Click here to

Session: Therapy General ePoster Viewing [Return to Session]

Dosimetrical Verification of Output of Small Cones in a Radiosurgery Device

L Jin*, S Hayes, C Ma, Fox Chase Cancer Center, Philadelphia, PA

Presentations

PO-GePV-T-347 (Sunday, 7/25/2021)   [Eastern Time (GMT-4)]

Purpose: Dosimetry for small fields presented a challenge for determining output factors for small-sized cones, so the accuracy delivered by small-sized cones could be affected if there are some errors in output factors. This work designed a method to verify the dose accuracy for small cones used in a Cyberknife.

Methods: A “a large uniform dose area” plan using 5 mm cone was generated for a prostate (60 mm lesion size) and a brain case with a large lesion (35 mm in size), and then recalculated with a phantom. A point dose (placed at a uniform dose area) was measured with a pinpoint ion chamber (0.015cc) for each phantom plan. The method to measure the point dose actually has been practiced in our routine clinical QA for plans that have a large uniform dose area, which has been approved correct. To compare, plans for these two cases are also generated with large cones (35 mm for the brain case and 60 mm for the prostate case), and their point doses for phantom plans were also measured.

Results: The average of three repeated measurements were taken as the measured dose. For plans using 5 mm cone, the measured dose is larger than the plan dose by 2.71% for the brain case while the prostate case shows that the measured dose is smaller than the plan dose by 3.8%. For the plan using large cones (35 mm above), the differences between the measured and plan doses are 1.41% and 0.8%, respectively for the brain and prostate case.

Conclusion: With acknowledging the method used for a point dose measurement is correct for a large uniform dose area, this work showed that the output for small cones in our Cyberkinfe are correct within about 3%.

ePosters

    Keywords

    Not Applicable / None Entered.

    Taxonomy

    Not Applicable / None Entered.

    Contact Email

    Share: