DARTEL

John Ashburner
2008
 Overview
• Motivation
– Dimensionality
– Inverse-consistency
• Principles
• Geeky stuff
• Example
• Validation
• Future directions
 Motivation
• More precise inter-subject alignment
– Improved fMRI data analysis
• Better group analysis
• More accurate localization
– Improve computational anatomy
• More easily interpreted VBM
• Better parameterization of brain shapes
– Other applications
• Tissue segmentation
• Structure labeling
 Image Registration
• Figure out how to
warp one image to
match another

• Normally, all
subjects’ scans are
matched with a
common template
 Current SPM approach
• Only about 1000 parameters.
– Unable model detailed
deformations
 A one-to-one mapping
• Many models simply add a smooth displacement to an
identity transform
– One-to-one mapping not enforced
• Inverses approximately obtained by subtracting the
displacement
– Not a real inverse Small deformation
approximation
 Overview
• Motivation
• Principles
• Optimisation
• Group-wise Registration
• Validation
• Future directions
 Principles
Deformations Diffeomorphic
parameterized by a
single flow field, which Anatomical
is considered to be
constant in time.
Registration
Through
Exponentiated
Lie Algebra
 DARTEL
• Parameterising the deformation
• φ(0)(x) = x
1

• φ(1)(x) = ∫ u(φ(t)(x))dt
t=0
• u is a flow field to be estimated
 Euler integration
• The differential equation is
dφ(x)/dt = u(φ(t)(x))
• By Euler integration
φ(t+h) = φ(t) + hu(φ(t))
• Equivalent to
φ(t+h) = (x + hu) o φ(t)
Flow Field
 For (e.g) 8 time steps
Simple integration Scaling and squaring
• φ(1/8) = x + u/8 • φ(1/8) = x + u/8
• φ(2/8) = φ(1/8) o φ(1/8) • φ(2/8) = φ(1/8) o φ(1/8)
• φ(3/8) = φ(1/8) o φ(2/8) • φ(4/8) = φ(2/8) o φ(2/8)
• φ(4/8) = φ(1/8) o φ(3/8) • φ(8/8) = φ(4/8) o φ(4/8)
• φ(5/8) = φ(1/8) o φ(4/8) 3 compositions

• φ(6/8) = φ(1/8) o φ(5/8)

• Similar procedure
• φ(7/8) = φ(1/8) o φ(6/8) used for the inverse.
• φ(8/8) = φ(1/8) o φ(7/8) Starts with
7 compositions
φ(-1/8) = x - u/8
Scaling and squaring example
DARTEL
Jacobian determinants remain
positive
 Overview
• Motivation
• Principles
• Optimisation
– Multi-grid
• Group-wise Registration
• Validation
• Future directions
 Registration objective function
• Simultaneously minimize the sum of
– Likelihood component
• From the sum of squares difference
• ½∑i(g(xi) – f(φ(1)(xi)))2
• φ(1) parameterized by u
– Prior component
• A measure of deformation roughness
• ½uTHu
 Regularization model
• DARTEL has three different models for H
– Membrane energy
– Linear elasticity
– Bending energy
• H is very sparse

An example H for 2D
registration of 6x6
images (linear
elasticity)
Regularization models
 Optimisation
• Uses Levenberg-Marquardt
– Requires a matrix solution to a very large set
of equations at each iteration
u(k+1) = u(k) - (H+A)-1 b

– b are the first derivatives of objective function

– A is a sparse matrix of second derivatives
– Computed efficiently, making use of scaling
and squaring
 Relaxation
• To solve Mx = c
Split M into E and F, where
• E is easy to invert
• F is more difficult

• Sometimes: x(k+1) = E-1(c – F x(k))

• Otherwise: x(k+1) = x(k) + (E+sI)-1(c – M x(k))

• Gauss-Siedel when done in place.

• Jacobi’s method if not

• Fits high frequencies quickly, but low frequencies slowly

H+A = E+F
 Highest resolution

Full Multi-Grid

Lowest
resolution
 Overview
• Motivation
• Principles
• Optimisation
• Group-wise Registration
– Simultaneous registration of GM & WM
– Tissue probability map creation
• Validation
• Future directions
 Generative Models for Images
• Treat the template as a deformable
probability density.
– Consider the intensity distribution at each
voxel of lots of aligned images.
• Each point in the template represents a probability
distribution of intensities.
– Spatially deform this intensity distribution to
the individual brain images.
• Likelihood of the deformations given by the
template (assuming spatial independence of
voxels).
 Generative models of anatomy
• Work with tissue class images.

Automatically
segmented
grey matter
images.

• Brains of differing shapes and sizes.

