Research

Image Reconstruction Methods

Advanced reconstructions including physical and/or data-driven models can produce high quality images from incomplete or noisy data. Effective use of these reconstructions introduce a number of new challenges, however, including dealing with variable image quality in training data, measuring image content complexity or image quality automatically, and effective reconstruction in the presence of motion or other perturbations. This research was a collaboration with Dr. Christopher Kramer and Dr. Michael Salerno, and was supported by the Thomas F. and Kate Miller Jeffress Memorial Trust, Bank of America, Trustee, the UVA Center for Engineering in Medicine, and the National Institutes of Health (NIH) under award R56 EB028254.

Selected Publications:

1. Haris Jeelani, Yang Yang, Ruixi Zhou, Christopher Kramer, Michael Salerno, and Daniel Weller. "A Myocardial T1-Mapping Framework with Recurrent and U-Net Convolutional Neural Networks." IEEE International Symposium on Biomedical Imaging. Virtual conference, April 2020, pp. 1941-1944. DOI: 10.1109/ISBI45749.2020.9098459

2. Tanjin Taher Toma and Daniel Weller. "Fast Automatic Parameter Selection for MRI Reconstruction." IEEE International Symposium on Biomedical Imaging. Virtual conference, April 2020, pp. 1078-1081. DOI: 10.1109/ISBI45749.2020.9098569

3. Haris Jeelani, Yang Yang, Roshin Mathew, Michael Salerno, and Daniel S. Weller. "Fast and Robust T1-mapping using Convolutional Neural Networks." ISMRM 27th Annual Meeting. Montreal, Canada, May 2019, p. 2134.

4. Daniel S. Weller, Michael Salerno, and Craig H. Meyer. "Content-Aware Compressive Magnetic Resonance Image Reconstruction." Magnetic Resonance Imaging, vol. 52, pp. 118-130, October 2018. DOI: 10.1016/j.mri.2018.06.008

5. Haris Jeelani, Jonathan Martin, Francis Vasquez, Michael Salerno, and Daniel S. Weller. "Image Quality Affects Deep Learning Reconstruction of MRI." IEEE International Symposium on Biomedical Imaging. Washington, DC, USA, April 2018, pp. 357-360. DOI: 10.1109/ISBI.2018.8363592 (preprint)

6. Daniel S. Weller. "Reconstruction with Dictionary Learning for Accelerated Parallel Magnetic Resonance Imaging." IEEE Southwest Symposium on Image Analysis and Interpretation. Santa Fe, NM, USA, March 2016, pp. 105-108. DOI: 10.1109/SSIAI.2016.7459186

Reconstruction Motion and Artifact Correction

One drawback of MRI is its susceptibility to a range of artifact-inducing phenomena, many of which can distort images to render them no longer useful. A frequent cause of such artifacts is subject motion, either body movements while lying in the scanner, or motion of the internal organs while breathing. Another such challenge, nonuniform magnetic fields can cause unacceptable signal loss or geometric distortions when imaging rapidly. Earlier work on head motion correction in collaboration with Jeffrey Fessler and Douglas Noll at the University of Michigan (supported by NIH F32 EB015914) investigated new ways for learning and correcting for head motion and nonuniform fields during the reconstruction. Recent efforts at UVA with Craig Meyer, John Mugler, and Michael Salerno addressed motion in MRI's in young children, who tend to move in the scanner, and in freely-breathing MRI's of cardiac patients, where the heart moves during the breathing cycle. The pediatric imaging work was supported by NIH R21 EB022309, and the cardiac imaging work was supported by NIH R56 EB028254.

Selected Publications:

1. Daniel S. Weller, Douglas C. Noll, and Jeffrey A. Fessler. "Real Time Filtering with Sparse Variations for Head Motion in Magnetic Resonance Imaging." Signal Processing, vol. 57, pp. 170-179, April 2019. DOI: 10.1016/j.sigpro.2018.12.001 (preprint) (supplement)

2. Daniel S. Weller, Luonan Wang, John P. Mugler III, and Craig H. Meyer. "Motion-compensated reconstruction of magnetic resonance images from undersampled data." Magnetic Resonance Imaging, vol. 55, pp. 36-45, January 2019. DOI: 10.1016/j.mri.2018.09.008 (preprint)

3. Haris Jeelani, Yang Yang, Michael Salerno, and Daniel S. Weller. "Evaluation of k-Space and Image-Space Motion Correction Schemes for CMR Perfusion." 20th Annual SCMR Scientific Sessions, Washington, DC, USA, February 2017, p. P046. (preprint)

4. Luonan Wang and Daniel S. Weller. "Joint Motion Estimation and Image Reconstruction Using Alternating Minimization." ISMRM 24nd Scientific Meeting. Singapore, May 2016, p. 1800. (preprint)

