Characterization of a high-elbow, fluoroscopic electronic portal imaging device for portal dosimetry

• Published in: Physics in medicine & biology Vol. 45; no. 1; pp. 197 - 216
• Main Authors: Boer, J C J de, Heijmen, B J M, Pasma, K L, Visser, A G
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.01.2000; Institute of Physics
• Subjects: Algorithms; Applied radiobiology (equipment, dosimetry...); Biological and medical sciences; Biological effects of radiation; Fluoroscopy - instrumentation; Fluoroscopy - methods; Fundamental and applied biological sciences. Psychology; Image Processing, Computer-Assisted; Phantoms, Imaging; Radiometry; Software; Tissues, organs and organisms biophysics; Performance evaluation; CCD camera; Radiobiology; Characteristic; Radioscopy; Instrumentation; Ionization chamber; Dosimetry; Radiotherapy; Portal imaging; Comparative study
• DOI: 10.1088/0031-9155/45/1/314
• ISSN: 0031-9155

The application of a newly developed fluoroscopic (CCD-camera based) electronic portal imaging device (EPID) in portal dosimetry is investigated. A description of the EPID response to dose is presented in terms of stability, linearity and optical cross-talk inside the mechanical structure. The EPID has a relatively large distance (41 cm on-axis) between the fluorescent screen and the mirror (high-elbow), which results in cross-talk with properties quite different from that of the low-elbow fluoroscopic EPIDs that have been studied in the literature. In contrast with low elbow systems, the maximum cross-talk is observed for points of the fluorescent screen that have the largest distance to the mirror, which is explained from the geometry of the system. An algorithm to convert the images of the EPID into portal dose images (PDIs) is presented. The correction applied for cross-talk is a position dependent additive operation on the EPID image pixel values, with a magnitude that depends on a calculated effective field width. Deconvolution with a point spread function, as applied for low-elbow systems, is not required. For a 25 MV beam, EPID PDIs and ionization chamber measurements in the EPID detector plane were obtained behind an anthropomorphic phantom and a homogeneous absorber for various field shapes. The difference in absolute dose between the EPID and ionization chamber measurements, averaged over the four test fields presented in this paper, was 0.1 +/- 0.5% (1 SD) over the entire irradiation field, with no deviation larger than 2%.