Projects: Medical and Computational Imaging

Here we use programmable commodity graphics hardware boards (GPUs) to accelerate a wide variety of 3D computer tomographic (CT) reconstruction algorithms. So far we have achieved speed-ups of 1-2 orders of magnitude, without significant loss in reconstruction quality. Our RapidCT system enables fast, accelerated 3D reconstruction in diagnostic imaging, radio-therapy applications, surgery planning, electron microscopy, and others, at a fraction of the cost of proprietary devices. We have also recently explored the use of deep learning amd generative adversial networks (GAN) for image augmentation.

Novel Lung CT Image Synthesis at Full Hounsfield Range with Expert Guided Visual Turing Test
A Krishna, S Yenneti, G Wang, K Mueller
International Meeting on Fully Three-Dimensional Image Reconstruction
Stony Brook, NY, July 2023
 
Improving CT Image Segmentation Accuracy Using StyleGAN Driven Data Augmentation
S Bhosale, A Krishna, G Wang, K. Mueller
International Meeting on Fully Three-Dimensional Image Reconstruction
Stony Brook, NY, July 2023
 
Development of Metaverse for Intelligent Healthcare
G. Wang, A. Badal, X. Jia, J. Maltz, K. Mueller, K. Myers, C. Niu, M. Vannier, P. Yan, Z. Yu, R. Zeng
Nature Machine Intelligence
4, pp. 922–929, 2022
 
Anatomy-guided synthesis of novel CT images at full Hounsfield range
A. Krishna, G. Wang, K Mueller
IEEE Medical Imaging Conference
Milano, Italy, November 5-11, 2022
 
Noise Entangled GAN For Low-Dose CT Simulation
C. Niu, G. Wang, P. Yan, J. Hahn, Y. Lai, X. Jia, A. Krishna, K. Mueller, A. Badal, K. Myers, R. Zen
International Meeting on Fully Three-Dimensional Image Reconstruction
Leuven, Belgium, July 2021. (shortlisted for best oral award)
 
Image Synthesis for Data Augmentation in Medical CT using Deep Reinforcement Learning
A. Krishna, K. Bartake, C. Niu, G. Wang, Y. Lai, X. Jia, K. Mueller
International Meeting on Fully Three-Dimensional Image Reconstruction
Leuven, Belgium, July 2021. (shortlisted for best oral award)
 
Interpreting Galaxy Deblender GAN from the Discriminator’s Perspective
H. Li, Y. Lin, W. Xu, K. Mueller
International Symposium on Visual Computing
Virtual, October 5-7, 2020.
 
Metal Artifact Reduction in X-ray CT via Ray Profile Correction
S. Ha, K. Mueller
Applied Sciences
10, 66, 2020
 
Medical (CT) Image Generation with Style
A. Krishna, K. Mueller
International Meeting on Fully Three-Dimensional Image Reconstruction (Fully3D)
Philadelphia, PA, June 2019
 
A GPU-Accelerated Multi-Voxel Update Scheme for Iterative Coordinate Descent (ICD) Optimization in Statistical Iterative CT Reconstruction (SIR)
S. Ha, K. Mueller
IEEE Transactions on Computational Imaging
4(3): 355-365, 2018
 
Image Reconstruction Is a New Frontier of Machine Learning
G. Wang. J. C. Ye, K. Mueller, J. A. Fessler,
IEEE Transactions on Medical Imaging
7(6), 1289-1296, 2018
 
Metal Artifact Reduction in Cone-Beam X-Ray CT via Ray Profile Correction
S. Ha, K. Mueller
Extended article to original SPIE Medical Imaging 2016 publication
July, 2018
 
A Look-Up Table-Based Ray Integration Framework for 2D/3D Forward and Back-projection in X-ray CT
S. Ha, K. Mueller
IEEE Transactions on Medical Imaging
37(2): 361-371, 2018
Low-Dose CT Streak Artifacts Removal using Deep Residual Neural Networks
H. Li, K. Mueller
Fully 3D Image Reconstruction in Radiology and Nuclear Medicine
Xi'an, China, June 2017
Efficient Area-Based Ray Integration Using Summed Area Tables and Regression Models
S. Ha, H. Li, K. Mueller
International Conference on Image Formation in X-Ray Computed Tomography
Bamberg, Germany, July, 2016
 
