Post on 12-Jul-2020
ShapeMI workshop
MICCAI conference
20 September 2018
Granada, Spain
Alexandre BÎne, Maxime Louis, Benoît Martin, Stanley Durrleman
Deformetrica 4: an open-source software for statistical shape analysis
I. Registration
II. Atlas
III.Regression
Deformetrica 4: an open-source software for statistical shape analysis
demo
Registration
ðœð,ð¶ð
ð ð
Registration
ðœð,ð¶ð
ð ð
Registration
cost
functionregularization
cost
attachment
cost
ðž ð, ðŒ =1
ðð2 Ίð,ðŒ
ð â ð â ðâ°
2+ð (ð, ðŒ)
ðœð,ð¶ð
ð ð
Registration
ðž ð, ðŒ =1
ðð2 Ίð,ðŒ
ð â ð â ðâ°
2+ð (ð, ðŒ)
ðœð,ð¶ð
ð ð
inputs
Registration
ðž ð, ðŒ =1
ðð2 Ίð,ðŒ
ð â ð â ðâ°
2+ð (ð, ðŒ)
ðœð,ð¶ð
ð ð
outputsinputs
Registration
ðž ð, ðŒ =1
ðð2 Ίð,ðŒ
ð â ð â ðâ°
2+ð (ð, ðŒ)
ð ð
Hyper-
parametersoutputsinputs
ðœð,ð¶ð
Registration
ðž ð, ðŒ =1
ðð2 Ίð,ðŒ
ð â ð â ðâ°
2+ð (ð, ðŒ)
ð ð
Hyper-
parametersoutputsinputs
ðœð,ð¶ð
133
190
Registration
ðž ð, ðŒ =1
ðð2 Ίð,ðŒ
ð â ð â ðâ°
2+ð (ð, ðŒ)
ð ð
Hyper-
parametersoutputsinputs
ðœð,ð¶ð
133
190
Registration
ðž ð, ðŒ =1
ðð2 Ίð,ðŒ
ð â ð â ðâ°
2+ð (ð, ðŒ)
Hyper-
parametersoutputsinputs
Registration
ðž ð, ðŒ =1
ðð2 Ίð,ðŒ
ð â ð â ðâ°
2+ð (ð, ðŒ)
Hyper-
parametersoutputsinputs
Registration
ð ð
Registration
ð ð
Registration
ð ð
Registration
ð ð
Registration
ð ð
>> deformetrica estimate
model.xml data_set.xml âp
optimization_parameters.xml
Registration
ð ð
I. Registration
II. Atlas
III.Regression
Deformetrica 4: an open-source software for statistical shape analysis
demo
Deterministic atlas
Deterministic atlas
Deterministic atlas
ðž ð, ð, (ðŒð)ð =1
ðð2
ð=1
ð
Ίð,ðŒðð â ð â ðð â°
2+ ð (ð, ðŒð)
Hyper-
parametersoutputsinputs
ðž ð, ð, (ðŒð)ð =1
ðð2
ð=1
ð
Ίð,ðŒðð â ð â ðð â°
2+ ð (ð, ðŒð)
Deterministic atlas
Hyper-
parametersoutputsinputs
ðž ð, ð, (ðŒð)ð =1
ðð2
ð=1
ð
Ίð,ðŒðð â ð â ðð â°
2+ ð (ð, ðŒð)
Deterministic atlas
Hyper-
parametersoutputsinputs
ðž ð, ð, (ðŒð)ð =1
ðð2
ð=1
ð
Ίð,ðŒðð â ð â ðð â°
2+ ð (ð, ðŒð)
Deterministic atlas
Hyper-
parametersoutputsinputs
ðž ð, ð, (ðŒð)ð =1
ðð2
ð=1
ð
Ίð,ðŒðð â ð â ðð â°
2+ ð (ð, ðŒð)
Deterministic atlas
Hyper-
parametersoutputsinputs
>> deformetrica estimate
model.xml data_set.xml âp
optimization_parameters.xml
Deterministic atlas
Deterministic atlas
Deterministic atlas
I. Registration
II. Atlas
III.Regression
Deformetrica 4: an open-source software for statistical shape analysis
demo
Geodesic regression
ð¡1 = 5 ð¡2 = 15 ð¡3 = 25 ð¡4 = 35Yin et al. 2008, âA High- Resolution 3D Dynamic Facial Expression Databaseâ
Geodesic regression
ðž ð, ð, ðŒ =1
ðð2
ð=1
ð
Ίð,ð¡ðâðŒð â ð â ðð
â°
2+ ð (ð, ðŒ)
Hyper-
parametersoutputsinputs
ð¡1 = 5 ð¡2 = 15 ð¡3 = 25 ð¡4 = 35Yin et al. 2008, âA High- Resolution 3D Dynamic Facial Expression Databaseâ
ð¡1 = 5 ð¡2 = 15 ð¡3 = 25 ð¡4 = 35Yin et al. 2008, âA High- Resolution 3D Dynamic Facial Expression Databaseâ
Geodesic regression
ð¡1 = 5 ð¡2 = 15 ð¡3 = 25 ð¡4 = 35Yin et al. 2008, âA High- Resolution 3D Dynamic Facial Expression Databaseâ
Geodesic regression
Geodesic regression
>> deformetrica estimate
model.xml data_set.xml
ð¡1 = 5 ð¡2 = 15 ð¡3 = 25 ð¡4 = 35Yin et al. 2008, âA High- Resolution 3D Dynamic Facial Expression Databaseâ
