Exploration of overlap volumes-based IMRT Rapidplan for cervical cancer patients with the aim of OARs sparing

Exploration of overlap volumes-based IMRT Rapidplan for cervical cancer patients with the aim of OARs sparing | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Exploration of overlap volumes-based IMRT Rapidplan for cervical cancer patients with the aim of OARs sparing Minfeng Huang, Changfei Gong, Biaoshui Liu, Yili Wang, Jinghua Zhong, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5971123/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 29 Apr, 2026 Read the published version in Physical and Engineering Sciences in Medicine → Version 1 posted 7 You are reading this latest preprint version Abstract Purpose: Developing a novel overlap volumes-basedRapidPlan model for cervical cancer radiotherapy, aiming to spare organs at risk (OARs). Methods: A retrospective analysis was conducted on 110 cases of cervical cancer patients. We utilized the planning target volume (PTV) and two organs at risk (OARs): bladder and bowelbag, as the primary protocol standards for planning. Original and unclassified dose-volume histogram (DVH) estimation models (Model-O) were trained using 80 initial plans, there are 40 cases in RMS and RML respectively. The classification criteria used the ratio of bladder volume overlapping with PTV to total bladder volume, and the ratio of bowelbag volume overlapping with PTV to total bowelbag volume. Based on the sum of these volume ratios (SVR), with thresholds of ≤0.4 and >0.4, new training sets and DVH estimation models were generated, named Model-S and Model-L respectively. The plans created using Model-O were named UMO, those configured with Model-S were referred to as RMS, and the plans configured with Model-L were called RML. An analysis of dosimetric parameters for the PTV and OARs was conducted for the three automated plans. Results: For 15 patients with SVR ≤ 0.40, UMO, RMS, and RML automatically produced clinically acceptable plans. Comparing UMO and RML, RMS reduced the high-dose region V47.25 (V105%) for the PTV and provided greater dose sparing in the Bladder and Bowelbag. In cases where SVR exceeded 0.40, the mean PTV coverage with RMS only 94.9%. Comparing to UMO, RML showed greater dose sparing for the Bladder, Bowelbag, Bladder overlap, and Bowelbag overlap, as well as improved V100% of the PTV. Conclusions: The results indicate that employing overlap volumes-based IMRT Rapidplan enhances sparing of OARs when dealing with significant overlap between PTV and OARs. overlap volumes-based Rapidplan model diversification OARs sparing Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. INTRODUCTION Cervical cancer ranks as the fourth most common malignancy among women globally, posing significant challenges to global health [1]. Treatment for cervical cancer typically involves multimodal approaches, with curative options primarily encompassing surgery and radiation therapy [2]. Radiation therapy stands as a key modality in the management of cervical cancer [3,4]. the quality of radiation therapy heavily relies on treatment planning, with one challenge being how to ensure adequate target dose delivery while sparing healthy tissues [5]. Intensity-modulated radiation therapy (IMRT) employs inverse planning to adjust the dose distribution within the target area, aiming to enhance dose uniformity within the target while minimizing doses to organs at risk (OARs) [6,7]. Currently, this technique is widely used in treating cervical cancer [8], contributing to the reduction of treatment-related side effects in patients with locally advanced cervical cancer [9]. However, due to the substantial overlap between OARs and the planning target volume (PTV), not only does it increase the complexity of planning design, but it also significantly affects the quality of IMRT plans [10]. Despite physicians typically prioritizing the protection of these organs, variations in clinical practices among different oncologists regarding planning requirements and reviews, as well as differences among planning optimization personnel, may ultimately lead to discrepancies between plans. Traditional IMRT techniques achieve optimal dose distribution through heuristic or iterative methods, heavily reliant on staff experience, resulting in lower inter-plan consistency [11,12]. Knowledge-based RapidPlan is a semi-automatic