page-title

Image reconstruction

Our group has done pioneering research in the following areas of image reconstruction:  4D (spatio-temporal) image processing algorithms in gated SPECT and dynamic PET, super-resolution in PET, 5D reconstruction in gated/dynamic SPECT, and mesh-based image reconstruction.

Mesh

Mesh model of the heart

Our current focus is the use of 4D reconstruction to improve image quality in gated cardiac SPECT, a standard imaging procedure to assess coronary artery disease. In 4D techniques, image sequences are treated as fully four-dimensional signals, consisting of three spatial dimensions plus time.  When combined with effective estimation of cardiac and respiratory motion, 4D reconstruction allows dramatic noise reduction to be achieved.  In published work, we have shown that 4D reconstruction can reduce radiation dose by a factor of 4, and recent work shows that a factor of 8 or more is possible.

In our mesh-based image reconstruction algorithms, the traditional pixel representation is replaced by a nonuniform sampling approach, which yields benefits for motion estimation, noise suppression, and computational efficiency.

In 5D methods, the time axis is split into a dynamic dimension (for large-scale time evolution) and a gated dimension (which captures a single cardiac cycle). The 5D approach may pave the way for alternative imaging protocols in which cardiac patients are evaluated in a single imaging session that provides information about cardiac perfusion, wall motion, and tracer kinetics simultaneously.

Our image reconstruction research has been sponsored by NIH/NHLBI and NIH/NINDS.

Selected publications

Fundamentals of mesh-based reconstruction

J. G. Brankov, Y. Yang, and M. N. Wernick, “Spatio-temporal processing of gated cardiac SPECT images using deformable mesh modeling,” Medical Physics, vol. 32, pp. 2839-2849, 2005.

J. Brankov, Y. Yang, M. N. Wernick, “Content-adaptive mesh modeling for tomographic image reconstruction,” IEEE Trans. on Medical Imaging, vol. 23, pp. 202-212, 2004.

Y. Yang, J. Brankov, and M. Wernick, “A computationally efficient approach for accurate content-adaptive mesh generation,” IEEE Trans. on Image Processing, vol. 12, no. 8, pp. 866-881, 2003.

T. Marin and J. G. Brankov, “Deformable left-ventricle mesh model for motion-compensated filtering in cardiac gated SPECT,” Medical Physics, vol. 37, Issue 10, 2010.

[back to top]

4D reconstruction for gated SPECT

Mingwu Jin, Xiaofeng Niu, Wenyuan Qi, Yongyi Yang, Joyoni Dey, Michael A. King, Seth Dahlberg, and Miles N. Wernick, “4D reconstruction for low-dose cardiac gated SPECT,” Medical Physics, vol. 40, no. 2, 2013.

W. Qi, Y. Yang, X. Niu, M. A. King, “A quantitative study of motion estimation methods on 4D cardiac gated SPECT reconstruction,” Medical Physics, vol. 39, no. 8, pp. 5182-5193, 2012.

X. Niu and Y. Yang, “Tomographic reconstruction of gated data acquisition using DFT basis functions,” IEEE Trans. on Image Processing, vol. 20, no. l, pp.176-185, 2011.

Xiaofeng Niu, Yongyi Yang, Mingwu Jin, Miles N. Wernick, Michael A. King, “Effects of motion, attenuation, and scatter corrections on gated cardiac SPECT reconstruction,” Medical Physics, vol. 38, pp. 6571-6584, 2011.

T. Marin and J. G. Brankov, “Deformable left-ventricle mesh model for motion-compensated filtering in cardiac gated SPECT,” Medical Physics, vol. 37, Issue 10, 2010.

Mingwu Jin, Yongyi Yang, Xiaofeng Niu, Thibault Marin, Jovan G. Brankov, Bing Feng, P. Hendrik Pretorius, Michael A. King, and Miles N. Wernick, “Quantitative evaluation study of four-dimensional gated cardiac
SPECT reconstruction,” Physics in Medicine and Biology, vol. 54, pp. 5643-5659, 2009.

E. Gravier, Y. Yang, M. A. King, and M. Jin, “Fully 4D motion-compensated reconstruction of cardiac SPECT images,” Phys. Med. Biol., vol. 51, pp. 4603-4619, 2006.