Metal Artifact Reduction in CT via Ray Profile Correction
S. Ha, K. Mueller
SPIE Medical Imaging Conference
San Diego, CA, February, 2016
 
MADR: Metal Artifact Detection and Reduction
S. Jaiswal, S. Ha, K. Mueller
SPIE Medical Imaging Conference
San Diego, CA, February, 2016
 
A Study of Sparse Detector Designs with Interpolation for Multi-Slice Spiral CT
D. Coelho, K. Mueller
IEEE Medical Imaging Conference, (poster)
San Diego, CA, November, 2015.
 
An Algorithm to Compute Independent Sets of Voxels for Parallelization of ICD-based Statistical Iterative Reconstruction
S. Ha, K. Mueller
Fully 3D Image Reconstruction in Radiology and Nuclear Medicine, (oral)
Newport, RI, June 2015.
A Study of Volume Integration Models for Iterative Cone-Beam Computed Tomography
S. Ha, A. Kumar, K. Mueller
Fully 3D Image Reconstruction in Radiology and Nuclear Medicine, (poster)
Newport, RI, June 2015.
 
Low Dose CT Image Restoration Using a Database of Image Patches
S. Ha, K. Mueller
Physics in Medicine and Biology
60(2):869-880, 2015.
DQS Advisor: A Visual Interface and Knowledge-Based System to Balance Dose, Quality, and Reconstruction Speed in Iterative CT Reconstruction with Application to NLM-Regularization
Z. Zheng, E. Papenhausen, K. Mueller
Physics in Medicine and Biology
58(21):7857-73, 2013
Rapid Rabbit: Highly Optimized GPU Accelerated Cone-Beam CT Reconstruction
E. Papenhausen, Z. Zheng, K. Mueller
IEEE Medical Imaging Conference
Seoul, Korea, November, 2013
 
DQS Advisor: a Visual Interface to Balance Dose, Quality and Reconstruction Speed in Iterative CT
Z. Zheng, E. Papenhausen, K. Mueller
IEEE Medical Imaging Conference
Seoul, Korea, November, 2013
 
Low Dose CT Image Restoration Using a Localized Patch Database
S. Ha, K. Mueller
IEEE Medical Imaging Conference
Seoul, Korea, November, 2013
 
VACT: Visualization-Aware CT Reconstruction
Z. Zheng, K. Mueller
IEEE Medical Imaging Conference
Seoul, Korea, November, 2013
 
Cloud X: A Platform as a Service for CT Reconstruction Research and Development
E. Papenhausen, Z. Zheng, K. Mueller
Fully 3D Image Reconstruction in Radiology and Nuclear Medicine (oral)
Lake Tahoe, CA, June 2013
 
A Comparative Study of Neighborhood Filters for Artifact Reduction in Iterative Low-Dose CT
W. Xu, S. Ha, Z. Zheng, K. Mueller
Fully 3D Image Reconstruction in Radiology and Nuclear Medicine (poster)
Lake Tahoe, CA, June 2013
 
Database-Assisted Low-Dose CT Image Restoration
W. Xu, S. Ha, K. Mueller
Medical Physics
3(40):031109, 2013
 
Creating Optimal Code for GPU-Accelerated CT Reconstruction Using Ant Colony Optimizationn
E. Papenhausen, Z. Zheng, K. Mueller
Medical Physics 
3(40): 031110, 2013
 
Detection of Low-Dose CT Reconstruction Artifacts Using a Bi-Modal Approach
S. Mahmood, K. Mueller
SPIE Medical Imaging Conference
Orlando, FL, February, 2013
 
GPU-Accelerated Forward and Back-Projections with Spatially Varying Kernels for 3D DIRECT TOF PET Reconstructionn
S. Ha, S. Matej, M.Ispiryan, K. Mueller
IEEE Transactions on Nuclear Science
60(1):166-173, 2013
 