technique that extracts relevant knowledge from existing treatment plans to generate high-quality plans [13], improving both plan quality and workflow efficiency [14, 15]. Within the industry, creating RapidPlan models is typically categorized by disease site, constructing universal models to generate high-quality plans that pass clinical evaluation [16–18]. However, a minority of researchers guide model construction within a single disease site based on specific conditions, or use general models to predict doses for categorized cases. In the study by Yu et al. [19], a special model was constructed based on conditions where the distance from the PTV to the right kidney was less than 3cm, demonstrating that specific models compared to general models can improve the conformity of liver cancer target areas while reducing doses to OARs. In a study by Yihang Xu et al. [20], universal models were used to predict doses for unilateral and bilateral cases of advanced head and neck (HN) cancer patients, showing different capabilities in protecting OARs. In this study, we explored model diversification by developing three automatic planning models for cervical cancer. This involved generating the original Unclassified Model-O based on 80 cases. Simultaneously, the sum of ratio (SVR): bladder volume overlapping with PTV to total bladder volume, and the ratio of bowelbag volume overlapping with PTV to total bowelbag volume, with a threshold value of 0.4 was used for classification. We then created two new Reclassification Models: Model-S (SVR ≤ 0.4) and Model-L (SVR > 0.4), using the remaining 30 cases for validation. Our aim is to analyze whether constructing different RapidPlan Models under specific conditions, especially in cases with significant overlap between the PTV and OARs, could better protect the overlapping region while ensuring high-quality target coverage. 2. METHODS 2.1. Clinical planning We retrospectively selected a cohort of 110 patients with stage IB1 to IIIC2 cervical cancer who underwent dynamic IMRT treatment. All patients have signed an informed consent for radiotherapy, The radiotherapy data were anonymized and approved by our hospital's ethics committee with the number LLSC-2024-133. Patients were prescribed a dose of 45 Gy, delivered at 1.8 Gy per day over 5 weeks. All plans were designed using the Eclipse 15.2 treatment planning system and utilized 7 coplanar fields (0°, 52°, 105°, 156°, 208°, 260°, 312°). Plan optimization and dose calculations were performed using the photon algorithm (PVO) version 15.5.12 and the anisotropic analytical algorithm (AAA) version 15.5.12, with a calculation grid of 2.5 mm. CT slices were 2.5 mm thick and obtained using the GE Discovery CT590. Target volumes were delineated by experienced senior oncologists following RTOG international guidelines and reviewed by another senior oncologist. Organs at risk (OARs), including the bowel bag, bladder, and rectum, were automatically contoured using AccuContour (Manteia Technologies, China) to minimize inter-observer variability, and then reviewed and adjusted (if necessary) by two senior radiation oncologists to ensure accuracy. The Contour of the Rectum: The lower border is the anal margin, which is marked with an opaque marker during simulated positioning. The rectum is no longer circular from the upper edge to the horizontal position and is connected to the sigmoid colon in the forward direction.The Contouring of the Bowel Bag: It starts from the bottom of the small intestine or large intestine loop, or from the anorectum, whichever is the lowest. If the rectum or anorectal portion appears in the axial section following the pattern of the intestinal loop, it should be included as part of the bag; otherwise, it should be excluded.Tip: The abdominal content outline excludes muscles and bones. [21]. All plans aimed to deliver effective doses to OARs overlapping with the PTV while minimizing dose to individual OARs, with at least 95% of the PTV receiving the prescribed dose. The original clinical plans were manually optimized by experienced physicists with ample experience in cervical cancer treatment, ensuring that each IMRT plan met clinical protocols. 