ð¡1 = 5 ð¡2 = 15 ð¡3 = 25 ð¡4 = 35Yin et al. 2008, âA High- Resolution 3D Dynamic Facial Expression Databaseâ
Geodesic regression
Parallel transport
Transfer a reference temporal evolution towards a new target geometry
Data courtesy of Paolo Piras, Sapienza Università di Roma, Italy
MR image registration performance
Registration of full-resolution MR images (7 millions voxels) in 2-3 minutes, with low GPU memory usage
Teaser: graphical user interface alpha
Teaser: python API beta
PyTorch
⢠Auto-differentiation, without memory
overflows
⢠Seamless CUDA code
PyTorch + PyKeops
⢠Auto-differentiation, without memory
overflows
⢠Seamless CUDA code
Thanks to Benjamin Charlier, Jean Feydy & Joan GlaunÚs
Conclusion
Implements many statistical shape analysis tasks ...
⢠Registration
⢠Deterministic atlas
⢠Bayesian atlas
⢠Geodesic regression
⢠Parallel transport
⢠Longitudinal atlas
⢠Principal geodesic
analysis
beta
alpha
Conclusion
Implements many statistical shape analysis tasks ...
⢠Registration
⢠Deterministic atlas
⢠Bayesian atlas
⢠Geodesic regression
⢠Parallel transport
⢠Longitudinal atlas
⢠Principal geodesic
analysis
beta
alpha
... with very few requirements about the data
⢠Image
⢠Meshes
⢠No required point
correspondence
⢠Multi-object
⢠Cross-sectional or
longitudinal datasets
⢠Linux or Mac
⢠Anaconda 3
Requirements
Thanks!
Install
conda install -c pytorch -c conda-
forge
-c anaconda -c aramislab deformetrica
www.deformetrica.org
Come see us at the lunch & demo session!
Future work
Grow the pool of users
⢠Graphical user
interface (GUI)
⢠Python API
⢠Windows platform
Add functionalities
⢠Longitudinal atlas
⢠Principal geodesic
analysis
⢠MCMC-SAEM
estimation algorithm
Improve performance
⢠Achieve massive parallelization on large clusters
⢠Emphasis on GPU-specific optimizations
A decade of development
Deformetrica 1 C++
Deformetrica 3 C++
Deformetrica 4 PythonDeformetrica 2
C++
2011 2013 2017 2018
Deterministic atlas: landmark/2d/skulls
Deterministic atlas: landmark/2d/skulls
Deterministic atlas: landmark/2d/skulls
A note on the Bayesian atlas
ð¶ ð, (ðð)ð, ðð2 =
1
ðð2
ð=1
ð
Ίðð â ð â ðð â°
2+ð (ðð , ðð
2)
cost
functionregularization
cost
attachment
cost
A note on the Bayesian atlas
ð¶ ð, (ðð)ð, ðð2 =
1
ðð2
ð=1
ð
Ίðð â ð â ðð â°
2+ð (ðð , ðð
2)
Gives a statistical interpretation of the regularization term, which arises from assumed underlying random
structures on the momenta and residuals
In practice, no need to specify ððºð anymore!
The optimal tradeoff between attachment and
regularity terms is estimated from the data
Bayesian atlas
ð¶ ð, (ðð)ð, ðð2 =
1
ðð2
ð=1
ð
Ίðð â ð â ðð â°
2+ð (ðð , ðð
2)
cost
functionregularization
cost
attachment
cost
Statistical interpretation of the regularization term, which arises from assumed underlying random structures on
the momenta and residuals
In practice, no need to specify ððºð anymore!
Bayesian atlas
Registration
ð ð
Registration
ð ð
Registration
ð ð
Registration
ð ð
Registration
ð ð
>> deformetrica estimate model.xml
data_set.xml âp
optimization_parameters.xml