Effcient Low-Dose CT Artifact Mitigation using an Artifact-Matched
Prior Scan

W. Xu, K. Mueller
Medical Physics 
39:4748-4760, 2012
 
GPU-Accelerated Forward and Backward Projection with Spatially Varying Kernels in 3D DIRECT TOF PET Reconstruction
S. Ha, S. Matej, M.Ispiryan, K. Mueller
IEEE Medical Imaging Conference 
Anaheim, CA, October 2012
 
Coding Ants: Using Ant Cclony Optimization to Accelerate CT Reconstruction
E. Papenhausen, Z. Zheng, K. Mueller
Proc. 2nd International Conference on Image Formation in X-ray Computed Tomography
(Premium Poster) 
Salt Lake City, UT, July 2012
 
Searching Effective Parameters for Low-Dose CT Reconstruction by Ant Colony Optimization
Z. Zheng, E. Papenhausen, K. Mueller
Proc. 2nd International Conference on Image Formation in X-ray Computed Tomography 
Salt Lake City, UT, July 2012
 
GPU-Accelerated First-Order Scattering Simulation for X-ray CT Image Reconstruction
S. Ha, J. Pi, K. Mueller
Proc. 2nd International Conference on Image Formation in X-ray Computed Tomography
Salt Lake City, UT, July 2012
 
Database-Assisted Low-Dose CT Image Restoration
W. Xu, S. Ha, K. Mueller
Proc. 2nd International Conference on Image Formation in X-ray Computed Tomography
(Oral presemtation)
Salt Lake City, UT, July 2012
 
A Reference Image Database Approach for NLM Filter-Regularized CT Reconstruction
W. Xu, K. Mueller
Fully 3D Image Reconstruction in Radiology and Nuclear Medicine 
Potsdam, Germany, July, 2011
 
GPU-Accelerated Back-Projection Revisited: Squeezing Performance by Careful Tuning
E. Papenhausen, Z. Zheng, K. Mueller
Workshop on High Performance Image Reconstruction (HPIR)  
Potsdam, Germany, July, 2011
 
Identifying Sets of Favorable Projections for Few-View Low-Dose Cone-Beam CT Scanning
Z. Zheng, K. Mueller
Fully 3D Image Reconstruction in Radiology and Nuclear Medicine
Potsdam, Germany, July, 2011
 
Performance Tuning for CUDA-Accelerated Neighborhood Denoising Filters
Z. Zheng, W. Xu, K. Mueller
Workshop on High Performance Image Reconstruction (HPIR) 
Potsdam, Germany, July, 2011
 
Efficiently GPU-Accelerating Long Kernel Projections in 3-D DIRECT TOF PET Reconstruction via a Memory Cache Optimization
S. Ha, S. Matej, K. Mueller
Workshop on High Performance Image Reconstruction (HPIR)
Potsdam, Germany, July, 2011
 
Evaluating Popular Non-Linear Image Processing Filters for their Use in Regularized Iterative CT
W. Xu, K. Mueller
IEEE Medical Imaging Conference
Knoxville, TN, October, 2010
 
A Cache-Aware GPU Memory Scheduling Scheme for CT Reconstruction Back-Projection
Z. Zheng, K. Mueller
IEEE Medical Imaging Conference
Knoxville, TN, October, 2010
 
Efficiently GPU-Accelerating Long Kernel Convolutions in 3-D DIRECT TOF PET Reconstruction via a Kernel Decomposition Scheme
S. Ha, Z. Zhang, S. Matej, K. Mueller
IEEE Medical Imaging Conference
Knoxville, TN, October, 2010
 
High-Performance Iterative Electron Tomography Reconstruction with Long-Object Compensation using Graphics Processing Units (GPUs)
W. Xu, F. Xu, M. Jones, B. Keszthelyi, J. Sedat, D. Agard, K. Mueller
Journal of Structural Biology
171(2):142-153, 2010
 
Refraction Corrected Transmission Ultrasound Computed Tomography for Application in Breast Imaging
S. Li, M. Jackowski, T. Varslot, D. Dione, L. Staib, K. Mueller
Medical Physics
37(5):2233-2246, 2010
 
