An EMCCD-array based high-resolution imaging system for fluoroscopy and angiography
The increasing demand for high-resolution, high-quality medical imaging systems in the field of endovascular image guided interventions (EIGI) serves as the motivation for this work that is focused on the design and implementation of an x-ray imager that uses an Electron Multiplying CCDs (EMCCDs) array. The large variable electronic-gain (up to 2000) and small pixel size of EMCCDs provide effective suppression of readout noise compared to signal, as well as high resolution, enabling the development of an x-ray detector with far superior performance compared to conventional x-ray image intensifiers or flat panel detectors. The modular array design was implemented to enable the x-ray detector to cover a full clinical field-of-view (FOV). This FOV was enabled by using fiber optic tapers (FOTs) that make the effective pixel size and FOV three times larger than the inherent EMCCD pixel size and FOV. The newly designed medical imaging system can enhance fluoroscopic and angiographic image quality by improving both the spatial resolution and dynamic range. This system consists of a detector front end and a graphic user interface (GUI). The system can be selected to run at up to 17.5 frames per second or even higher frame rates with different binning schemes. The integration time for the sensors can be adjusted from 1 ms to 1000 ms to match the output pulses of commercial x-ray machines. Twelve-bit correlated double sampling analog to digital converters were used to digitize the images—which were acquired by a National Instruments dual-channel CameraLink PC board in real time. A user-friendly interface was programmed using LabVIEW to send commands, and acquire and to display 2K × 1K pixel matrix digital images for the 2 × 1 array. Image processing techniques were also programmed in LabVIEW to perform image geometric corrections which were introduced by the inevitable mechanical misalignment between the two sensors, and to complete the automatic brightness matching for the array image that was acquired by two identical sensors but that do not have the same gain value due to the slightly different analog circuitry driving the sensors. A unique automatic gain control algorithm was implemented for the EMCCD sensors. This algorithm allowed the imaging system to maintain the brightness of an arbitrarily-shaped operator-selected region-of-interest while keeping the x-ray imaging tube parameters unchanged. The developed x-ray imaging system was evaluated in terms of the modulation transfer function (MTF), detective quantum efficiency (DQE), and noise power spectrum (NPS). In this dissertation, the x-ray detector basics will be introduced in the first chapter, followed by the design theory and realization of the new EMCCD-based detector in chapter two. The measurement of the new detector system's optical and x-ray performance will be presented and discussed in chapter three. In chapter four, image processing techniques will be explored to make this EMCCD-based array imager more suitable for the clinical applications of interest. Finally, chapter five summarizes the results and conclusions that drawn from the work presented in this dissertation and presents a discussion of the future work.