MULTI-METRIC EVALUATION OF IMAGING OUTPUT AND AFFINE CORRECTION AS A VIABLE ALTERNATIVE TO CONE BEAM COMPUTED TOMOGRAPHY IN IMAGE-GUIDED RADIOTHERAPY AT THE UNIVERSITY OF NIGERIA TEACHING HOSPITAL, ENUGU STATE
DOI:
https://doi.org/10.65360/ajemates.v18i1.1001Keywords:
Image-Guided Radiotherapy (IGRT); Cone Beam Computed Tomography (CBCT); Planar kV Imaging; Imaging Output; Affine Calibration; Setup Errors; Radiation Dose OptimizationAbstract
Background: Image-guided radiotherapy (IGRT) improves targeting accuracy but three-dimensional Cone Beam Computed Tomography (CBCT) adds substantial imaging dose, time, and resource burdens. Planar kilovoltage (kV) imaging offers faster acquisition and lower radiation output, making it an attractive low-dose alternative, particularly in resource-limited settings where daily CBCT may be impractical. Aim: To compare the radiation imaging output between kV and CBCT procedures and to evaluate whether site-specific affine calibration models can reliably reduce positional couch-shift discrepancies between the two modalities across different anatomical regions. Methods: A prospective within-patient method-comparison study was conducted at the University of Nigeria Teaching Hospital. A cohort of 39 cancer patients undergoing IGRT was recruited across four stratified treatment sites: head and neck, breast/thorax, spine, and pelvis. Multiple fraction-level couch-shift vector measurements and system-recorded imaging output indices were collected. Data analysis involved paired t-tests to evaluate imaging output differences, Bland-Altman analysis for bias, and the development/cross-validation of site-specific linear affine correction models to align kV shifts with CBCT references. Results: The radiation output of kV imaging was statistically significantly lower ($p < 0.001$) than CBCT across all four treatment sites, confirming a distinct dose advantage. Prior to calibration, geometric agreement was strongest in rigid anatomy (spine, head/neck) and weakest in the pelvis due to soft tissue variations. Post-calibration, the affine models successfully minimized systematic bias, increased tolerance pass rates, and lowered Mean Absolute Error (MAE) across all sites. Cross-validation confirmed good mathematical generalization for the spine, breast/thorax, and head and neck, whereas the pelvis remained the most unstable and difficult region to correct. Conclusion: Planar kV imaging serves as a valid, lower-dose complement to CBCT in image guidance workflows. Site-specific affine calibration significantly narrows the geometric gap in rigid and semi-rigid regions (spine, head and neck, breast), supporting selective substitution. However, CBCT remains the mandatory standard for anatomically variable and deformable soft tissue regions like the pelvis.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 African Journal of Educational Management, Teaching and Entrepreneurship Studies

This work is licensed under a Creative Commons Attribution 4.0 International License.