A Computationally Efficient Approach for 2D-3D Image
Registration
Juri Minxha Medical Image Analysis
Professor Benjamin KimiaSpring 2011
Brown University
Problem Statement2 Signal Sources - 3D volumetric data (CT scan, MRI) - 2D images (ex. frame from fluoroscopy
video)
Project 3D data onto a 2D plane and compare it to existing 2D image. The projected image is also known as the digitally reconstructed radiograph (DRR)
Brown University
Why do we need this?Image guided surgery
Pre-operative data (CT/MRI acquisitions) Good resolution 3D data Slow Acquisition
Intra-operative data (fluoroscopy images) Can be quickly acquired Poor resolution, more noise (ex. temporal blurring)
Brown University
Typical Approach to Registration
Similarity Metric Optimization
1. Similarity Metric Mutual Information, Cross-Correlation, Correlation Ratio,
Cross Correlation Residual Entropy
2. Optimization, Non-gradient vs. GradientGauss-Newton, steepest descent, Levenberg-Larquardt, simplex method etc.
The main challenge is:Minimize computation time
Brown University
Approach Outlined in this paper
1. Similarity Measure: Sum of Conditional Variances
2. Optimization Algorithm: Gauss-Newton
1. Requires computation of gradient2. Fast convergence
Brown University
Similarity Metric: SoCVI0 Ro
R0 =100·ln(256-I0)-300
1. Quantize images to 64 possible values2. Each pixel in the image on the left corresponds to
a bin in the histogram (64 x 64 bins)3. Notice the non-linear relationship between
I and R
Similarity Metric: SoCV
What happens if I0 is translated to the right?
For each value of R, we have a range of values in I’
So far…Similarity metric: Sum of Conditional
VariancesOptimization StepThe optimization step only converges for
certain casesOptimization over 1 variable only (needs to
be debugged)Testing with MRI data built into Matlab
Plan of Action
Brown University
Fix optimization over all 6 parameters [rx ry rz
tx ty tz ]Test on a real data set Implement the computationally efficient
approach to this algorithm from their follow up paper
Test on real data set and compare computation speed to original
April 28 - May 2 May 2 – May 10
May 10 - May 14
ReferencesA computationally efficient approach for 2D-3D image
registrationHaque, M.N.; Pickering, M.R.; Biswas, M.; Frater, M.R.; Scarvell, J.M.; Smith, P.N.; 2010 Annual International Conference of the IEEEEngineering in Medicine and Biology Society (EMBC), Issue Date: Aug. 31 2010-Sept. 4 2010, On page(s): 6268 – 6271
M. Pickering, A. Muhit, J. Scarvell, and P. Smith, "A new multimodalsimilarity measure for fast gradient-based 2D-3D imageregistration," in Proc. IEEE Int. Conf. on Engineering in Medicineand Biology (EMBC), Minneapolis, USA, 2009, pp. 5821-5824.
Brown University