On the Efficiency of Iterative Ordered Subset Reconstruction Algorithms for Acceleration on GPUs
F. Xu, W. Xu, M. Jones, B. Keszthelyi, J. Sedat, D. Agard, K. Mueller
Computer Methods and Programs in Biomedicine
98(3):261-270, 2010 (erratum file)
GPU-Based Ultrafast IMRT Plan Optimization
C. Men, X. Gu, D. Choi, A. Majundar. Z. Zheng, K. Mueller, S. Jiang
Physics in Medicine and Biology
54, pp. 6565-6573, 2009
 
Optimal Sampling Lattices for High-Fidelity CT Reconstruction
F. Xu and K. Mueller
IEEE Medical Imaging Comference
Orlando, FL, October, 2009
Learning Effective Parameter Settings for Iterative CT Reconstruction Algorithms
W. Xu, K. Mueller
Fully 3D Image Reconstruction in Radiology and Nuclear Medicine
Beijing, China, September, 2009
A Performance-Driven Study of Regularization Methods for GPU-Accelerated Iterative CT
W. Xu, K. Mueller
Workshop on High Performance Image Reconstruction (HPIR)
Beijing, China, September, 2009
Accelerating Regularized Iterative CT Reconstruction on Commodity Graphics Hardware (GPU)
W. Xu, K. Mueller
IEEE International Symposium on Biomedical Imaging (ISBI)
Boston, MA, July, 2009

On the Efficiency of Iterative Ordered Subset Reconstruction Algorithms for Acceleration on GPUs
F. Xu, K. Mueller, M. Jones, B. Keszthelyi, J. Sedat, D. Agard
MICCAI '08 (Workshop on High-Performance Medical Image Computing & Computer Aided Intervention)
New York, NY, September 2008

Physical-Space Refraction-Corrected Transmission Ultrasound Computed Tomography Made Computationally Practical
S. Li, K. Mueller, M. Jackowski, D. Dione, L. Staib
MICCAI '08
New York, NY, September 2008
Springer Lecture Notes In Computer Science, vol. 5242, pp. 280-288

GPU-Acceleration of Attenuation and Scattering Compensation in Emission Computed Tomography
F. Xu and K. Mueller
9th Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine '07
High Performance Image Reconstruction Workshop
Lindau, Germany, July 2007

Real-Time 3D Computed Tomographic Reconstruction Using Commodity Graphics Hardware
F. Xu and K. Mueller
Physics in Medicine and Biology
vol. 52, pp. 3405–3419
, 2007
Hardware acceleration vs. algorithmic acceleration: Can GPU-based processing beat complexity optimization for CT?
N. Neophytou, F. Xu, and K. Mueller
SPIE Medical Imaging '07
San Diego, February 2007
Why do GPUs work so well for acceleration of CT?
K. Mueller, F. Xu, and N. Neophytou
SPIE Electronic Imaging '07 (Keynote, Computational Imaging V)
San Jose, January 2007

Fast Marching Method to correct for refraction in ultrasound computed tomography
S. Li, K. Mueller, M. Jackowski, D. Dione, and L. Staib
IEEE 2006 International Symposium on Biomedical Imaging (ISBI '06)
Arlington, VA, April 2006

pp. 896-899, 2006

 
Practical considerations for GPU-accelerated CT
K. Mueller and F. Xu
IEEE 2006 International Symposium on Biomedical Imaging (ISBI '06)
Arlington, VA, April 2006

pp. 1184-1187, 2006

A comparative study of popular interpolation and integration methods for use in computed tomography
F. Xu and K. Mueller
IEEE 2006 International Symposium on Biomedical Imaging (ISBI '06)
Arlington, VA, April 2006

pp. 1252-1255, 2006


Accelerating Popular Tomographic Reconstruction Algorithms On Commodity PC Graphics Hardware
F. Xu and K. Mueller
IEEE Transactions on Nuclear Science
vol. 52, no. 3, pp. 654-663, 2005

Ultra-Fast 3D Filtered Backprojection on Commodity Graphics Hardware
F. Xu and K. Mueller
IEEE International Symposium on Biomedical Imaging (ISBI '04)
Arlington, VA, April 2004


