DOS 542 - Weeks 1&2 Discussion
Initial Post: QA of Linear Accelerators
The morning startup QA process for the linacs at Loyola University Medical Center tests a range of important machine parameters, including mechanical condition, dosimetric accuracy, and functionality of safety systems. After beginning the powerup process for the linac and control computers, the therapist checks several gauges inside the linac's housing. The first check is a water level reading, where a floating indicator must fall between two marks on a tube. If the water level is low, the therapist can add water with a funnel and water jug. The next check in the water system pressure, which should be approximately 80 psi, and the water temperature, which should be around 40 F.1 The next check is the gas system pressure, which should be around 32 psi. If the water system is out of spec, the physicist must be called and then environmental services needs to be contacted, as this could indicate a problem with the facility's chiller system. If the gas pressure is too low, a physicist can add more gas from a tank.
Once the basic machinery has been checked, the optical distance indicator (ODI) is checked by placing a small device on the table and raising or lowering the table to align the room lasers to the top edge of the device. With the lasers on the top edge, the ODI should read 100 cm SSD, with a tolerance of 1mm.2 At this point, the device is replaced with a Daily QA3 device from Sun Nuclear. This device has physical markings on its surface to check the size of a 20x20 cm field at 100 cm SSD, as well as a series of diodes and ion chambers to measure beam flatness and symmetry and beam edge agreement with the projected light field. A 20x20 cm electron cone is placed on the linac head and the therapist runs 200 monitor units (MU) on each energy available on the machine. Each run is recorded by the Daily QA3 device and the results are automatically recorded and compared against an acceptable range of values, which is +/-2%, though TG-142 only calls for +/-3%.3 These test are repeated for every available energy of photons once the electron cone has been removed. If the field size check (tolerance 2mm) or any dosimetric parameter of the beam such as output or symmetry fails the check, the test can be repeated. If the test fails again, that beam can not be used for treatment until it has been cleared by a physicist. The next test is a test of a dynamic wedge set at 30 degrees, to verify correct multileaf collimator motion.
The wedge test can be interrupted by the first of several safety checks. If the vault door is opened while the wedge check is in progress, the door interlock should cause the beam to immediately turn off. Other safety system checks include checking the functionality of the "BEAM ON" lighted sign over the treatment room door, and the functionality of the closed circuit TV monitors and audio intercom. The door itself is a heavy sliding door design, so there are pressure sensors on its edge to prevent it from closing on a person or other obstruction. These pressure sensors need to be tested. If power is lost, there is a manual crank that can open the door, but it is installed in a panel above a false ceiling, so a ladder must be present in the linac's power supply closet.
Once these tests are done, the onboard imaging is tested in conjunction with the room lasers and couch motion system. A small box with two offset marked locations on the surface and radiopaque internal markers is placed on the table and the room's lasers are aligned to one of the sets of surface marks. Three of Loyola's 4 linacs have megavolt imaging (two 21 EX machines and a TrueBeam) so each of them tests the megavolt imagers by capturing anterior/posterior (AP) and lateral images, which reveal the radiopaque internal markers. Both the observed size of the box and the locations of the markers are compared against a template, and shifts to the second marked location are measured in the onboard imaging software, with a tolerance of 1.5 mm.2 The shifts are applied to the table either manually (on one of the 21 EXs) or by computer control (on the other two machines). The therapist then checks to make sure that the lasers are now aligned to the second set of surface markings, with a tolerance of 1 mm. This is a test of both the laser system and the onboard imaging system.
One of Loyola's 21 EX machines has been upgraded with kilovolt imaging and cone beam CT capability, and these functions are standard equipment on our TrueBeam. To test the KV imaging systems, the same phantom is once again scanned with an AP and lateral pair using KV or a cone beam CT, and the shifts are once again measured, applied, and verified by checking the laser position on the surface of the phantom. This tests not only the imaging panels, but the software that runs them. These machines also have an "X-Ray In Use" lighted sign that must come on while the KV systems are operating.
Loyola's 4th linac is a Novalis stereotactic radiosurgery system. It uses the ExacTrac X-Ray monitoring system made by BrainLab, which incorporates two fixed superior-posterior oblique X-Ray tubes and corresponding imager panels installed on the ceiling of the vault. There is also an infrared tracking system that scans the treatment area for reflective spheres that are attached to positioning frames or QA devices. Two different phantom devices are placed on the table; one is used to calibrate and then test the isocenter location. It must be set up perfectly with the lasers, because the position of its reflective spheres during calibration determines the software's understanding of the isocenter location for all treatments that day. Once isocenter is set, the phantom is moved, and the ExacTrac system uses the sphere tracking cameras to shift the couch so that the phantom is once again centered at the isocenter. The isocenter accuracy is also checked with a special attachment that slots into the SRS frame. All lasers must perfectly align on its surface markings, with a tolerance of 1mm. Finally, another phantom with internatl radiopaque markers is placed on the table, and the oblique X-Ray panels are used to determine shifts to place the phantom correctly at isocenter.
Many of these procedures are inspired by the TG-142 report from AAPM, but they have been adapted to the capabilities and requirements of each of our different machines.
- Rusu I, Roeske J. Loyola University Medical Center TueBeam warmup procedure. [Department Memorandum]. 2015
- Roeske J, Diak A. Loyola University Medical Center monthly linac quality assurance procedures. [Department Memorandum]. 2014.
- Klein EE, Hanley J, Bayouth J, et al. Task Group 142 report: quality assurance of medical accelerators. Med Phys. 2009;36(9):4197-4212. http://dx.doi.org/10.1118/1.3190392