Evaluation of four volume-based image registration algorithms.

We evaluated 4 volume-based automatic image registration algorithms from 2 commercially available treatment planning systems (Philips Syntegra and BrainScan). The algorithms based on cross correlation (CC), local correlation (LC), normalized mutual information (NMI), and BrainScan mutual information (BSMI) were evaluated with: (1) the synthetic computed tomography (CT) images, (2) the CT and magnetic resonance (MR) phantom images, and (3) the CT and MR head image pairs from 12 patients with brain tumors. For the synthetic images, the registration results were compared with known transformation parameters, and all algorithms achieved accuracy of submillimeter in translation and subdegree in rotation. For the phantom images, the registration results were compared with those provided by frame and marker-based manual registration. For the patient images, the results were compared with anatomical landmark-based manual registration to qualitatively determine how the results were close to a clinically acceptable registration. NMI and LC outperformed CC and BSMI, with the sense of being closer to a clinically acceptable result. As for the robustness, NMI and BSMI outperformed CC and LC. A guideline of image registration in our institution was given, and final visual assessment is necessary to guarantee reasonable results.

A feasibility study of Dynamic Phantom scanner for quality assurance of photon beam profiles at various gantry angles.

The effect of gantry rotation on beam profiles of photon and electron beams is an important issue in quality assurance for radiotherapy. To address variations in the profiles of photon and electron beams at different gantry angles, a Dynamic Phantom scanner composed of a 20 x 12 x 6 cm3 scanning Lucite block was designed as a cross-beam-profile scanner. To our knowledge, differences between scanned profiles acquired at different gantry angles with a small size Lucite block and those acquired a full-size (60 x 60 x 50 cm3) water phantom have not been previously investigated. We therefore performed a feasibility study for a first prototype Dynamic Phantom scanner without a gantry attachment mount. Radiation beams from a Varian LINAC 21EX and 2100C were used. Photon beams (6 MV and 18 MV) were shaped by either collimator jaws or a Varian 120 Multileaf (MLC) collimator, and electron beams (6 MeV, 12 MeV, and 20 MeV) were shaped by a treatment cone. To investigate the effect on profiles by using a Lucite block, a quantitative comparison of scanned profiles with the Dynamic Phantom and a full-size water phantom was first performed at a 0 degrees gantry angle for both photon and electron beams. For photon beam profiles defined by jaws at 1.0 cm and 5.0 cm depths of Lucite (i.e., at 1.1 cm and 5.7 cm depth of water), a good agreement (less than 1% variation) inside the field edge was observed between profiles scanned with the Dynamic Phantom and with a water phantom. The use of Lucite in the Dynamic Phantom resulted in reduced penumbra width (about 0.5 mm out of 5 mm to 8mm) and reduced (1% to 2%) scatter dose beyond the field edges for both 6 MV and 18 MV beams, compared with the water phantom scanner. For profiles of the MLC-shaped 6 MV photon beam, a similar agreement was observed. For profiles of electron beams scanned at 2.9 cm depth of Lucite (i.e., at 3.3 cm depth of water), larger disagreements in profiles (3% to 4%) and penumbra width (3 mm to 4 mm out of 12 mm) were observed. Additional profiles with the gantry at 90 degrees and 270 degrees were performed for both MLC- and jaw-shaped photon beams and electron beams to evaluate the effect of gantry rotation. General good agreement is seen (less than 1 % variation) at all field sizes for collimator-shaped 6 MV and 18 MV photon beams. Similar variations observed for MLC-shaped photon beams indicate that the uncertainty in MLC position is similar to that for the collimator jaws. We conclude that the Dynamic Phantom scanner is a useful device for the routine quality assurance on beam profiles of photon beams and for constancy check on electron beams at various gantry angles. Caution should be taken when using this device to acquire basic electron dosimetry data.