Other Ongoing projects in our lab include the following:
Generation of numerical optical and acoustic breast phantoms from MRI data
we develop and implement a methodology for generating anatomically realistic numerical breast phantoms from clinical contrast-enhanced magnetic resonance imaging data. The phantoms will depict vascular structures and the volumetric distribution of different tissue types in the breast. By assigning optical and acoustic parameters to different tissue structures, both optical and acoustic breast phantoms will be established for use in PACT and USCT studies. Details regarding the database of the generated phantoms is here.
Optical and Acoustic Diffraction Tomography
Diffraction tomography (DT) is a 3D imaging method that is applicable to a wide range of studies in optics and acoustics. We have developed a variety of novel analytic image reconstruction algorithms for DT that exploit symmetries of the imaging operator to reduce noise levels in the reconstructed images. In collaboration with Dr. Greg Gbur, we have also developed a variety of image reconstruction algorithms for reconstructing the refractive index of a weakly scattering object from intensity measurements. This class of methods has been referred to as intensity diffraction tomography (I-DT). We have developed I-DT reconstruction methods for use with spherical waves, novel scanning geometries, and multi-spectral measurements.
Improved Image Reconstruction Methods for Radiation Therapy Applications
(New Project. Collaboration with Dr. Deshan Yang)
Cone-beam computed tomography (CBCT) is widely employed in image-guided radiation therapy (IGRT) to obtain patient’s anatomy for precise localization before and during radiation delivery. However, there remains an important need to improve image quality and reduce imaging doses. In recent years, there have been numerous advancements in iterative CBCT image reconstruction algorithms based on the burgeoning field of compressive sampling. While such algorithms will likely facilitate dose reduction in CBCT, they are computationally burdensome, even by use of graphics processing units (GPUs), and long image reconstruction times render them unsuitable for current clinical implementation. We are developing a novel fast convergent iterative image reconstruction algorithm that can be used to improve image quality and reduce the imaging dose in radiation therapy applications of CBCT. By use of multiple GPUs, our methods can reconstruct volumetric images in a few minuets or less. We will be evaluating our reconstruction methods in an IRB-approved study employing patient data, to determine if our advanced image reconstruction methods can improve clinical outcomes. Qiaofeng Xu, a doctoral student in our Lab, was named a finalist in the John R. Cameron Young investigator Competition at the 2012 American Association of Physicists in Medicine (AAPM) meeting, where he presented work related to this project.
(New Project. Collaboration with Dr. Carsten Schirra, Philips Medical Systems)
The broad objective of this project is to develop and evaluate innovative image reconstruction algorithms for spectral CT that will facilitate significant reductions (a factor of 10 or more) in data-acquisition times. We seek to minimize data-acquisition times by reducing the number of tomographic projection measurements that are required to produce sufficient image quality for a given image reconstruction algorithm and defined diagnostic task. Statistically principled iterative image reconstruction algorithms based on modern compressive sensing theory are being developed to mitigate image artifacts associated with the use of few-view projection data, thereby permitting larger reductions in data-acquisition times than would be possible by use of conventional image reconstruction algorithms. The application of our reconstruction methods to accelerated k-edge and dual k-edge plaque imaging is being explored.