Comparative evaluation of image quality from three CT simulation scanners.

Abstract

Today, radiation therapy (RT) is moving toward increased radiation dose to the tumor as a result of 3D conformal RT (3DCRT) and intensity-modulated RT (IMRT), which have been made possible by advances in volumetric-based image planning with digital imaging systems such as computed tomography (CT). Treatment planning for such RT requires superior CT image quality. Our goal in this study was to evaluate and to compare the image quality of three unique CT simulation scanners available at our center for both single- and multiple-slice helical scanners. These scanners included a conventional 70-cm bore single-slice scanner (Philips Medical Systems), a large 85-cm bore single-slice scanner (Philips Medical Systems), and a 70-cm bore multislice scanner (GE Medical Systems). Image quality was evaluated in terms of image noise, low-contrast detectability (LCD), limiting spatial resolution (modulation transfer function), and slice thickness accuracy in accordance with guidelines set out by the AAPM. A commercially available Catphan phantom was used to characterize image quality for both axial and helical modes of scanning. We found that image quality was generally comparable for all scanners. Limiting spatial resolution and slice thickness accuracy were comparable for all three scanners for both scanning modes. The multislice unit was superior in terms of noise content, resulting in improved visualization of small, low-contrast objects, which is of significant clinical importance, particularly for soft tissue delineation. In addition, the multislice unit optimizes volume coverage speed and longitudinal resolution without compromising image quality, a significant advantage for the radiation oncology environment.