2.2. Model Building, Training, and Reclassification In this study, we utilized a PTV and three OARs—bladder, bowelbag, and rectum—to develop the RapidPlan DVH estimation model. Initially, we included 80 clinically approved IMRT plans for cervical cancer from our institution in the training dataset. The DVH estimation modeling engine processed the dataset for each patient, correlating dose-volume histogram information with the geometric characteristics of the respective OARs [22]. Subsequently, the RapidPlan modeling procedure was applied to these initial 80 plans, resulting in the creation of the original Unreclassification model (Model-O). Previous research has indicated that the volume of individual cases' OARs overlapping with the PTV is a crucial anatomical factor influencing OARs dose sparing [23]. Additionally, several scholars have demonstrated that the overall quality of training plans can significantly influence the outcomes produced by the model [10]. Therefore, selecting a case classification method based on the extent of overlap between different organs and the PTV is critical, especially in cervical cancer patients, where there may be substantial overlap between the bowelbag and bladder regions with the PTV. Research by O'Toole et al. suggests that combining the sparing of OARs yields greater advantages compared to independent optimization [24]. Consequently, we utilize the ratio of the bladder's overlap volume with the PTV (Bladder overlap ) to its original volume (V bladder ), and the ratio of the bowelbag's overlap volume with the PTV (Bowelbag overlap ) to its original volume (V bowelbag ), summed together as the Sum of these Volume Ratios (SVR), as the criteria for grouping. This is represented by Equation 1: Considering the number of cases we trained with, our goal was to maintain consistency in the sample size for training the RapidPlan model. Therefore, we divided the 80 patients into two groups, using the median of the SVR from Equation 1, which is 0.4, as the threshold for grouping. Among these, 40 original plans with SVR ≤ 0.40 were used to construct Reclassification Model-S, while Reclassification Model-L was trained using the remaining 40 original plans with SVR > 0.40.One advantage of this approach is that by consolidating the modeling of several OAR DVHs within the training plans, the model can achieve a comprehensive balance in predicting dose protection among OARs, resulting in more balanced DVH estimates. However, a concern arises under specific patient requirements (such as mandatory targets for critical organs): the necessary trade-offs in combining OAR DVH modeling may lead to the prediction of unattainable DVH estimates. In such cases, manually setting fixed priorities for these special OARs during model construction may provide a more reliable solution 2.3. Classification of training samples and automatic generation of plans The performance of the DVH estimation models in automatically generating plans was validated under reclassification scenarios. The remaining 30 cases of cervical cancer were divided into two groups based on Equation 1: SVR ≤ 0.4 and SVR > 0.4, without any additional specific criteria applied. Both groups of test cases were assigned the three models mentioned earlier. Plans configured with the Unclassification Model-O were denoted as UMO, those configured with Reclassification Model-S as RMS, and those with Reclassification Model-L as RML. When using the RapidPlan model for automatic plan generation, efforts were made not only to ensure the model's capability to generate automated plans but also to consider multiple criteria for OAR/PTV during the optimization process. However, determining the optimal combination of specific details such as OAR targets and priorities, especially in light of the patient's pathological characteristics, is not straightforward. The process of constructing the RapidPlan model and validating automated plans is illustrated in Figure 1. The primary steps involved in the models used were as follows: (1) A model has been selected. (2) Manually aligned the structure of the validation plan with that of the model. (3) Utilizing a sophisticated regression model, we generated an estimation range for dose-volume histogram (DVH) constraints and automatically established the corresponding dose-volume limits. (4) Finally, we optimized the verification plan. 2.4. Analysis and validation of plans The validation of the evaluation set were completed by comparing the PTV and OAR DVH results among RMS, RML, and the original plan UMO. Criteria for