Click here to


Are you sure ?

Yes, do it No, cancel

Impact of Image Registration Region/intensity Selection On H&N Patient Daily Setup / Dose Delivery Accuracy

M Su, V Osborn, S Sura , Icahn School of Medicine at Mount Sinai, Elmhurst, NY


(Tuesday, 7/16/2019) 1:15 PM - 1:45 PM

Room: Exhibit Hall | Forum 5

Purpose: Study the impact of image registration region/intensity selection on daily setup/planned dose delivery accuracy.

Methods: A left maxillary alveolar region cancer patient was used in this study. Immobilization technique included: CIVICO thermoplastic head, neck and shoulder mask, headrest, shoulder retractor, knee cushion. Two-lateral and one-anterior isocenter markers and one middle anterior tattoo which just outside the inferior board of the mark were used for setup laser alignment. Two CTs were scanned with 3mm slice thickness under same immobilization and setup in 10 days interval. Two CTs were registered with region selected as case1/caseA: superior of temporal bone to C3; case2/caseB: C4 to T1; case3/caseC: T1 to chest; case4: entire vertebrae body in the CT and case5/caseD: entire treatment area by using Eclipse V11 auto registration function. A “bones� box was checked in intensity range window during the case1 to case5 registration but not for others. To study dose variation, a PET-CT was registered with CTs to get PTVs. A 7-field SIB-IMRT plan to treat PTV66, PTV60 and PTV54 in 33 fractions was created on 1st CT image and assigned to 2nd CT image based on each image registration. Dose was calculated with the setup shift but not rotation correction. The amount of patient shift, rotation, the dose variation at D98, D95, D50 and D2 of PTV66, PTV60 and PTV54, Dmax of spine, brainstem, mandible bone, brachial plexus and Dmean of parotid were evaluated.

Results: In this study, the maximum setup correction difference was 0.51cm, 0.34cm, 0.26cm in x-axis, y-axis, z-axis and 2.29°, 1.68°, 1.08° in pitch, yaw and roll. The caseD showed the smallest dose variation from the planned dose for both PTVs and OARs.

Conclusion: Without couch rotation correction, an entire treatment region image registration for the setup correction showed a closest results to the planned dose.


Setup Verification, Image Fusion, 3D


IM/TH- Image registration : General (Most aspects)

Contact Email