Development of a Statistical Framework for Optimizing Image-Guidance Frequency in Whole-Breast Radiotherapy

BACKGROUND: Radiation therapy after breast-conserving surgery reduces local recurrence. Three-dimensional conformal radiotherapy (3D-CRT) is widely used, although large setup errors may compromise treatment reproducibility. Frequent image guidance improves setup verification but increases imaging exposure. This study developed a statistical framework to assess setup deviations and inform image-guidance frequency. Methods: Couch coordinates from 35 patients (596 treatment fractions) receiving postoperative whole-breast radiotherapy were retrospectively analyzed. Absolute Z-coordinate deviations from the mean couch position of the first three fractions were calculated, and correlations with image-guidance corrections were assessed. Statistical thresholds were derived from cumulative distributions. Results: A positive correlation was observed between couch-coordinate deviations and image-guidance corrections (Spearman's ρ = 0.75, p < 0.05). The 95th and 99th percentile statistical thresholds for the absolute Z-coordinate deviation were 4.8 mm and 6.3 mm, respectively. Retrospective application of the proposed flowchart identified 45 fractions exceeding the P95 threshold, including 36 fractions with consecutive threshold exceedances suggesting potential systematic setup deviations. Of 596 fractions, 81 met the framework's predefined criteria for image guidance. Of 183 fractions actually imaged, 81 met these criteria and 102 did not. This classification does not establish that those sessions could safely or appropriately have been omitted. Hypothetically omitting these 102 fractions from imaging would correspond to an estimated exposure difference of 58-145 mGy per patient, depending on protocol; this was not directly measured. Conclusion: The proposed framework provides a quantitative method for evaluating setup deviations using routinely available couch-coordinate information. Patient-level bootstrap resampling demonstrated statistical stability of the thresholds within this dataset. However, the thresholds have not been calibrated against clinically meaningful geometric or dosimetric endpoints, and the safety of deferring image guidance remains unestablished. External validation and prospective assessment of treatment accuracy, dosimetric impact, and clinical outcomes are needed before implementation.

Authors

Institutions

Publication Details

Journal
Journal of Radiological Protection
Published
2026-11-18
DOI
https://doi.org/10.1088/1361-6498/aec3b1
Primary Topic
Advanced Radiotherapy Techniques
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Development of a Statistical Framework for Optimizing Image-Guidance Frequency in Whole-Breast Radiotherapy

Ririko Kosugi, Hiroki Kiyohara, Hayato Tsuno, YUKIHIKO YOSHIMATSU et al.
Journal of Radiological Protection
Advanced Radiotherapy Techniques
article

Development of a Statistical Framework for Optimizing Image-Guidance Frequency in Whole-Breast Radiotherapy

Ririko Kosugi, Hiroki Kiyohara, Hayato Tsuno, YUKIHIKO YOSHIMATSU, Yasuhiro Kawashima, Koji Sasaki
article en

Abstract

BACKGROUND: Radiation therapy after breast-conserving surgery reduces local recurrence. Three-dimensional conformal radiotherapy (3D-CRT) is widely used, although large setup errors may compromise treatment reproducibility. Frequent image guidance improves setup verification but increases imaging exposure. This study developed a statistical framework to assess setup deviations and inform image-guidance frequency. Methods: Couch coordinates from 35 patients (596 treatment fractions) receiving postoperative whole-breast radiotherapy were retrospectively analyzed. Absolute Z-coordinate deviations from the mean couch position of the first three fractions were calculated, and correlations with image-guidance corrections were assessed. Statistical thresholds were derived from cumulative distributions. Results: A positive correlation was observed between couch-coordinate deviations and image-guidance corrections (Spearman's ρ = 0.75, p < 0.05). The 95th and 99th percentile statistical thresholds for the absolute Z-coordinate deviation were 4.8 mm and 6.3 mm, respectively. Retrospective application of the proposed flowchart identified 45 fractions exceeding the P95 threshold, including 36 fractions with consecutive threshold exceedances suggesting potential systematic setup deviations. Of 596 fractions, 81 met the framework's predefined criteria for image guidance. Of 183 fractions actually imaged, 81 met these criteria and 102 did not. This classification does not establish that those sessions could safely or appropriately have been omitted. Hypothetically omitting these 102 fractions from imaging would correspond to an estimated exposure difference of 58-145 mGy per patient, depending on protocol; this was not directly measured. Conclusion: The proposed framework provides a quantitative method for evaluating setup deviations using routinely available couch-coordinate information. Patient-level bootstrap resampling demonstrated statistical stability of the thresholds within this dataset. However, the thresholds have not been calibrated against clinically meaningful geometric or dosimetric endpoints, and the safety of deferring image guidance remains unestablished. External validation and prospective assessment of treatment accuracy, dosimetric impact, and clinical outcomes are needed before implementation.

Journal of Radiological Protection
Hamamatsu University School of Medicine (JP), Maebashi Red Cross Hospital (JP), Hamamatsu University Hospital (JP), Gunma Prefectural College of Health Sciences (JP)
Good health and well-being
Advanced Radiotherapy Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon