JE
J.R. Edelman
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
2 records found
1
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
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.
...
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.
Battery Free Jogger Light
Lighting and Casing
This document describes the design process and implementation of the lighting and casing of a battery free jogger light. This light is meant to increase the safety of joggers in dark environments. The most effective way of increasing the jogger's conspicuity will be researched by considering different light sources and driver circuits to efficiently power the light source in a blinking manner. A casing will be designed to encapsulate the components. Light emitting diodes were chosen as a light source due to their energy efficiency, colour optimisation and small size. Two LEDs are part of a rectifier, while three other LEDs are powered by a driver circuit that uses a clocked decade counter 4017 IC that receives its clock signal from a NAND based oscillator circuit. The casing is designed with 3D modelling software and a prototype is 3D printed. The intended light intensity was not reached, but the brightness in dark environments was deemed suitable for the project's goals. The casing has bigger dimensions than intended; however these can be optimised for mass production.
...
This document describes the design process and implementation of the lighting and casing of a battery free jogger light. This light is meant to increase the safety of joggers in dark environments. The most effective way of increasing the jogger's conspicuity will be researched by considering different light sources and driver circuits to efficiently power the light source in a blinking manner. A casing will be designed to encapsulate the components. Light emitting diodes were chosen as a light source due to their energy efficiency, colour optimisation and small size. Two LEDs are part of a rectifier, while three other LEDs are powered by a driver circuit that uses a clocked decade counter 4017 IC that receives its clock signal from a NAND based oscillator circuit. The casing is designed with 3D modelling software and a prototype is 3D printed. The intended light intensity was not reached, but the brightness in dark environments was deemed suitable for the project's goals. The casing has bigger dimensions than intended; however these can be optimised for mass production.