May, 2011

Human neuroimaging data Human neuroimaging data Simulations: Model Comparison HRF > Boxcar TR = 4s L p < 0.001, FDR q = 0.05 Tyler Perrachione -- tkp@mit.edu 12 May, 2011

Human neuroimaging data Human neuroimaging data Difference between task conditions Go > No-Go TR = 4 TR = 8 TR = 12 p < 0.01, FDR q = 0.05 p < 0.01, FDR q = 0.05 p < 0.05, FDR q = 0.05 Tyler Perrachione -- tkp@mit.edu 13 May, 2011

Conclusions Conclusions • Optimizing sparse fMRI designs: – HRF convolution significantly improves power of short (4s) TR sparse experiments – Short TRs improve detection of non-auditory activity – Very long TRs (12s) obscure activity in both auditory and non- auditory regions – HRF convolution can distinguish multiple conditions, allowing event-related sparse designs • Increased power from optimized sparse designs: – Reduces amount of scantime needed – Reduces length participants are in scanner – Reduces cost of fMRI experiments – Increases reliability of findings Tyler Perrachione -- tkp@mit.edu 14 May, 2011

Acknowledgments Acknowledgments Satrajit S. Ghosh We thanks: Carlos Cardenas-Iniguez, Stephanie Del Tufo, John Lymberis, Steve Shannon, and Christina Triantafyllou. John D.E. Gabrieli Imaging was performed at the Athinoula A. Martinos Imaging Center at McGovern Institute for Brain tkp@mit.edu Research, MIT. This research was supported by the Ellison Medical http://web.mit.edu/tkp/www/ Foundation, as well as NIH grant R03-EB008673 to S.G., and a NSF GRF to T.P. Tyler Perrachione -- tkp@mit.edu 15 May, 2011