Abstract

The flat-panel x-ray detectors based on large area integrated circuit called active matrix array ensure excellent image quality and provide wide dynamic range. As a result, active matrix flat panel imagers (AMFPIs) are now commercially available for chest radiography and mammography. However, at low level exposures, most of the AMFPIs are not quantum noise limited due to the electronic noise of the readout circuitry. Therefore, AMFPIs are not fully commercialized for x-ray fluoroscopy (used for interventional procedures and deployment of endovascular devices) which requires maintaining a very low x-ray exposure. An active research is underway to make AMFPIs quantum noise limited at fluoroscopic exposure level. This work investigates the feasibility of avalanche gain as a solution.

At high electric field, avalanche multiplication of charge carrier improves the signal strength to overcome the effects of electronic noise in both direct and indirect conversion x-ray detectors. Indirect conversion detectors are suitable for the avalanche multiplication. It is because the gain fluctuation is minimum since the x-ray absorption and electric charge collection occur in two separate layers. However, in indirect conversion detectors, the image resolution in terms of modulation transfer function (MTF) deteriorates due to depth dependent x-ray absorption (Lubberts effect) and omni-directional propagation of light photons in the phosphor. In this research work, a cascaded linear-system model is proposed to calculate the image quality of CsI-based indirect conversion a-Se avalanche x-ray detectors in terms of spatial frequency dependent detective quantum efficiency (DQE). The depth dependent MTF and noise power spectrum (NPS) are modeled by incorporating the Lubberts effect. The theoretical model also considers MTF due to K-fluorescence reabsorption. The model is then compared with experimentally determined DQE(f) and shows a better fit than previously published models.
On the other hand, direct conversion directors show a better performance in terms of image resolution as the x-ray photons are directly converted into electron and hole pairs. However, the direct conversion detectors are still vulnerable to the electronic noise at low exposures. Utilization of mesh electrode in order to separate the x-ray absorption and gain region has been proposed in the literature to reduce the avalanche gain fluctuation in direct conversion detectors. This work includes a cascaded linear-system model to calculate the DQE(f) of an a-Se based direct conversion avalanche x-ray detector. The proposed model evaluates charge collection efficiency using the Ramo-Shockley theorem and the actual weighting potential of an individual pixel. A 2-Dimensional simulation is performed to calculate the actual weighting potential in the presence of a mesh electrode. The optimal design parameters and operational condition for a-Se based direct conversion multilayer avalanche x-ray detectors are described in this work.

In order to ease the fabrication process and eliminate the need of applying two different voltages to the aforementioned mesh electrode based direct conversion avalanche detector, a novel structure for direct conversion avalanche detector is proposed. The proposed structure contains a hole trapping layer instead of a mesh electrode to separate the absorption layer from the gain region. A numerical model is developed using Semiconductor Module of COMSOL Multiphysics to analyze the device performance. The electric field profile as a function of various device parameters is calculated. A detailed analysis on the transient behavior of the dark current in presence of blocking and trapping layers is performed. A modified cascaded linear-system model that considers the effect of reabsorption of K-fluorescent x-rays, carrier trapping in different layers and avalanche multiplication of charge carrier is used to calculate the DQE(f) and the MTF of the proposed structure. The DQE(f) of the proposed structure is then compared with published experimental results of a commercially available detector at low x-ray exposures (e.g., exposures used in tomosynthesis).

The relative performance of these detector structures influences their clinical effectiveness. Therefore, a comparison of the performance of these detectors for different x-ray imaging modalities is also presented in this research work.