CT and MR image fusion of tandem and ring applicator using rigid registration in intracavitary brachytherapy planning

Abstract

The purpose of this study is to find the uncertainties in the reconstruction of MR compatible ring‐tandem intracavitary applicators of high‐dose rate image‐based brachytherapy treatment planning using rigid registration of 3D MR and CT image fusion. Tandem and ring reconstruction in MR image based brachytherapy planning was done using rigid registration of CT and MR applicator geometries. Verifications of registration for applicator fusion were performed in six verification steps at three different sites of tandem ring applicator set. The first site consists of three errors at the level of ring plane in (1) cranio–caudal shift (Cranial Shift) of ring plane along tandem axis, (2) antero–posterior shift (AP Shift) perpendicular to tandem axis on the plane containing the tandem, and (3) lateral shift (Lat Shift) perpendicular to the plane containing the tandem at the level of ring plane. The other two sites are the verifications at the tip of tandem and neck of the ring. The verification at the tip of tandem consists of two errors in (1) antero–posterior shift (AP Shift) perpendicular to tandem axis on the plane containing the tandem, and (2) lateral shift (Lat Shift) perpendicular to the plane containing the tandem. The third site of verification at the neck of the ring is the error due to the rotation of ring about tandem axis. The impact of translational errors from −5 mm to 5 mm in the step of 1 mm along x‐, y‐, and z‐axis and three rotational errors about these axes from −19.1° to 19.1° in the step of 3.28° on dose‐volume histogram parameters (normalD2cc,normalD1cc,normalD0.1cc, and normalD5cc of bladder, rectum, and sigmoid, and D90 and D98 of HRCTV were also analyzed. Maximum registration errors along cranio–caudal direction was 2.2 mm (1 case), whereas the errors of 31 out of 34 cases of registration were found within 1.5 mm, and those of two cases were less than 2 mm but greater than 1.5 mm. Maximum rotational error of ring about tandem axis was 3.15° (1.1 mm). In other direction and different sites of the ring applicator set, the errors were within 1.5 mm. The impacts of registration errors on DVH parameters of bladder, rectum, and sigmoid were very sensitive to antero–posterior shift. Cranio‐caudal errors of registration also largely affected the rectum DVH parameters. Largest change of 17.95% per mm and 20.65% per mm in all the DVH parameters of all OARs and HRCTV were observed for ϕ and normalΨ rotational errors as compare to other translational and rotational errors. Catheter reconstruction in MR image using rigid registration of applicator geometries of CT and MR images is a feasible technique for MR image‐based intracavitary brachytherapy planning. The applicator registration using the contours of tandem and neck of the ring of CT and MR images decreased the rotational error about tandem axis. Verification of CT MR image fusion using applicator registration which consists of six steps of verification at three different sites in ring applicator set can report all the errors due to translation and rota...

Bladder–Rectum Spacer Balloon versus Vaginal Gauze Packing in High Dose Rate Brachytherapy in Cervical Cancer: A Randomised Study (Part II)

“…OAR, namely the bladder, rectum and sigmoid, were contoured on computed tomography at each fraction, as recommended by the GEC-ESTRO guidelines [2]. Applicators were reconstructed on the computed tomography scan on the multiplanar reconstruction views, as previously described [14,15]. A dose of 9 Gy was prescribed to point A. Doses to 0.1, 1, 2 and 5 cm 3 volumes for the bladder, rectum and sigmoid were documented at each fraction.…”

Review of strategies for MRI based reconstruction of endocavitary and interstitial applicators in brachytherapy of cervical cancer

“…It should be based on applicator geometry instead of bony structures, and its uncertainty should be previously assessed. Oinam et al 71 report a reconstruction method of the tandem and ring applicator by using rigid registration between a CT image set and MRI. They also evaluate the registration uncertainties and conclude that their impact on DVH parameters of the bladder, rectum and sigmoid are very sensitive to antero-posterior shifts.…”