Deformation-Invariant Sparse Coding for Modeling Spatial Variability of Functional Patterns in the Brain George Chen, Evelina Fedorenko, Nancy Kanwisher,

Slides:

Advertisements

Similar presentations

FMRI Methods Lecture 10 – Using natural stimuli. Reductionism Reducing complex things into simpler components Explaining the whole as a sum of its parts.

Advertisements

J. Daunizeau Institute of Empirical Research in Economics, Zurich, Switzerland Brain and Spine Institute, Paris, France Bayesian inference.

1 Challenge the future Multi-scale mining of fMRI data with hierarchical structured sparsity – R. Jenatton et al, SIAM Journal of Imaging Sciences, 2012.

Basis Functions. What’s a basis ? Can be used to describe any point in space. e.g. the common Euclidian basis (x, y, z) forms a basis according to which.

INTRODUCTION Assessing the size of objects rapidly and accurately clearly has survival value. Thus, a central multi-sensory module for magnitude assessment.

The General Linear Model Or, What the Hell’s Going on During Estimation?

OverviewOverview Motion correction Smoothing kernel Spatial normalisation Standard template fMRI time-series Statistical Parametric Map General Linear.

Multi-Group Functional MRI Analysis Using Statistical Activation Priors Deepti Bathula, Larry Staib, Hemant Tagare, Xenios Papademetris, Bob Schultz, Jim.

Realigning and Unwarping MfD

For stimulus s, have estimated s est Bias: Cramer-Rao bound: Mean square error: Variance: Fisher information How good is our estimate? (ML is unbiased:

1 Haskins fMRI Workshop Part III: Across Subjects Analysis - Univariate, Multivariate, Connectivity.

Amos Storkey, School of Informatics. Density Traversal Clustering and Generative Kernels a generative framework for spectral clustering Amos Storkey, Tom.

Preprocessing II: Between Subjects John Ashburner Wellcome Trust Centre for Neuroimaging, 12 Queen Square, London, UK.

Surface-based Analysis: Intersubject Registration and Smoothing

Haskins fMRI Workshop Part I: Data Acquisition & Preprocessing.

Voxel-Based Morphometry John Ashburner Wellcome Trust Centre for Neuroimaging, 12 Queen Square, London, UK.

Modeling fMRI data generated by overlapping cognitive processes with unknown onsets using Hidden Process Models Rebecca A. Hutchinson (1) Tom M. Mitchell.

Statistical Shape Models Eigenpatches model regions –Assume shape is fixed –What if it isn’t? Faces with expression changes, organs in medical images etc.

Connectivity of aMRI and fMRI data Keith Worsley Arnaud Charil Jason Lerch Francesco Tomaiuolo Department of Mathematics and Statistics, McConnell Brain.

1st Level Analysis Design Matrix, Contrasts & Inference

Lecture 24: Cross-correlation and spectral analysis MP574.

2nd Level Analysis Jennifer Marchant & Tessa Dekker

Multivariate Approaches to Analyze fMRI Data Yuanxin Hu.

Analysis of fMRI data with linear models Typical fMRI processing steps Image reconstruction Slice time correction Motion correction Temporal filtering.

Sparse Shape Representation using the Laplace-Beltrami Eigenfunctions and Its Application to Modeling Subcortical Structures Xuejiao Chen.

Keith Worsley Department of Mathematics and Statistics, and McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University Correlation.

Coregistration and Spatial Normalisation

Network modelling using resting-state fMRI: effects of age and APOE Lars T. Westlye University of Oslo CAS kickoff meeting 23/

SPM short course – Oct Linear Models and Contrasts Jean-Baptiste Poline Neurospin, I2BM, CEA Saclay, France.

Functional Connectivity in an fMRI Working Memory Task in High-functioning Autism (Koshino et al., 2005) Computational Modeling of Intelligence (Fri)

2004 All Hands Meeting Analysis of a Multi-Site fMRI Study Using Parametric Response Surface Models Seyoung Kim Padhraic Smyth Hal Stern (University of.

Reduced Local BOLD Correlation in Pre-Frontal Cortex during Deep Sleep

1/35 Structural and Probabilistic Approaches in Group Analysis of fMRI Data Brain and ICT workshop Bertrand Thirion 1, Alan Tucholka 2, Philippe Pinel.

Contrasts & Inference - EEG & MEG Himn Sabir 1. Topics 1 st level analysis 2 nd level analysis Space-Time SPMs Time-frequency analysis Conclusion 2.

FMRI ROI Analysis 7/18/2014 Friday Yingying Wang

MIT Computer Science and Artificial Intelligence Laboratory

TipiX Rapid Visualization of Large Datasets Adrian V. Dalca, Ramesh Sridharan, Natalia Rost, Polina Golland 1.

Correlation random fields, brain connectivity, and astrophysics Keith Worsley Arnaud Charil Jason Lerch Francesco Tomaiuolo Department of Mathematics and.

Computational Modeling of Anatomical and Functional Variability in Populations Polina Golland Computer Science and Artificial Intelligence Laboratory Massachusetts.

Retain H o Refute hypothesis and model MODELS Explanations or Theories OBSERVATIONS Pattern in Space or Time HYPOTHESIS Predictions based on model NULL.

A Unified Feature Registration Framework for Brain Anatomical Alignment Haili Chui, Robert Schultz, Lawrence Win, James Duncan and Anand Rangarajan* Image.

Spatial Smoothing and Multiple Comparisons Correction for Dummies Alexa Morcom, Matthew Brett Acknowledgements.

