Faculty of Medicine, Graduate School of Health Sciences, Okayama University, 2-5-1 Shikata, Kita-ku, Okayama, Japan , tanabey@okayama-u.ac.jp
Abstract: (6 Views)
Background:Predicting the direction and amount of movement of the patient-specific prostate at an early stage of treatment is important for estimating systematic errors and avoiding large dose differences between planning and actual treatment. This study aimed to evaluate a two-plane analysis of prostate movement for multiple-image matching and examine the accuracy of predicting the amount of prostate movement for inter- and intra-fraction setup errors at the early stage of treatment. Materials and Methods: Sixty-five patients who underwent prostate intensity-modulated rotating radiotherapy with fiducial markers were examined for setup errors in bone matching and inter- and intra-marker matching. The two-plane setup errors in the anterior-posterior (AP), left-right (LR), and superior-inferior (SI) directions were analyzed. Correlation analysis was performed by calculating the relationship between the total average setup error and each average setup error (2‒5 fractions), increasing from the first to the sixth fraction. Results: The inter- and intra-fraction setup errors between the AP and SI directions of prostate movement were moderately correlated (r: 0.63, r =0.58, respectively). The average setup error of >4 fractions was strongly correlated (r >0.7), and the standard deviation of the >3 fraction setup errors was moderately correlated (r >0.4) with total and early setup errors. Conclusions: Prostate movement during radiotherapy was linear in the AP-SI direction. The evaluation of early fraction setup errors may be used to predict prostate movement in individual patients during the treatment period.