5. Daniel S. Weller and Jeffrey A. Fessler. "Fast non-Cartesian L1-SPIRiT with Field Inhomogeneity Correction." ISMRM 22nd Scientific Meeting. Milan, Italy, May 2014, p. 84. Summa cum laude award. (preprint)

6. Daniel S. Weller, Douglas C. Noll, and Jeffrey A. Fessler. "Prospective Motion Correction for Functional MRI Using Sparsity and Kalman Filtering." SPIE Wavelets and Sparsity XV, vol. 8858, August 2013, pp. 885823-1-10. DOI: 10.1117/12.2023074

Automatic Enhancement, Reconstruction, and Analysis for Light Microscopy

Automatic methods are needed to deal with the high volumes of light microscope imaging data obtained by neuroscientists studying the behavior and interactions among individual neurons and glia in the brain. This research had two thrusts: enhancing and analyzing high-resolution video of the behavior of microglia and similar cells in two-photon microscope videos in different states, and enhancing, reconstructing, and analyzing three-dimensional volumes describing targeted neurons activated during memory formation, epileptic seizures, and other stimuli in the dentate gyrus, dorsal hippocampus, and other interconnected brain regions. The first of these was supported by NSF Grant 1759802, in collaboration with Dr. Scott Acton and Dr. Gustavo Rohde. More information can be found on the Neuroglia Image Toolkit project website. The second project was a collaboration with Dr. Jaideep Kapur and Dr. Cedric Williams. Both of these projects involved developing automatic image quality comparison and parameter selection tools and their integrating them with new image enhancement and reconstruction methods.

Preprints: (unpublished)

1. Haoyi Liang, Aijaz Naik, Cedric L. Williams, Jaideep Kapur, and Daniel S. Weller. "Enhanced Center Coding for Cell Detection with Convolutional Neural Networks." Available online: arXiv:1904.08864

Selected Publications:

1. Tanjin Taher Toma, Kanchan Bisht, Ukpong Eyo, and Daniel Weller. "VBET: Vesselness and Blob Enhancement Technique for 2D and 3D Microscopy Images of Microglia." IEEE Asilomar Conference on Signals, Systems, and Computers. Virtual conference, November 2020, pp. 256-260. DOI: 10.1109/IEEECONF51394.2020.9443366

2. Haris Jeelani, Haoyi Liang, Scott T. Acton, and Daniel S. Weller. "Content-Aware Enhancement of Images with Filamentous Structures." IEEE Transactions on Image Processing, vol. 28, no. 7, pp. 3451-3461, July 2019. DOI: 10.1109/TIP.2019.2897289 (preprint) (supplement)

3. Haoyi Liang, Natalia Dabrowska, Jaideep Kapur, and Daniel S. Weller. "Structure-based Intensity Propagation for 3D Brain Reconstruction with Multilayer Section Microscopy." IEEE Transactions on Medical Imaging, vol. 38, no. 5, pp. 1106-1115, May 2019. DOI: 10.1109/TMI.2018.2878488

4. Haoyi Liang, Natalia Dabrowska, Jaideep Kapur, and Daniel Weller. "Structure Correction for 3D Mouse Brain Reconstruction." IEEE International Symposium on Biomedical Imaging. Washington, DC, USA, April 2018, abstract. (preprint)

5. Haoyi Liang, Natalia Dabrowska, Jaideep Kapur, and Daniel Weller. "Whole Brain Reconstruction from Multilayered Sections of a Mouse Model of Status Epilepticus." IEEE Asilomar Conference on Signals, Systems, and Computers. Pacific Grove, CA, USA, October 2017, pp. 1260-1263. DOI: 10.1109/ACSSC.2017.8335554

6. Haoyi Liang, Scott T. Acton, and Daniel S. Weller. "Content-Aware Neuron Image Enhancement." IEEE International Conference on Image Processing. Beijing, China, September 2017, pp. 3510-3514. DOI: 10.1109/ICIP.2017.8296935

7. Haoyi Liang and Daniel S. Weller. "Comparison-based Image Quality Assessment for Selecting Image Restoration Parameters." IEEE Transactions on Image Processing, vol. 25, no. 11, pp. 5118-5130, November 2016. DOI: 10.1109/TIP.2016.2601783

8. Haoyi Liang and Daniel S. Weller. "Denoising method selection by comparison-based image quality assessment." IEEE International Conference on Image Processing. Phoenix, AZ, USA, September 2016, pp. 3106-3110. DOI: 10.1109/ICIP.2016.7532931

9. Haoyi Liang and Daniel S. Weller. "Regularization Parameter Trimming for Iterative Image Reconstruction." IEEE Asilomar Conference on Signals, Systems, and Computers. Pacific Grove, CA, USA, November 2015, pp. 755-759. DOI: 10.1109/ACSSC.2015.7421235