Date of Award

Fall 2019

Document Type

Dissertation - Restricted

Degree Name

Doctor of Philosophy (PhD)

Department

Biomedical Engineering

First Advisor

Schmidt, Taly G.

Second Advisor

Clough, Anne

Third Advisor

Sidky, Emil

Abstract

Energy-resolved photon counting detectors have potential improvements overconventional scintillating x-ray detectors and can be used for spectral computed tomog-raphy (CT) applications such as material decomposition. However, photon counting de-tector are difficult to model because they have many non-ideal physical phenomena. Wedeveloped machine learning algorithms to model the spectral measurements of energy-resolved photon counting detectors for the purposes of material decomposition and imagereconstruction.A feed-forward neural network was used to estimate path lengths of different ma-terials that the x-ray path traversed from spectral measurements of a photon counting de-tector using calibration measurements on an experimental bench-top CT system. The neu-ral network relied only on calibration measurements and no prior knowledge of the sourcespectrum, material attenuation coefficients, or detector response. The neural networksdecomposed each ray in the measured sinograms into material separated sinograms andreconstructed into basis material images. The capabilities of the machine learning algo-rithm were evaluated experimentally with a rod phantom with calculated biases of 0.3%to 7.6% compared to biases of 1.3% to 16% of a previously proposed empirical decompo-sition algorithm.A special case of material decomposition occurs when materials with high atomicnumbers are in the measured ray path. These materials have K-absorption edges in theirlinear attenuation coefficients in the relevant x-ray energy range. In a separate study, theneural network mode was extended for K-edge imaging to separate contrast agents andnanoparticles, such as iodine and gadolinium, from other materials in the image. Transferlearning was used to further train the neural networks between detector pixels and fromsimulated calibration data to improve the qualitative and quantitative image quality in arod phantom and an ex-vivo rat leg specimen.Similarly, a neural network was developed to model the forward spectral measure-ment from basis material parameters of the object. The forward model was able to fit thecalibration measurements to within 6% of all measurement and was used with an opti-mization algorithm for material decomposition. Such a model could be used for iterativereconstruction algorithms and iterative material decomposition algorithms to optimize theestimates using the noise properties of the measurement.

Share

COinS

Restricted Access Item

Having trouble?