Print Email Facebook Twitter Deep Learning Enhanced Contrast Source Inversion And Phase Error Based Conductivity Correction For Electrical Properties Tomography Title Deep Learning Enhanced Contrast Source Inversion And Phase Error Based Conductivity Correction For Electrical Properties Tomography: Exploration Of Physics-Assisted Deep Learning Methodology For Magnetic Resonance Based Electrical Properties Tomography Author Edelman, Jory (TU Delft Electrical Engineering, Mathematics and Computer Science) Contributor Remis, R.F. (mentor) Mandija, S. (graduation committee) Degree granting institution Delft University of Technology Programme Electrical Engineering | Circuits and Systems Date 2023-12-05 Abstract Magnetic resonance electrical properties tomography is a type of quantitative magnetic resonance imaging that aims to reconstruct the conductivity and permittivity of biological tissue. These electrical properties of the tissue can be used to compute the specific absorption rate, to differentiate tumours from healthy tissue and for hyperthermia treatment planning. Several methods to reconstruct these electrical properties exist with different degrees of success. Combining analytical reconstruction methods with deep learning methods is left relatively unexplored in the field of magnetic resonance electrical properties tomography. Hence, this work explores such hybrid methods in which deep learning is embedded in an analytical reconstruction method. A recurrent inference machine is integrated into the iterative reconstruction scheme called Contrast Source Inversion, in an attempt to decrease its high computational load. Additionally, a U-net is trained to correct reconstructed conductivity maps using discrepancies in measured- and reconstructed phase data, which is based on the relation between conductivity and phase in the Helmholtz equation. The recurrent inference machine embedded version of contrast source inversion failed to achieve a desirable reconstruction quality with its current implementation. However, the large amount of potential improvements to its implementation motivates further research into its application before discarding it. The conductivity correction U-net is able to correct conductivity errors as small as 0.13 S/m when used iteratively or 0.05 S/m when used a single time when noiseless data is used. Further research in its capabilities of handling noisy data is required to assess practical usage. Subject EPTElectrical Property TomographyCSIcontrast source inversionDeep LearningRecurrent inference machineU-NetMRI To reference this document use: http://resolver.tudelft.nl/uuid:8a829af8-88a0-41a0-8bee-c7d529685ece Part of collection Student theses Document type master thesis Rights © 2023 Jory Edelman Files PDF ThesisJoryEdelman.pdf 3.5 MB Close viewer /islandora/object/uuid:8a829af8-88a0-41a0-8bee-c7d529685ece/datastream/OBJ/view