assessing the quality of radiation therapy plans for cervical cancer were established based on dose constraints outlined in the International Commission on Radiation Units and Measurements (ICRU) Report 83, as well as recommendations from the GEC-ESTRO Gynecological Working Group [25–27], taking into account specific dose requirements for cervical cancer at our center. Key dosimetric evaluation indices for the target volume (PTV) include minimum dose (Dmin), maximum dose (Dmax), mean dose (Dmean), D2%, D50%, D98%, V45, and V47.25. Conformity Index (CI) and Homogeneity Index (HI) are important metrics used to assess target dose. The calculation method for CI is as follows [28]: A Homogeneity Index (HI) closer to 0 indicates better homogeneity within the target volume. The irradiation dose to organs at risk should be kept as low as reasonably achievable, considering their maximum tolerated doses, while ensuring that the overlap regions between organs at risk and target volumes meet clinical dosimetric criteria. The primary dosimetric evaluation parameters for organs at risk are as follows: bowel V40, V45 (V100%), and V47.25 (V105%); bladder V40, V45, and V47.25. Additionally, for organs at risk overlapping with targets, the primary dosimetric evaluation parameters include bowelbag overlap of V40, V45, and V47.25; bladder overlap of V40, V45, and V47.25. 2.5. Statistical Analysis Plan data generated by the three automatic planning models were compared between groups using SPSS 26.0 software. The Wilcoxon signed-rank test was conducted on the datasets, with statistical significance defined as P < 0.05. 3. RESULTS In the study of 110 cases, the SVR ranged from [0.148–0.961] in 80 training cases, with a mean and standard deviation of 0.442 ± 0.16. For 30 training cases, the SVR ranged from [0.166–0.827], with a mean and standard deviation of 0.438 ± 0.17. Detailed patient characteristics are provided in Table 1. We compared automatic plan quality differences using models with different configurations across two sets of validation cases. Table 2 shows PTV dose metrics for 15 patients with SVR ≤ 0.40, revealing no significant differences among the three plans for Dmin, Dmax, and D2% (p > 0.05). However, the Conformity Index (CI) of RMS differed significantly from UMO (p = 0.015). In terms of Dmean, RMS closely aligned with the prescription dose compared to UMO and RML (p = 0.006, p = 0.044, respectively). The D98% of RMS was significantly lower than RML (p = 0.033). RMS exhibited a smaller V45 compared to UMO by 0.22 (p = 0.007) and compared to RML by 0.24 (p = 0.038), remaining at 95.53%. The high-dose region V47.25 (V105%) of RMS was significantly lower than UMO (p = 0.007), showing a decrease of 1.19 percentage points. Although not significant (p = 0.0965), RMS's V47.25 was 1.08 percentage points lower than RML. Table 3 presents PTV dose metrics for automated plans in patients with SVR > 0.40. For V45, RML was significantly higher than UMO and RMS (p = 0.006, p = 0.001, respectively), with RMS averaging only 94.9%, indicating that many RMS plans fail to meet clinical requirements. RML's D98% and CI were slightly inferior to UMO, these differences were not statistically significant. Meanwhile, RML outperformed UMO significantly in Dmax, Dmean, D2%, D50%, V45, V47.25 (V105%), and Homogeneity Index (HI) (p = 0.047, p = 0.013, p = 0.003, p = 0.019, p = 0.06, p = 0.002, p = 0.002, respectively). Notably, RML's V47.25 was 1.58 percentage points lower than UMO. Table 1 Characteristics of the 110 patients enrolled in this study Characteristic n Number of patients 110 Age (years) Mean±SD 55.65±10.63 Median[range] 56[25-86] Pathological Type and Staging Squamous Cell Carcinoma 85 Adenocarcinoma 17 Adenosquamous Carcinoma 3 Other 5 Staging IB1~IIIC2 Prescription dose (Gy) 45 Sum of the volume ratios 80 Initial Patients Mean±SD Median[range] 0.442±0.16 0.400[0.148-0.961] 30 Validation Patients Mean±SD Median[range] 0.438±0.17 0.400[0.166-0.827] Abbreviations: SD =Standard deviation. Table 2 The PTV’s DVH metrics for UMO, RMS and RML with SVR≤0.40 Parameter UMO RMS RML P Dmin[Gy] Mean±SD 39.19±0.71 38.5±2.14 38.51±2.14 i=0.307;ii=0.156;iii=0.51 Median[range] 39.21[37.43 -40.56] 39.35[31.63-40.5] 39.06[31.31-40.49] Dmax[Gy] Mean±SD 49.32±0.57 49.34±0.46 49.36±0.52 i=1;ii=0.379;iii=0.73 Median[range] 49.26[48.71-50.84] 49.31[48.79-50.57] 