Towards a Unified Framework for Rapid Computed Tomography on Commodity GPUs
F. Xu and K. Mueller
IEEE Medical Imaging Conference (MIC) 2003
Portland, OR, October 2003


Conceptual design of a proton computed tomography system for applications in proton radiation therapy
R. Schulte, V. Bashkirov, T. Li, Z. Liang, K. Mueller,  et al.,
IEEE Transactions on Nuclear Science
vol. 51, no. 3, pp. 866-872, 2004


Reconstruction for Proton Computed Tomography: A Practical Approach
K. Mueller, Z. Liang, T. Li, F. Xu, et al.
IEEE Medical Imaging Conference (MIC) 2003
Portland, OR, October 2003

Design of a Proton Computed Tomography System for Applications in Proton Radiation Therapy 
R. Schulte, V. Bashkirov, T. Li, J. Liang, K. Mueller, et al.
IEEE Medical Imaging Conference (MIC) 2003
Portland, OR, October 2003


Reconstruction for Proton Computed Tomography: A Monte Carlo Study
T. Li, Z. Liang, K. Mueller, et al.
IEEE Medical Imaging Conference (MIC) 2003
Portland, OR, October 2003

Rapid 3D cone-beam reconstruction with the Simultaneous Algebraic Reconstruction Technique (SART) using 2D texture mapping hardware
K. Mueller and R. Yagel
IEEE Transactions on Medical Imaging
vol. 19, no. 12, pp. 1227-1237, 2000

Anti-aliased 3D cone-beam reconstruction of low-contrast objects with algebraic methods
K. Mueller, R. Yagel, and J.J. Wheller
IEEE Transactions on Medical Imaging
vol. 18, no. 6, pp. 519-537, 1999

Fast implementations of algebraic methods for the 3D reconstruction from cone-beam data
K. Mueller, R. Yagel, and J.J. Wheller
IEEE Transactions on Medical Imaging
vol. 18, no. 6, pp. 538-547, 1999

On the use of graphics hardware to accelerate algebraic reconstruction methods
K. Mueller and R. Yagel
1999 SPIE Medical Imaging Conference, No. 3659-62
San Diego, February 1999

Fast and accurate three-dimensional reconstruction from cone-beam projection data using algebraic methods
K. Mueller
PhD dissertation, The Ohio State University
June, 1998


A fast and accurate projection algorithm for the Algebraic Reconstruction Technique (ART)
K. Mueller, R. Yagel, and J.J. Wheller
1998 SPIE Medical Imaging Conference, Vol. SPIE 3336
San Diego, February 1998
pp. 24-732 (won Honorary Mention Award) 
 

The weighted distance scheme: a globally optimizing projection ordering method for the Algebraic Reconstruction Technique (ART)
K. Mueller, R. Yagel, and J.F. Cornhill
IEEE Transactions on Medical Imaging
vol. 16, no. 2, pp. 223-230, 1997
 

Accelerating the anti-aliased Algebraic Reconstruction Technique (ART) by table-based voxel backward projection
K. Mueller, R. Yagel, and J.F. Cornhill
EMBS'95 (Conference of the IEEE Engineering in Medicine and Biology Society)
Montreal, June 1995
pp. 579-580
 

Some current projects include the use of programmable commodity graphics hardware (GPUs) to accelerate a wide variety of 3D tomographic reconstruction algorithms. So far we have achieved speedups of 1-2 orders of magnitude (a factor of 10-100), without significant losses in reconstruction quality (see here for more information). Our approach enables rapid 3D reconstruction in diagnostic imaging, radio-therapy applications, surgery planning, electron microscopy, and others, at a fraction of the cost of proprietary hardware. Related projects are the 3D reconstruction from transmission ultrasound, transmission electron microscopy, MV-CT and proton-CT data, projections obtained via mobile X-ray source/detector pairs, and functional imaging applications, including MRI, functional MRI, SPECT, and PET.

 

Home

 


Last update: June 27, 2005
http://www.cs.stonybrook.edu/~mueller