The research project proposes a scientific, preclinical and clinical evaluation of an experimental digital radiography system incorporating a laboratory prototype Kinestatic Charge Detector (KCD) used with an x-ray scanning strip beam geometry. The major impact on diagnostic health care is expected to result from the system's simultaneously high spatial resolution (> 5 cy/mm) and high contrast resolution (approximately twice that of film- screen radiography) as well as its immediately available digital images. During the firse two years of the project, a small field of view detector producing 4.3 cm x 43 cm images will be available, and a large (43 cm x 43 cm) field detector is expected to be used during the last three years. A comprehensive evaluation of the imaging parameters will be made on both small and large field of view systems. Parameters to be studied include spatial resolution, contrast resolution (detective quantum efficiency, scatter acceptance, beam spectral effects) and radiation dose, temporal resolution, artifacts and other parameters such as dynamic range, uniformity and potential for further field enlargement. The methodology includes Monte Carlo simulations and other theoretical modeling as well as phantom experiments to measure spatial resolution, contrast resolution, scatter rejection and other parameters. To quantify the diagnostic capability of the KCD compared with film-screen radiography and digital fluorography, an observer study designed for lesion detection will be conducted. Images will be obtained from the three modalities at different exposure rates using phantoms consisting of simulated lesions with differnt sizes and contrasts which are embedded in scattering media with different thicknesses. The results from the observer study will be compared with those obtained from a theoretical model. Comparative data analysis will provide detailed information on the contrast- detail-dose characteristics of the KCD in lesion detection compared with the two conventional radiography techniques. To compare the KCD system quantitatively with film-screen radiography and digital fluorography, three anatomic regions: chest, kidney, and lumbar spine have been selected as representative of the areas in which projection radiography is expected to remain a preferred diagnostic tool. Anthropomorphic phantoms will be used to simulate clinical imaging yet retain quantitative experimental control. Images obtained from the phantoms will either contain one or no abnormality. The observer task is the detection of the presence or absence of an abnormality with a confidence rating. The data from the KCD and radiography/fluorography observer studies will be analyzed by ROC techniques. The principal clinical studies are scheduled for years 3-5 pending satisfactory results from phantom studies. Using a large field detector, a limited number of normal anatomic structures in the chest, abdomen, and extremities will be characterized. Subsequently, formal observer studies will compare KCD imaging with film-screen in disease states, including interstitial lung disease, pulmonary nodules, mediastinal masses, lumbar spine abnormalities, extremity fractures and renal calculus disease. ROC analysis will be performed.
| Wagenaar, D J; DiBianca, F A; Tenney, C R et al. (1991) Space charge effects in a kinestatic charge detector. Phys Med Biol 36:61-76 |
