Holt_IMRT QA Using an EPID.ppt

Holt EPI Dosimetry ACMP 2005 Orlando IMRT QA using an EPID  ?Randy Holt, Ph.D., DABR North Valley Radiation Oncology Enloe Medical Center, Chico, CA Advantages of EPID 1) No Film Processing 2) Immediate feedback 3) No heavy awkward cassettes 4) Large Fields are a problem – Solution coming? 5) DIGITAL IMAGE PROCESSING 6) Review at any workstation Amorphous Silicon Amorphous Silicon EPID EPID as a physics tool No film to take or store No film scanner to warm up No cumbersome film scanning software Linear Dose response Programmable via RV New analysis tools Can be used for both commissioning and long term QA Can’t completely replace film EPID as a physics tool Flatness/Symmetry Collimator radiation Light Field Radiation Field MLC position tests Does it have a role in IMRT??? EPID relative dose linearity Predicting the EPI Dose Meausrements at cassete are not “fluence” Cassette has inherent scatter. Must measure and model scatter kernel. McCurty, et.al. EPI Dose Kernel Assume convolution holds, start with known f(t). Measure output, h(t) Compute smearing function, g(t) = scatter kernel Direct solution, g(t) = IFT(FT(h(t))/FT(g(t))) Noise sensitive – use iterative de-convolution EPI Dose Kernel Method Errors No tolerance for “delay” error (spatial delay is a simple translation). Cassette must be perfectly centered, and at correct distance. Point source model based, so non-focal output elements will introduce error at gradients. Output not a pure radial function, pure symmetry and axis-variable flatness. Commissioning: dMLC Stress Test dMLC Commish: Fluence Edge vs. Gantry Angle Dose Gradient Analysis TG-23, TG-40, TG-53 How to apply to IMRT? Gamma Analysis Gamma is a Difference-Distance Function with thresholds, e.g. 3mm, 3% Gamma = 1 ? Pass Gamma 1 ? Fail Marginal Gamma Better Gamma EPI-Dosimetry: Some caution dMLC Wedge: Morning Warmup dMLC Wedge Gamma Monthly QA: dMLC H-stress test Current “Standard” for IMRT QA Test each fluence map for agreeme