SPM Pre-Processing Oli Gearing + Jack Kelly Methods for Dummies

Dynamic Data Analysis Projects in the Image Analysis and Motion Capture Labs Figure: functional brain MRI of a monetary reward task; left: 16 cocaine subjects,

Group-wise Registration in NAMIC-kit Serdar K Balci (MIT) Lilla Zöllei (MGH) Kinh Tieu (BWH) Mert R Sabuncu (MIT) Polina Golland (MIT)

NA-MIC National Alliance for Medical Image Computing fMRI in NAMIC Facilitator: Polina Golland Presenters: Jim Fallon and Andy Saykin.

Methods for Dummies Second level Analysis (for fMRI) Chris Hardy, Alex Fellows Expert: Guillaume Flandin.

Statistical Analysis An Introduction to MRI Physics and Analysis Michael Jay Schillaci, PhD Monday, April 7 th, 2007.

Group Analysis ‘Ōiwi Parker Jones SPM Course, London May 2015.

Hierarchical statistical analysis of fMRI data across runs/sessions/subjects/studies using BRAINSTAT / FMRISTAT Jonathan Taylor, Stanford Keith Worsley,

NA-MIC National Alliance for Medical Image Computing Statistical Models of Anatomy and Pathology Polina Golland.

Fast Human Detection in Crowded Scenes by Contour Integration and Local Shape Estimation Csaba Beleznai, Horst Bischof Computer Vision and Pattern Recognition,

NA-MIC National Alliance for Medical Image Computing Modeling Populations and Pathology Kayhan N. Batmanghelich PI: Polina Golland MIT.

SPM short course – Mai 2008 Linear Models and Contrasts Jean-Baptiste Poline Neurospin, I2BM, CEA Saclay, France.

A 2 veto for Continuous Wave Searches

Group Averaging of fMRI Data

Polina Golland Core 1, MIT

Surface-based Analysis: Inter-subject Registration and Smoothing

Registration of Pathological Images

Surface-based Analysis: Intersubject Registration and Smoothing

Surface-based Analysis: Intersubject Registration and Smoothing

Keith Worsley Keith Worsley

Computational Neuroanatomy for Dummies

Sam Norman-Haignere, Nancy G. Kanwisher, Josh H. McDermott Neuron

Bayesian inference J. Daunizeau

Anatomical Measures John Ashburner

Volume 47, Issue 5, Pages (September 2005)

Group DCM analysis for cognitive & clinical studies

Evelina Fedorenko, John Duncan, Nancy Kanwisher Current Biology

Presentation transcript:

Deformation-Invariant Sparse Coding for Modeling Spatial Variability of Functional Patterns in the Brain George Chen, Evelina Fedorenko, Nancy Kanwisher, Polina Golland 12/16/2011NIPS MLINI Workshop 20111

Talk Outline 1.Finding correspondences between functional regions in the brain 2.A new generative model 3.Results for language fMRI study 12/16/2011NIPS MLINI Workshop 20112

Functional Region Correspondences 12/16/2011NIPS MLINI Workshop Given stimulus, get functional activation regions Subject 1 Subject 2 Align to common anatomical space Functional variability! Goal: Find correspondences between “parcels” contiguous region in brain group-level parcels Parcel : contiguous region in brain Biology : brain compartmentalized into functional modules  parcels represent these modules

Functional Variability Standard approach: just average in common anatomical space 12/16/2011NIPS MLINI Workshop Functional variability  less pronounced activation in group average space Subject 1 Subject 2 space Average space Aligned space

Previous Work Thirion et al. 2007: treat parcels as discrete objects and find parcel correspondences across subjects by matching Xu et al. 2009: generative, hierarchical model representing activation regions as Gaussian mixtures Sabuncu et al. 2010: groupwise functional registration 12/16/2011NIPS MLINI Workshop 20115

Previous Work Thirion et al. 2007: treat parcels as discrete objects and find parcel correspondences across subjects by matching Xu et al. 2009: generative, hierarchical model representing activation regions as Gaussian mixtures Sabuncu et al. 2010: groupwise functional registration 12/16/2011NIPS MLINI Workshop 20116

Our Generative Model 12/16/2011NIPS MLINI Workshop To generate image for a subject: 1.Choose weights for each group-level parcel 2.Form weighted sum of group-level parcels 3.Deform pre-image and add noise Pre-image e.g. (0.2, 1) Group-level parcels 1: 2: Goal: Estimate group-level parcels and deformations

Estimating Group-level Parcels and Deformations 12/16/2011NIPS MLINI Workshop sparsitysmoothnessparcelidentifiability

Language fMRI Study 12/16/2011NIPS MLINI Workshop 20119

Left frontal lobe Left temporal lobe Estimated Group-level Parcels Correspond to known language processing regions 12/16/2011NIPS MLINI Workshop Spatial support of group-level parcels Right temporal lobe Right cerebellum Example group-level parcels

Apply estimated deformation to fMRI data for each subject and redo standard fMRI group analysis on separate data 12/16/2011NIPS MLINI Workshop Modeling functional variability increases statistical significance in each group-level parcel Group-level Parcel Index Negative log p-value Improving fMRI Group Analysis with Estimated Deformations

12/16/2011NIPS MLINI Workshop space Subject 1 Subject 2 Average Aligned space Why is the variance so high for statistical significance values for our model?

Improving fMRI Group Analysis with Estimated Deformations 12/16/2011NIPS MLINI Workshop Average space Why is the variance so high for statistical significance values for our model? Group-level parcel support Variation using anatomical alignment only Variation using our model

Apply estimated deformation to fMRI data for each subject and redo standard fMRI group analysis 12/16/2011NIPS MLINI Workshop Modeling functional variability increases statistical significance in each group-level parcel Group-level Parcel Index Negative log p-value Improving fMRI Group Analysis with Estimated Deformations

Contributions 12/16/2011NIPS MLINI Workshop