This proposal addresses the marriage of two novel technologies, with the overall objective to develop a practical single- and dual-energy digital radiography system, dedicated and optimized for chest imaging. A filter wheel equalization system compensates for over- and under-penetrated areas in the image, by reducing or increasing the entrance exposure to those areas. A series of computer controlled wheels are mounted near the x-ray tube, and the filter wheels have organ-specific compensation patterns, determined by analysis of numerous chest images, machined into them. A low-dose prescan image is acquired with only homogeneous-filtration present, and based on analysis of the prescan image the computer automatically rotates the most appropriate filter wheel patterns into position to achieve compensation and the radiographic exposure is acquired. The heart, left lung and right lung are the principle thoracic features which demand enormous detector latitude; independent compensation of these structures will reduce the exposure dynamic range and will yield better uniformity in the signal-to-noise performance of the system. This will improve the detectability of pulmo- nary nodules since detectability is fundamentally related to the signal to noise ratio. The proposed detector is a novel dual phosphor (binary) screen system, composed of two phosphors differing in energy-dependent x-ray absorption (different k-edges), but also differing in their optical emission properties. One phosphor (Gd2O2S:Eu, k-edge = 50 keV) emits principally at 624 nm, and the other (y2O3:Gd, k-edge = 17 keV) emits at 315 nm. Using optical filtration, dual CCD cameras are individually tuned to the optical emission spectrum of each phosphor, thereby allowing the low energy and high energy images to be acquired simultaneously during one x-ray pulse. The marriage of the filter wheel equalization system with the binary screen detector will result in a synergistic and comprehensive approach to chest radiography. The research methodology includes the acquisition of a large data base of chest images (digitized from film) to facilitate a statistical approach to the compensation pattern design and analysis, computer simulation and design optimization of both the compensation and detector subsystems, construction and testing of a prototype system including image performance characterization (MTF & DQE), and ROC analysis of experimentally acquired images. The comprehensive approach to digital chest imaging proposed here should increase the effectiveness of chest radiography, by increasing the sensitivity and specificity to pulmonary nodules and other chest pathologies.
