Sci-Thurs PM: Delivery-02: Improving image quality produced by CCD cameras exposed to stray radiation from a medical linac.

Abstract

Charge coupled devices (CCDs) are increasingly used in radiation therapy. CCDs are ideal for applications such as two-dimensional dosimetry of scintillator sheets or to read arrays of miniature scintillation detectors. However, CCDs are sensitive to stray radiation. Radiation-induced noise strongly alters images and limits their quantitative analysis. We have characterized radiation-induced noise and developed filtration algorithms to restore image quality. Two models of CCD cameras were used for measurements in linac environments. Images were acquired with and without radiation. The structure of the transient noise was characterized. Then, four methods of noise filtration were compared: median filtering of a time series of images, uniform median filtering of single images, an adaptive filter with switching mechanism and a modified version of the adaptive filter. The intensity distribution of noisy pixels was similar in both cameras. However, the spatial distribution of the noise was different: the average noise cluster size was 1.2±0.6 and 3.2±2.7 pixels for each of the two cameras. The median of a time series of image resulted in the best filtration and minimal image distortion. For applications where time series is impractical, adaptive filtration must be used to reduce image distortion. We have characterized the transient noise produced in CCDs by scattered radiation from a linac and have developed an efficient filtration scheme to remove this noise and restore image quality. Use of our filtration scheme allows detailed quantitative analysis of an image even when subjected to scattered radiation.