Bulletin of the American Physical Society
APS March Meeting 2022
Volume 67, Number 3
Monday–Friday, March 14–18, 2022; Chicago
Session B43: Inverse Problems: From Biomedicine to MaterialsInvited Live Streamed Undergrad Friendly
|
Hide Abstracts |
Sponsoring Units: GMED DCOMP Chair: Wojciech Zbijewski, Johns Hopkins University Room: McCormick Place W-375B |
Monday, March 14, 2022 11:30AM - 12:06PM |
B43.00001: Introduction to Inverse Problems with Applications to Magnetic Resonance Relaxometry and Myelin Mapping in the Brain Invited Speaker: Richard G Spencer The success of conventional magnetic resonance imaging (MRI) is partly attributable to the fact that it is a Fourier technique. Data is collected in the space reciprocal to spatial coordinates, known as k-space, and then (inverse) Fourier transformed to produce an image. This reconstruction has the very attractive property of being mathematically well-conditioned, with condition number of unity, so that noise in the acquired data is necessarily transmitted to the image domain but is not magnified. Because of this, early studies performed at low magnetic field and with relatively unsophisticated radio frequency technology were able to yield useful images. Correspondingly, the quality of conventional MRI has increased roughly in proportion to improvements in acquisition SNR. However, the situation with the newer technique of MR relaxometry is not so rosy; the reconstruction of acquired data to obtain the distribution function of the parameter of interest is via a version of the inverse Laplace transform. This arises from the classically ill-posed problem of solving the Fredholm equation of the first kind. One implication is that noise is amplified in the reconstruction process. A corollary of this is that brute force efforts to improve SNR rapidly reach the point of diminishing returns, and other means must be undertaken to produce useful results. For this, the inverse problems perspective has proven to be enormously fruitful. We will discuss the basics of inverse problem theory and some developments applicable to MR relaxometry and related experiments. Our main application is to myelin mapping in the brain, and we will show how more accurate myelin quantification permits physiological correlations to be established. Our studies have the twofold goal of improving the capacity of MR to diagnose pathology and monitor disease progression, and of developing methods of general use for inverse problems. |
Monday, March 14, 2022 12:06PM - 12:42PM |
B43.00002: X-ray tomography with sparse data: physics, mathematics, and application Invited Speaker: Emil Sidky Due to advances in large-area flat-panel X-ray digital detectors, applications for X-ray tomography have diversified to a wide range of applications in medical imaging such as image-guided radiation therapy and surgery, and devices such as extremity scanners, dental CT, and digital breast tomosynthesis. An important part of all of these specialized scanners, is the development of new image reconstruction algorithms that result from the study of inverse problems in X-ray tomography. Of particular interest for this talk are sparse-data inverse problems where the number of image voxels is larger than the number of measurements in the X-ray projection data. This presentation will explain how prior information on the scanned subject can be used to arrive at accurate solution to the inverse problem associated with under-sampled image reconstruction. The presentation will mainly rely on images and examples with few equations. After explaining the theory, application will be presented in the domain of tomographic mammography imaging, exploiting under-sampled image reconstruction for novel scan configurations and X-ray dose and scan-time reduction. |
Monday, March 14, 2022 12:42PM - 1:18PM |
B43.00003: Inverse problems in radiation cancer therapy Invited Speaker: Kamil M Yenice One of the most remarkable realizations in the field of inverse problems was the invention of an inversion algorithm for Computed Tomography (CT) and its experimental demonstration in CT imaging. After five decades of its invention, CT has become the primary imaging modality for radiation therapy dose calculations and image guidance for accurate dose localization in patients for the last two decades. After the invention of CT imaging, another inverse (optimization) method, namely, intensity modulated radiation therapy (IMRT) and in its more recent version- volumetric arc therapy (VMAT) have arguably made the most profound impact in the field of radiation therapy in terms of the ability to exquisitely shape dose distributions delivered to patients for treatment of various diseases. In IMRT, radiation beams are divided into many small beamlets, whose intensities are individually adjusted in order to deliver a dose of radiation that conforms to the shape of the tumor. Having the availability of the physical and biological parameters of the irradiated object (patient) and the specifications of the treatment machine, finding the radiation intensity function (individual beamlets) from a desired dose distribution prescribed by the physician is the inverse problem in IMRT. In clinical applications, robust decision making about the optimality of the solution, deliverability, and organ motion are some of the factors that confound the problem. This presentation will provide an overview of the inverse planning process in radiation therapy and the current state of its relevant issues prom the perspective of a practicing clinical physicist. |
Monday, March 14, 2022 1:18PM - 1:54PM |
B43.00004: Materials physics from microscopy: statistical and machine learning methods for tackling inverse problems Invited Speaker: Rama K Vasudevan Materials imaging has undergone a dramatic transformation in the previous two decades, caused by the proliferation of new modalities in scanning probe microscopy that enable better functional property measurements at high resolution, along with advances in electron microscopy, that now make atomic-scale imaging nearly routine for a wide variety of material systems. This explosion of new imaging data brings forth challenges related to data compression, analysis, and ultimately, physics knowledge extraction. |
Monday, March 14, 2022 1:54PM - 2:30PM |
B43.00005: Reinforcement Learning for Inverse Materials Design Invited Speaker: Subramanian Sankaranarayanan The most common and popular method for structure search and optimization are based on evolutionary design. This can often be cumbersome, limited to few tens of parameters and fails for large structural configurations or design problems with high degrees of freedom. Reinforcement learning approaches mostly operate in discrete action space such as in Go game but the applications of that to inverse problems is limited since most inverse problems deal with continuous action space. There are a large number of inverse structural search problems ranging from crystal structure search in material sciences to topology design in Quantum information, where it is highly desirable to optimize structure/configuration to target desired properties or functionalities. This talk will provide an overview of our current efforts to perform scalable crystal structure and topology search to discover and design metastable or non-equilibrium phases with desired functionality. We will also discuss our efforts on fingerprinting and use of unsupervised learning to identify crystal structures and critical nuclei from amorphous melts, using zeolites as a representative example. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700