• Need strategies to encode such variability.
Simultaneous registration of GM to
GM and WM to WM
Grey matter
Subject
1 White matter Grey matter Subject
3
White matter

Grey matter

White matter
Grey matter
Template Grey matter
White matter
White matter
Subject 2
Subject 4
 Template Creation
• Template is an average shaped brain.
– Less bias in subsequent analysis.
• Iteratively created mean using DARTEL
algorithm.
– Generative model of data.
– Multinomial noise model. Grey matter
average of
471 subjects
ϕ1
t1
ϕ2
μ
t2
t5
White matter
ϕ3
t3 average of
ϕ5 t4 ϕ4 471 subjects
 Average Shaped Template
• For CA, work in the tangent space of the
manifold, using linear approximations.
– Average-shaped templates give less bias, as the
tangent-space at this point is a closer approximation.
• For spatial normalisation of fMRI, warping to a
more average shaped template is less likely to
cause signal to disappear.
– If a structure is very small in the template, then it will
be very small in the spatially normalised individuals.
• Smaller deformations are needed to match with
an average-shaped template.
– Smaller errors.
 Average shaped templates
Average on
Riemannian
manifold

Linear Average
(Not on Riemannian manifold)
 Template
Initial
Average

Iteratively
generated from
471 subjects
After a few
Began with rigidly iterations
aligned tissue
probability maps

Used an inverse
consistent
Final
formulation
template
Grey matter
average of 452
subjects –
affine
Grey matter
average of 471
subjects
 Multinomial Model
• Current DARTEL model is multinomial for
matching tissue class images.
log p(t|μ,ϕ) = ΣjΣk tjk log(μk(ϕj))
t – individual GM, WM and
background
μ – template GM, WM and background
ϕ – deformation
• A general purpose template should not
have regions where log(μ) is –Inf.
 Laplacian Smoothness Priors on
template
2D
Nicely
scale
invariant

3D
Not quite
scale
invariant –
but probably
close enough
 Template modelled as softmax of a Gaussian process
μk(x) = exp(ak(x))/(Σj exp(aj(x)))
Rather than compute mean images and convolve with a Gaussian, the
smoothing is done by maximising a log-likelihood for a MAP solution.

Note that Jacobian transformations are required (cf “modulated

VBM”) to properly account for expansion/contraction during warping.

Smoothing by solving
matrix equations
using multi-grid
 Determining amount of
regularisation
• Matrices too big for
REML estimates.
Nonlinear registered
• Used cross-
validation.
• Smooth an image
by different Rigidly aligned

amounts, see how

well it predicts other
images:
log p(t|μ) = ΣjΣk tjk log(μjk)
ML and MAP templates from 6
subjects
Nonlinear Registered Rigid registered

ML

MAP

log
 Overview
• Motivation
• Principles
• Optimisation
• Group-wise Registration
• Validation
– Sex classification
– Age regression
• Future directions
 Validation
• There is no “ground truth”
• Looked at predictive accuracy
– Can information encoded by the method make
predictions?
• Registration method blind to the predicted information
• Could have used an overlap of fMRI results
– Chose to see whether ages and sexes of subjects
could be predicted from the deformations
• Comparison with small deformation model
 Training and Classifying

?
Control
Training Data
?
?
?
Patient
Training Data
 Classifying

?
Controls

?
?
?
Patients
y=f(aTx+b
)
Support Vector Classifier
Support Vector Classifier (SVC)

Suppo a is a weighted linear

rt
combination of the
Vector
support vectors
Support
Vector

Support
Vector
Nonlinear SVC
 Support-vector classification
• Guess sexes of 471 subjects from brain
shapes
– 207 Females / 264 Males
• Use a random sample of 400 for training.
• Test on the remaining 71.
• Repeat 50 times.
 Sex classification results
• Small Deformation • DARTEL
– Linear classifier – Linear classifier
• 87.0% correct • 87.7% correct
• Kappa = 0.736 • Kappa = 0.749
– RBF classifier – RBF classifier
• 87.1% correct • 87.6% correct
• Kappa = 0.737 • Kappa = 0.748

An unconvincing improvement
 Regression
40

30
23

29

26
18

32
 Relevance-vector regression
• A Bayesian method, related to SVMs
– Developed by Mike Tipping
• Guess ages of 471 subjects from brain
shapes.
• Use a random sample of 400 for training.
• Test on the remaining 71.
• Repeat 50 times.
 Age regression results
• Small deformation • DARTEL
– Linear regression – Linear regression
• RMS error = 7.55 • RMS error = 7.90
• Correlation = 0.836 • Correlation = 0.813
– RBF regression – RBF regression
• RMS error = 6.68 • RMS error = 6.50
• Correlation = 0.856 • Correlation = 0.867

An unconvincing improvement
(slightly worse for linear regression)
 Overview
• Motivation
• Principles
• Optimisation
• Group-wise Registration
• Validation
• Future directions
 Future directions
• Compare with variable velocity methods
– Beg’s LDDMM algorithm.
• Classification/regression from “initial
momentum”.
• Combine with tissue classification model.
• Develop a proper EM framework for
generating tissue probability maps.
u

Hu
Variable velocity framework (as in LDDMM)

“Initial
momentum”
Variable velocity framework (as in LDDMM)

“Initial
momentum”
Thank you