Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


Dissertation - Mingdong Fan.pdf (8.16 MB)

ETD Abstract Container

Abstract Header

THREE INITIATIVES ADDRESSING MRI PROBLEMS

Fan, Mingdong
http://rave.ohiolink.edu/etdc/view?acc_num=case1585863940821908

Abstract Details

2020, Doctor of Philosophy, Case Western Reserve University, Physics.
MRI is one of the most commonly used medical imaging techniques and the quest for higher image quality and reduced scan time has been driving various developments in this modality. In the thesis, three advances in MRI hardware and software are presented. Magnetic resonance fingerprinting (MRF) is a new imaging technique that allows quantification of intrinsic tissue parameters. Two deep learning models are built to achieve fast and accurate reconstruction for spatial-temporal MRF signals. The first model performs image de-aliasing before the traditional MRF reconstruction. The second model provides an end-to-end MRF reconstruction without any additional processing. The end-to-end model returns the highest reconstruction accuracy while showing a great advantage in time efficiency in both signal acquisition time and reconstruction time. A shim coil is designed to cancel the susceptibility-induced field inhomogeneities at the back of the knee. The current density of the shim coil is inversely calculated using a target-field approach. The resulting field inhomogeneities can be reduced by more than 50% in the region of interest with a finite-length cylindrical coil and a current consumption of only 242 mA. As an attempt to replace the coaxial cabling in the MRI interconnect, a digital fiber-optic transmission system using delta-sigma modulation (DSM) is presented. Initial bench tests are conducted with a commercial DSM. System linearity is measured and a dynamic range of approximately 70 dB is observed, which is only about 10 dB away from the requirement of full-bandwidth (±500 kHz) MR signal transmission. The system is expected to provide sufficient dynamic range for practical MR signal transmission using a custom DSM chip currently being developed in our collaboration.
Robert Brown (Advisor)
Michael Martens (Advisor)
Soumyajit Mandal (Committee Member)
Gary Chottiner (Committee Member)
129 p.
Artificial Intelligence; Biomedical Engineering; Electrical Engineering; Medical Imaging; Physics
MRI; MRF; deep learning; U-Net; image de-aliasing; image reconstruction; B0; B0 shimming; knee shimming; target field; stream function; RF-over-fiber; fiber-optic; signal transmission; delta-sigma modulation; DSM; dynamic range;

Recommended Citations

Citations

  • Fan, M. (2020). THREE INITIATIVES ADDRESSING MRI PROBLEMS [Doctoral dissertation, Case Western Reserve University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=case1585863940821908

    APA Style (7th edition)

  • Fan, Mingdong. THREE INITIATIVES ADDRESSING MRI PROBLEMS. 2020. Case Western Reserve University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=case1585863940821908.

    MLA Style (8th edition)

  • Fan, Mingdong. "THREE INITIATIVES ADDRESSING MRI PROBLEMS." Doctoral dissertation, Case Western Reserve University, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=case1585863940821908

    Chicago Manual of Style (17th edition)

case1585863940821908
359
© 2020, all rights reserved.
This open access ETD is published by Case Western Reserve University School of Graduate Studies and OhioLINK.