Objective:To investigate the feasibility and optimal expansion width of replacing the left anterior descending coronary artery (LADCA) with the region of heart sparing (RHS) to reduce cardiac radiation dose during breast cancer radiotherapy.Methods:Retrospective analysis was conducted on data from 88 patients with left-sided breast cancer who underwent radiotherapy at 2 centers: Nanfang Hospital of Southern Medical University (50 cases for the training set, 15 cases for the internal test set) and Ganzhou Hospital of Nanfang Hospital (23 cases for the external test set) from March 2022 to January 2024. All patients had left-sided invasive ductal carcinoma with axillary lymph node metastasis, and had undergone modified radical mastectomy and chemotherapy. Based on simulation CT images, 2 radiation oncologists delineated the LADCA and 8 RHSs. The RHSs were delineated by expanding the LADCA contour by 0.5 cm increments, totaling 8 expansions. The RHS widths were defined as 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, and 4.0 cm. The nnU-Net model was trained for 3D automatic segmentation of the LADCA and RHSs. Model performance was evaluated using the Dice similarity coefficient (DSC), relative volume error (RVE), sensitivity, specificity, and 95% Hausdorff distance (HD95). Additionally, the minimum, maximum, and average relative dose variations (RDV) as well as V5% and V20% indicators were calculated for the LADCA and each RHS. Correlation analysis was performed using the least squares regression, with the slope and coefficient of determination ( R2) employed to evaluate the accuracy of the model fitting, the relationship between the LADCA and RHS, and the degree of their correlation, thereby assessing the substitutive effect of the RHS for the LADCA. Results:The DSC for the LADCA was 0.415, while the DSCs for RHS widths of 0.5 cm and 4.0 cm were 0.718 and 0.835, respectively. Overall, the automatic segmentation performance improved with increasing RHS width. The DSC, RVE, sensitivity, specificity, and HD95 for the external test set were largely consistent with those of the internal test set, demonstrating the model's good robustness across different datasets. All RDVmin values were negative, while RDVmax and RDVmean showed a positive correlation with RHS width. RDVmean increased from 39.01% to 75.89% as the RHS width increased. In the correlation analysis, the slopes for RHS widths of 1.5 cm and 2.0 cm were 0.95 and 1.05, respectively, with R2 values and coefficients of variation of 0.79 and 0.73, and 21.11% and 24.03%, respectively. Conclusions:The automatic segmentation model trained on nnU-Net can accurately segment RHSs. Based on geometric and dosimetric indicators, a 1.5 cm-wide RHS is the most suitable substitute for the LADCA, effectively limiting the radiation dose to the LADCA without compromising target dose coverage.

Objective:To propose a semi-supervised learning method for dynamic generation of organ geometric contours, leveraging bladder volume variations and its relative position to the uterus to accurately generate uterine contours in cervical cancer radiotherapy.Methods:A total of 120 sets of pelvic planning CT images (including both full and empty bladder scans) from 60 patients with cervical cancer treated at the Department of Radiation Oncology, Nanfang Hospital of Southern Medical University between January and December 2023 were retrospectively collected. A conditional generative adversarial network (CGAN) based on a squeeze-and-excitation channel attention mechanism was proposed to accurately generate uterine geometric contours under varying bladder filling states. By emphasizing the critical spatial relationships between the bladder and uterus, the model learned the relative anatomical positions of pelvic organs and their motion correlations. The generative performance was quantitatively evaluated using the average Dice similarity coefficient (DSC), intersection over union (IoU), and the 95 th percentile Hausdorff distance (HD95), and was compared with GAN model, CGAN model, and Pix2Pix model. Pairwise comparisons were perfomed by paired-sample t-test. Results:The proposed SE-CGAN model achieved the best performance on the test set, with DSC of 0.83±0.09, IoU of 0.71±0.05, HD95 of (6.74±1.23) mm, improving DSC by 7.5%, 4.9%, and 3.6% compared to the GAN, CGAN, and Pix2Pix models, respectively (all P<0.001), and reducing the mean HD95 by 32.9%-45.3%. Statistical analysis revealed significant differences between SE-CGAN model and the other 3 baseline models, whereas no significant difference was observed between CGAN model and Pix2Pix model. The visualization results further demonstrated that the GAN model produced uterine contours deviated greatly from the real shape, and the edge was fuzzy; CGAN and Pix2Pix model achieved better overlap but lacked of precision in boundary reconstruction. In contrast, the contours generated by SE-CGAN model closely matched the ground truth with clearly defined edges, indicating superior reconstruction accuracy. Conclusions:In this study, we propose a generative adversarial network method that establishes a dynamic modulation mechanism by which the bladder state influences the uterine geometric contour, enabling accurate generation of the uterine contours from the bladder contours of any given localization CT scan. This approach effectively addresses the uncertainty in radiotherapy target delineation caused by pelvic organ motion.

