Fully automated rigid image registration versus human registration in postoperative spine SBRT
PO-1681
Abstract
Fully automated rigid image registration versus human registration in postoperative spine SBRT
Authors: Yutaro Koide1, Hidetoshi Shimizu1, Risei Miyauchi1, Shouichi Haimoto2, Hiroshi Tanaka3, Yui Watanabe4, Sou Adachi5, Daiki Kato6, Takahiro Aoyama1, Tomoki Kitagawa1, Hiroyuki Tachibana1, Takeshi Kodaira1
1Aichi Cancer Center, Radiation Oncology, Nagoya, Japan; 2Aichi Cancer Center, Neurosurgery, Nagoya, Japan; 3Shiokawa Hospital Gamma Knife Center, Radiation Oncology, Suzuka, Japan; 4Yachiyo Hospital Radiation Therapy Center, Radiation Oncology, Anjo, Japan; 5Aichi Medical University Hospital, Radiology, Nagakute, Japan; 6Daiyukai General Hospital, Radiology, Ichinomiya, Japan
Show Affiliations
Hide Affiliations
Purpose or Objective
Spine stereotactic body
radiation therapy (SBRT) planning
requires accurate image registration of MRI/CT-myelogram to the planning CT for
spinal cord delineation. To confirm
the fully automated rigid image registration (A-RIR) accuracy in postoperative
spine SBRT, we conducted a multicenter
non-inferiority study compared to the human registration (H-RIR).
Material and Methods
Twenty-eight
metastatic cancer patients who underwent postoperative spine SBRT are
enrolled. All patients underwent
planning CT scans in the immobilized supine position under the following conditions: field of view =
550 mm, slice thickness = 1 mm, the spatial resolution = 1.074 mm/pixel. CT-myelogram was
obtained for spinal cord delineation through an intrathecal injection of iohexol contrast, followed by CT simulation. The adopted A-RIR workflow is a
contour-focused algorithm performing a rigid registration by maximizing
normalized mutual information (NMI) restricted to the data contained within the
automatically extracted contour. The A-RIR and H-RIR were performed after an NMI-based RIR on the whole image as the baseline. Three radiation oncologists from multicenter
were prompted to review two blinded registrations and choose one for clinical
use. Indistinguishable cases were allowed to vote equivalent, counted A-RIR side. A-RIR is considered non-inferior to
H-RIR if the lower limit of the 95% confidence interval (CI) of A-RIR
preferable/equivalent is greater than 0.45. We also evaluated the NMI
improvement from the baseline and the translational/rotational errors between
A-RIR and H-RIR.
Results
The A-RIR
preferable/equivalent was selected in twenty-one patients (0.75, 95% CI: 0.55–0.89), demonstrating non-inferiority
to H-RIR. The NMI improvement of A-RIR was
greater than that of H-RIR in twenty-four patients: the mean value ± SD was 0.225 ± 0.115 in A-RIR and 0.196 ± 0.114 in H-RIR (P <0.001). The absolute
translational error was 0.38 ± 0.31 mm. The rotational error was –0.03 ± 0.20,
0.05 ± 0.19, –0.04 ± 0.20 degrees in axial, coronal, and sagittal planes
(range: –0.66–0.52).
Conclusion
A-RIR shows
non-inferior to H-RIR in CT and CT-myelogram registration for postoperative
spine SBRT planning.