Not available / text

Thesis (Ph.D.)--Rochester Institute of Technology, 2009. / Typescript. Includes bibliographical references (leaves 173-179).

Title from PDF of title page (University of Missouri--Columbia, viewed on Feb 26, 2010). The entire thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file; a non-technical public abstract appears in the public.pdf file. Dissertation advisor: Dr. Ping Yu. Vita. Includes bibliographical references.

Thesis (Ph. D.)--University of Chicago, Dept. of Radiology, August 2000. / Includes bibliographical references. Also available on the Internet.

Thesis: Sc. D., Massachusetts Institute of Technology, Department of Nuclear Engineering, 1980 / Bibliography: leaves 348-372. / by Kimberlee Jane Kearfott. / Sc. D. / Sc. D. Massachusetts Institute of Technology, Department of Nuclear Engineering

High resolution imaging devices have made a digital medical archiving system feasible. The large volumes of information generated must be stored and retrieved at high data rates in order to insure the timely diagnosis of patients. This creates some unique technological challenges that must be resolved, including the problem dealing with multiple vendor products interacting in one environment. The high speed interface card design presented in this thesis is able to deal with different computer host busses as well as different interprocessor communication protocols. The ACR-NEMA standard has been implemented in the design as one possible network protocol that provides a solution that can be easily adapted to different vendors. The design has been analyzed using the Network II.5 simulation language. The simulation was performed to insure that the original objectives are met and to determine the impact on the protocols rated throughput.

The goal of digital image noise smoothing is to smooth noise in the image without smoothing edges and other high frequency information. Statistically optimal methods must use accurate statistical models of the image and noise. Subjective methods must also characterize the image. Two methods using high frequency information to augment existing noise smoothing methods are investigated: two component model (TCM) smoothing and second derivative enhancement (SDE) smoothing. TCM smoothing applies an optimal noise smoothing filter to a high frequency residual, extracted from the noisy image using a two component source model. The lower variance and increased stationarity of the residual compared to the original image increases this filters effectiveness. SDE smoothing enhances the edges of the low pass filtered noisy image with the second derivative, extracted from the noisy image. Both methods are shown to perform better than the methods they augment, through objective (statistical) and subjective (visual) comparisons.

No description available.

The EPID (electronic portal imaging device) was designed for in vivo imaging of patients during radiotherapy treatment. The ability of EPIDs to promptly acquire two dimensional data, lends them to be considered for use in quality assurance of the linac. This thesis set out to investigate the possibility of using a radionuclide, technetium 99 m (Tc99m), to produce a flood field for the calibration of an EPID, because using a beam calibrated EPID to measure the beam is self-referential. The difference in relative response between the energy spectrum of a 6MV beam and the Tc99m was investigated using EGSNRC DoseXYZ Monte Carlo Modelling. The relative output ratio was calculated to be less than 1.6%. The dose response of the EPID with respect to dose rate was checked using different activities of Tc99m and found to be linear. The flatness from a phantom was calculated, with a model in MATLAB, for a range of heights, overlaps, thickness, and deformations, to find the optimum balances between signal strength and flatness. This model was checked for accuracy using diagnostic radiographic film. The culmination of the energy response, linearity and the calculated flatness is a flood field taken with a flood phantom on the EPID with low signal strength. To get a signal to noise ratio of 3% the mean of over 2000 flood field images were used. This accuracy was not adequate for clinical use but the averaging of pixels it is accurate enough for QA.

It is a common opinion in the image processing field that a software system cannot be both portable and efficient. Furthermore, when efficiency prevails in the design of an image processing system, the system typically becomes hard to use and complicated to program in. Most efficient systems require the user to worry about external data format, buffer allocation, overlapped input and output and similar details. To show that it is feasible to design an image processing system that is efficient, portable and easy to use, a system has been designed and implemented using the "Software Tools" philosophy. This system has been implemented on two computer systems (the PDP-11 and the VAX-11) under three different operating systems (Unix, RSX-11M and VMS). The image operations have been implemented in Ratfor with the bulk of the image i/o routines written in C. Details such as double-buffered overlapped i/o, conversion of external data formats and virtual memory support (on the VAX-11) are handled automatically by the system. Additionally, the system allows one to perform image operations "on the fly" under Unix, allowing one to use software pipelines for image processing. It is shown that this system is portable and efficient. The system is analyzed to determine where the time is spent during typical image operations. This analysis shows that the features that make the system easy to use account for a small fraction of its time. The performance of the system is also measured under different operating systems.