Objective:To investigate the feasibility and optimal expansion width of replacing the left anterior descending coronary artery (LADCA) with the region of heart sparing (RHS) to reduce cardiac radiation dose during breast cancer radiotherapy.Methods:Retrospective analysis was conducted on data from 88 patients with left-sided breast cancer who underwent radiotherapy at 2 centers: Nanfang Hospital of Southern Medical University (50 cases for the training set, 15 cases for the internal test set) and Ganzhou Hospital of Nanfang Hospital (23 cases for the external test set) from March 2022 to January 2024. All patients had left-sided invasive ductal carcinoma with axillary lymph node metastasis, and had undergone modified radical mastectomy and chemotherapy. Based on simulation CT images, 2 radiation oncologists delineated the LADCA and 8 RHSs. The RHSs were delineated by expanding the LADCA contour by 0.5 cm increments, totaling 8 expansions. The RHS widths were defined as 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, and 4.0 cm. The nnU-Net model was trained for 3D automatic segmentation of the LADCA and RHSs. Model performance was evaluated using the Dice similarity coefficient (DSC), relative volume error (RVE), sensitivity, specificity, and 95% Hausdorff distance (HD95). Additionally, the minimum, maximum, and average relative dose variations (RDV) as well as V5% and V20% indicators were calculated for the LADCA and each RHS. Correlation analysis was performed using the least squares regression, with the slope and coefficient of determination ( R2) employed to evaluate the accuracy of the model fitting, the relationship between the LADCA and RHS, and the degree of their correlation, thereby assessing the substitutive effect of the RHS for the LADCA. Results:The DSC for the LADCA was 0.415, while the DSCs for RHS widths of 0.5 cm and 4.0 cm were 0.718 and 0.835, respectively. Overall, the automatic segmentation performance improved with increasing RHS width. The DSC, RVE, sensitivity, specificity, and HD95 for the external test set were largely consistent with those of the internal test set, demonstrating the model's good robustness across different datasets. All RDVmin values were negative, while RDVmax and RDVmean showed a positive correlation with RHS width. RDVmean increased from 39.01% to 75.89% as the RHS width increased. In the correlation analysis, the slopes for RHS widths of 1.5 cm and 2.0 cm were 0.95 and 1.05, respectively, with R2 values and coefficients of variation of 0.79 and 0.73, and 21.11% and 24.03%, respectively. Conclusions:The automatic segmentation model trained on nnU-Net can accurately segment RHSs. Based on geometric and dosimetric indicators, a 1.5 cm-wide RHS is the most suitable substitute for the LADCA, effectively limiting the radiation dose to the LADCA without compromising target dose coverage.

Objective:To propose a semi-supervised learning method for dynamic generation of organ geometric contours, leveraging bladder volume variations and its relative position to the uterus to accurately generate uterine contours in cervical cancer radiotherapy.Methods:A total of 120 sets of pelvic planning CT images (including both full and empty bladder scans) from 60 patients with cervical cancer treated at the Department of Radiation Oncology, Nanfang Hospital of Southern Medical University between January and December 2023 were retrospectively collected. A conditional generative adversarial network (CGAN) based on a squeeze-and-excitation channel attention mechanism was proposed to accurately generate uterine geometric contours under varying bladder filling states. By emphasizing the critical spatial relationships between the bladder and uterus, the model learned the relative anatomical positions of pelvic organs and their motion correlations. The generative performance was quantitatively evaluated using the average Dice similarity coefficient (DSC), intersection over union (IoU), and the 95 th percentile Hausdorff distance (HD95), and was compared with GAN model, CGAN model, and Pix2Pix model. Pairwise comparisons were perfomed by paired-sample t-test. Results:The proposed SE-CGAN model achieved the best performance on the test set, with DSC of 0.83±0.09, IoU of 0.71±0.05, HD95 of (6.74±1.23) mm, improving DSC by 7.5%, 4.9%, and 3.6% compared to the GAN, CGAN, and Pix2Pix models, respectively (all P<0.001), and reducing the mean HD95 by 32.9%-45.3%. Statistical analysis revealed significant differences between SE-CGAN model and the other 3 baseline models, whereas no significant difference was observed between CGAN model and Pix2Pix model. The visualization results further demonstrated that the GAN model produced uterine contours deviated greatly from the real shape, and the edge was fuzzy; CGAN and Pix2Pix model achieved better overlap but lacked of precision in boundary reconstruction. In contrast, the contours generated by SE-CGAN model closely matched the ground truth with clearly defined edges, indicating superior reconstruction accuracy. Conclusions:In this study, we propose a generative adversarial network method that establishes a dynamic modulation mechanism by which the bladder state influences the uterine geometric contour, enabling accurate generation of the uterine contours from the bladder contours of any given localization CT scan. This approach effectively addresses the uncertainty in radiotherapy target delineation caused by pelvic organ motion.