Assessment of inter-observer variability in delineating high-risk clinical target volumes and organs at risk in cervical cancer brachytherapy

Atumaini A. Makoba, Tanzania Atomic Energy Commission, P.O. Box 1585, Dodoma, TanzaniaFollow
Julius S. Chiuyo, Department of Physics, Mathematics and Informatics, Dar es Salaam University College of Education, University of Dar es Salaam, P.O. Box 2329, Dar es Salaam, TanzaniaFollow
Innocent J. Lugendo, Nuclear Sciences and Technology, Department of Physics, University of Dar es Salaam, P.O. Box 35063, Dar es Salaam, TanzaniaFollow

Section

Physical Sciences

Abstract

Achieving precise target volume delineation during cervical cancer treatment poses challenges due to several reasons, including difficulties in establishing the precise geometrical location of the gross target volume. Inter-observer variability in contouring anatomical structures poses a critical challenge to treatment consistency, delineation, and dosimetry accuracy. Volumetric and dosimetric records of 20 patients treated with gynecologic high-dose rate brachytherapy were analyzed to assess the interobserver variability between oncologists. Three radiation oncologists independently examined the organs at risk and high-risk critical organs on 20 computed tomography scans. The volumetric and dosimetric relationship between the contours delineated by radiation oncologists was described using the Python algorithm. The average volume and dose across oncologists, along with the corresponding median lines, were plotted. The interobserver variability in the single fraction related to the CT scan used for contouring was measured by the minimum dose to 2% and the minimum dose to 98% of the target volume. For statistical analysis, a non-parametric version of the ANOVA- equivalent test (Kruskal-Wallis H-Test) was used. The most peaked and statistically significant interobserver variability was found in the contouring of the sigmoid and the high-risk critical target volume between radiation oncologists. Although the statistical difference among radiation oncologists was not significant (p < 0.05), the results revealed dosimetric and volumetric differences among them. The findings indicate that systematic evaluation of interobserver variability can serve as a constructive mechanism for improving target and organs at risk delineation consistency by identifying sources of contouring variation. Such assessments can support refinement of clinical practice and contribute to enhanced treatment accuracy and clinical outcomes.

Recommended Citation

Makoba, Atumaini A.; Chiuyo, Julius S.; and Lugendo, Innocent J. (2026) "Assessment of inter-observer variability in delineating high-risk clinical target volumes and organs at risk in cervical cancer brachytherapy," Tanzania Journal of Science: Vol. 52: Iss. 3, Article 18.
Available at:https://doi.org/10.65085/2507-7961.2161