J. G. Brankov, Y. Yang, and M. N. Wernick, “Spatio-temporal processing of gated cardiac SPECT images using deformable mesh modeling,” Medical Physics, vol. 32, pp. 2839-2849, 2005.

[back to top]

5D reconstruction for gated + dynamic SPECT

L. Li and Y. Yang, “Optical flow estimation for a periodic image sequence,” IEEE Trans. on Medical Imaging, vol. 19, pp.1-10, 2010.

Xiaofeng Niu, Mingwu Jin, Yongyi Yang, Michael A. King, and Miles N. Wernick, “Regularized fully 5D reconstruction of cardiac gated dynamic SPECT images,” IEEE Transactions on Nuclear Science, vol. 57, no. 3, pp. 1085-1095, 2010.

M. Jin, Y. Yang, and M. A. King, “Reconstruction of dynamic gated cardiac SPECT,” Medical Physics, vol. 33, pp. 4384-4394, 2006.

[back to top]

4D reconstruction for dynamic SPECT

E. Gravier, Y. Yang, and M. Jin, “Tomographic reconstruction of dynamic cardiac image sequences,” IEEE Trans. on Image Processing, vol. 16, pp. 932-942, 2007.

Jovan G. Brankov, Miles N. Wernick, Michael A. King, Yongyi Yang, and Manoj V. Narayanan, “Spatially varying temporal smoothing for dynamic image sequences” IEEE Transactions on Nuclear Science, vol. 53, pp. 2769-2777, 2006.

V. Manoj Narayanan, Michael A. King, Miles N. Wernick, Charles L. Byrne, Edward J. Soares, and P. Hendrik Pretorius, “Improved image quality and computation reduction in 4D reconstruction of cardiac-gated SPECT images,” IEEE Transactions on Medical Imaging, vol. 19, pp. 423-433, 2000.

V. Manoj Narayanan, Michael A. King, Edward Soares, Charles Byrne, Hendrik Pretorius and Miles N. Wernick, “Application of the Karhunen-Loeve transform to 4D reconstruction of gated cardiac SPECT images,” IEEE Transactions on Nuclear Science, vol. 46, pp. 1001-1008, 1999.
5D reconstruction for gated + dynamic imaging SPECT

Xiaofeng Niu, Yongyi Yang, Michael A. King, and Miles N. Wernick, “Detectability of perfusion defect in five-dimensional gated-dynamic cardiac SPECT images,” Medical Physics, vol. 37, pp. 5102-5112, 2010.

Xiaofeng Niu, Mingwu Jin, Yongyi Yang, Michael A. King, and Miles N. Wernick, “Regularized fully 5D reconstruction of cardiac gated dynamic SPECT images,” IEEE Transactions on Nuclear Science, vol. 57, no. 3, pp. 1085-1095, 2010.

B. Feng, P. H. Pretorius, T. H. Farncombe, S. T. Dahlberg, M. V. Narayanan, M. N. Wernick, A. M. Celler, M.A. King, J.A. Leppo, “Imaging time-varying Tc-99m teboroxime localization and cardiac function simultaneously by five-dimensional (5D) gated-dynamic SPECT imaging and reconstruction,” presented at the American Society of Nuclear Cardiologists, 2003.

[back to top]

4D reconstruction for dynamic PET

Miles N. Wernick, E. James Infusino, and Milos Milosevic, “Fast spatio-temporal image reconstruction for dynamic PET,” IEEE Transactions on Medical Imaging, vol. 18, pp. 185-195, 1999.

Chien-Min Kao, Jeffrey T. Yap, Jogeshwar Mukherjee, and Miles N. Wernick, “Image reconstruction for dynamic PET based on low-order approximation and restoration of the sinogram,” IEEE Transactions on Medical Imaging, vol. 16, pp. 738-749, 1997.

[back to top]

Segmentation of dynamic PET and fMRI

Jovan G. Brankov, Nikolas P. Galatsanos, Yongyi Yang, and Miles N. Wernick, “Segmentation of dynamic PET or fMRI images based on a similarity metric,” IEEE Transactions on Nuclear Science, vol. 50, pp. 1410-1414, 2003.

[back to top]

Super-resolution in PET
Miles N. Wernick and Chin-Tu Chen, “Superresolved tomography by convex projections and detector motion,” Journal of the Optical Society of America A, vol. 9, pp. 1547-1553, 1992.

[back to top]