BIM 252/EEC 205 Computational Imaging Winter 2022


Project Description

The project is to provide you opportunities to explore contemporary issues in image reconstruction
beyond what have been covered in class. The project consists of two parts:

1. The first part is a paper review. You will pick one or more paper(s) on an image reconstruction
topic and write a critical review on the topic. It can be something related to your Ph.D. work or
some problem that you found interesting. The topic should be beyond what we already covered
in the lectures.

2. The second part is an experimental component. You can either implement a new reconstruction
method or perform a more in-depth study of a reconstruction method that we covered in the
lectures. The experimental study can be related to the topic that is reviewed in part 1 (for
example, you review a new reconstruction method in part 1 and implement it in part 2), but it
can also be independent of part 1. The scope of the problem should be something beyond what
you have done in the homework. However, it should also be restricted in scope, to fit the limited
time available. You can use any program language and any online resources, but you need to
perform your own experiments.

You are encouraged to team up with another student on the project.

The projects will be assessed based on the difficulty and importance of the problem, and on how well
you do with your problem. Both parts are required, although you can choose to put more effort on one
component than the other. It is OK to have a small experimental study if you already have an extensive
in-depth review, or to have a short review but with an extensive experimental component.

Submissions:
1. (20%) a short description of the project plan due Feb 27. It should describe the topic that you will
review and the experiment study you will perform. The entire description should only be a paragraph or
so.
2. (80%) project report due March 15. Your project report should be about 10 pages covering at least
the following components:
a) Paper review should include
a. an introduction to the topic
b. explanation of the new method proposed in the paper
c. discussion on its advantages and limitations.
b) Experimental section should include
a. description of the method and computer implementation
b. results from your program
c. analysis and discussion on your results.
If you use any existing codes that are either in public domain or borrowed from your friend, please
provide proper acknowledgement.

A few papers are listed below as examples. One topic is Fourier rebinning of Time-of-flight (TOF) PET
data. Another paper deals with different backprojector in FDK reconstruction. Other papers are related
to model-based iterative reconstruction and learning based reconstruction. You can also find many
other papers on image reconstruction.

Sample papers:
Fourier rebinning of TOF PET
A unified Fourier theory for time-of-flight PET data
Yusheng Li et al Physics in Medicine and Biology
2016 61 601
IOPscience

MAP reconstruction for Fourier rebinned TOF-PET data
Bing Bai et al Physics in Medicine and Biology
2014 59 925
IOPscience

Exact and approximate Fourier rebinning of PET data from time-of-flight to non time-of-flight -
http://dx.doi.org/10.1088/0031-9155/54/3/001
Phys Med Biol. 2009 Feb 7;54(3):467-84. doi: 10.1088/0031-9155/54/3/001.

Ahn S, Cho S, Li Q, Lin Y and Leahy R 2011 Optimal rebinning of time-of-flight PET data IEEE Trans. Med.
Imaging 30 1808–18 CrossRef

Distance-driven Projector
Bruno De Man and Samit Basu. Distance-driven projection and backprojection in three dimensions.
Phys. Med. Biol. 49 No 11 (7 June 2004) 2463-2475
Acrobat PDF (256 KB)


Image reconstruction from sparse views
Emil Y Sidky and Xiaochuan Pan. Image reconstruction in circular cone-beam computed tomography by
constrained, total-variation minimization
Phys. Med. Biol. 53 No 17 (7 September 2008) 4777-4807
Acrobat PDF (1.70 MB)


Fast reconstruction algorithms
Se Young Chun; Dewaraja, Y.K.; Fessler, J.A., "Alternating Direction Method of Multiplier for Tomography
With Nonlocal Regularizers," in Medical Imaging, IEEE Transactions on , vol.33, no.10, pp.1960-1968, Oct.
2014
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6825888&isnumber=6913038


Image regularization
G. Wang and J. Qi, "PET Image Reconstruction Using Kernel Method," in IEEE Transactions on Medical
Imaging, vol. 34, no. 1, pp. 61-71, Jan. 2015, doi: 10.1109/TMI.2014.2343916.

K. Gong et al., "Iterative PET Image Reconstruction Using Convolutional Neural Network Representation,"
in IEEE Transactions on Medical Imaging, vol. 38, no. 3, pp. 675-685, March 2019, doi:
10.1109/TMI.2018.2869871.

K. Gong, C. Catana, J. Qi and Q. Li, "PET Image Reconstruction Using Deep Image Prior," in IEEE
Transactions on Medical Imaging, vol. 38, no. 7, pp. 1655-1665, July 2019, doi:
10.1109/TMI.2018.2888491.

Learning based reconstruction
Zhu, B., Liu, J., Cauley, S. et al. Image reconstruction by domain-transform manifold learning. Nature 555,
487–492 (2018). https://doi.org/10.1038/nature25988

Häggström I, Schmidtlein CR, Campanella G, Fuchs TJ. DeepPET: A deep encoder-decoder network for
directly solving the PET image reconstruction inverse problem. Med Image Anal. 2019 May;54:253-262.
doi: 10.1016/j.media.2019.03.013.

A. Mehranian and A. J. Reader, "Model-Based Deep Learning PET Image Reconstruction Using Forward–
Backward Splitting Expectation–Maximization," in IEEE Transactions on Radiation and Plasma Medical
Sciences, vol. 5, no. 1, pp. 54-64, Jan. 2021, doi: 10.1109/TRPMS.2020.3004408.

Whiteley W, Luk WK, Gregor J. DirectPET: full-size neural network PET reconstruction from sinogram
data. J Med Imaging (Bellingham). 2020 May;7(3):032503. doi: 10.1117/1.JMI.7.3.032503. Epub 2020
Feb 28. PMID: 32206686; PMCID: PMC7048204.