A method has been devised which enables the modulation transfer function (MTF) of computerized tomographic (CT) scanners to be simply and rapidly determined. The method relies upon the measurement of the standard deviation of the pixel values within the image of cyclic bar patterns. The method is shown to produce results which are indistinguishable from those derived by less practical conventional methods.

A technique to measure the MTF of metal screen-film detectors at megavoltage x-ray energies has been devised. It employs an exposure through a slit 25 micron wide and a method to average numerous line spread functions from a single film image. That the slit is wide compared to those used at diagnostic x-ray energies serves two purposes: it facilitates the use of a laser alignment procedure designed to ensure correct orientation of the slit-forming tungsten blocks in the x-ray beam, and it also increases the contrast in the slit image, thereby reducing the data analysis problems caused by film noise. Further reduction in the effects of film noise is accomplished by the averaging method. A correction is made to account for the effect of the wide slit on the MTF. The validity of this correction, the laser alignment, and the averaging method have been confirmed by simulations.

The radiographic contrast of metal screen-film detectors was investigated in order to determine the contrast capabilities of these detectors applied to megavoltage x-ray imaging. The film contrast gamma was found to be independent of the metal screen composition. Measurement of the scatter-to-primary film dose ratio in contact geometry demonstrated that a thick front screen of either 1.5 g/cm2 copper of 2.5 g/cm2 lead provides optimum contrast for the photon energies studied (60Co and 4- and 8-MV x rays). The same thickness were also found to be suitable in an air gap geometry which significantly improved the contrast compared to the contact geometry. Rear lead screens were found to provide no contrast improvement.

The MTF of metal screen film detectors used in radiation treatment verification has been measured at 4 and 8 MV x-ray energies. The results show that lead screens provide better resolution than copper screens, and a single-emulsion film offers considerable advantage over the traditional double-emulsion film. A rear lead screen was found to seriously degrade the resolution properties of a front lead screen single-emulsion film detector. The detector MTF was found to be energy dependent. In general, both the low and the high spatial frequency response decreased with increasing x-ray energy. This, in part, accounts for the noticeable image quality difference between 4 and 8 MV radiographs.

A quality assurance (QA) test has been developed as a practical means to routinely monitor digital image resolution. The method is based upon a parameter termed the image modulation M. When noise is present M cannot be directly measured. Instead, a biased estimate of M, designated M', is measured. This report describes the application of the parameter M' to QA monitoring of digital image resolution. A measurement of M' is generally simpler than alternative measurements, and is shown to outperform visual evaluation in the clinical QA environment.