49.25[48.6-50.57] Dmean[Gy] Mean±SD 46.47±0.1 46.45±0.09 46.48±0.11 i=0.006;Ii=0.533;iii=0.044 Median[range] 46.44[46.33-46.63] 46.42[46.32-46.62] 46.44[46.34-46.75] D2%[Gy] Mean±SD 47.83±0.11 47.81±0.08 47.82±0.11 i=0.131;ii=0.561;iii=0.925 Median[range] 47.83[47.67-48.03] 47.84[47.66-47.93] 47.8[47.67-48.08] D50%[Gy] Mean±SD 46.5±0.1 46.47±0.1 46.51±0.11 i=0.004;ii=0.329;iii=0.012 Median[range] 46.46[46.39-46.69] 46.46[46.37-46.67] 46.48[46.4-46.82] D98%[Gy] Mean±SD 44.54±0.14 44.52±0.12 44.55±0.12 i=0.052;ii=0.381;iii=0.033 Median[range] 44.51[44.32-44.76] 44.53[44.3-44.68] 44.55[44.34-44.74] V45[%] Mean±SD 95.75±0.78 95.53±0.72 95.77±0.71 i=0.007;ii=0.505;iii=0.038 Median[range] 95.7[94.1-96.9] 95.7[93.7-96.5] 95.7[94.2-97] V47.25[%] Mean±SD 17.25±3.87 16.06±3.29 17.14±4.78 i=0.007;ii=0.798;iii=0.096 Median[range] 16[12.4-24.2] 15.2[10.9-21.9] 16.1[11.5-30.3] CI Mean±SD 0.997±0.016 0.993±0.014 0.996±0.016 i=0.015;ii=0.57;iii=0.069 Median[range] 0.994[0.972-1.023] 0.989[0.968-1.019] 0.994[0.973-1.028] HI Mean±SD 0.071±0.003 0.071±0.003 0.070±0.002 i=0.733;ii=0.069;iii=0.011 Median[range] 0.07[0.063-0.076] 0.071[0.064-0.075] 0.07[0.064-0.075] Abbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P0.40 Parameter UMO RMS RML P Dmin[Gy] Mean±SD 39.27±0.71 39.04±0.89 39.33±0.76 i=0.017;ii=0.733;iii=0.011 Median[range] 39.28[38-40.81] 38.95[37.29-40.96] 39.37[37.83-40.77] Dmax[Gy] Mean±SD 49.25±0.53 49.15±0.48 49.17±0.51 i=0.055;ii=0.047;iii=0.851 Median[range] 49.26[48.58-50.49] 49.07[48.55-50.4] 49.08[48.62-50.31] Dmean[Gy] Mean±SD 46.48±0.11 46.43±0.1 46.46±0.1 i=0.003;ii=0.013;iii=0.014 Median[range] 46.47[46.32-46.71] 46.42[46.31-46.67] 46.45[46.29-46.72] D2%[Gy] Mean±SD 47.87±0.16 47.82±0.16 47.82±0.18 i=0.017;ii=0.003;iii=0.889 Median[range] 47.81[47.68-48.35] 47.76[47.65-48.31] 47.76[47.64-48.39] D50%[Gy] Mean±SD 46.5±0.12 46.44±0.11 46.48±0.11 i=0.002;ii=0.019;iii=0.004 Median[range] 46.48[46.33-46.73] 46.44[46.31-46.68] 46.47[46.3-46.73] D98%[Gy] Mean±SD 44.54±0.12 44.49±0.09 44.54±0.11 i=0.011;ii=0.899;iii=0.003 Median[range] 44.54[44.38-44.73] 44.5[44.35-44.61] 44.54[44.37-44.75] V45[%] Mean±SD 95.02±2.37 94.9±2.27 95.33±2.05 i=0.137;ii=0.006;iii=0.001 Median[range] 95.5[86.7-96.8] 95.3 [87-96.2] 95.7 [88.4-97.1] V47.25[%] Mean±SD 17.97±4.92 15.97±4.4 16.39±4.6 i=0.004;ii=0.002;iii=0.293 Median[range] 17[11.8-29.1] 15.1[10.1-27.4] 15.6[10.8-29.1] CI Mean±SD 0.99±0.011 0.986±0.01 0.989±0.01 i=0.011;ii=0.069;iii=0.006 Median[range] 0.99[0.976-1.008] 0.985[0.971-1.002] 0.988[0.975-1.006] HI Mean±SD 0.071±0.003 0.072±0.003 0.071±0.004 i=0.307;ii=0.002;iii=0.003 Median[range] 0.071[0.065-0.081] 0.072[0.066-0.081] 0.07[0.064-0.082] Abbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P<0.05 indicates a statistical significance.SD =Standard deviation. When the SVR was ≤0.40 in the test group, dose variations in the Bladder and Bowelbag are shown in Table 4. For the Bladder, RML exhibited significantly higher Dmean (p=0.012) and V20 (p=0.004) compared to RMS, while V30 (p=0.826) and V45 (p=0.052) showed no significant differences between RMS and RML. There were no significant differences in Bladder dose between RMS and UMO (p > 0.05). RMS showed a decrease of 0.25 percentage points in the high-dose region V47.25 compared to UMO (p=0.011) and 0.33 percentage points compared to RML (p=0.011). For the Bowelbag, there were no significant differences in Dmean and V40Gy among UMO, RMS, and RML. However, V45% of RMS was significantly lower than UMO and RML (p=0.039, p=0.039, respectively). In terms of V47.25, RMS decreased by 0.15% compared to UMO (p=0.003) and by 0.19% compared to RML (p=0.001). For the Rectum, except for significantly lower Dmean in RMS compared to UMO and RML, and significantly lower V45% in UMO and RMS compared to RML, there were no significant differences in other parameters. Table 5 presents dose variations in the Bladder and Bowelbag when SVR was greater than 0.40. For the Bladder, there were no significant differences in Dmean, V20, and V30 among UMO, RMS, and RML. However, RMS had the lowest V45 (p=0.001, p=0.003, respectively) and also the lowest V47.25 (V105%), being only 2.80% (p=0.002, p=0.004, respectively). In the Bowelbag, although there were no significant differences in Dmean and V40Gy among UMO, RMS, and RML, RMS provided the lowest values. RMS had significantly lower V45 compared to RML and UMO (p=0.007, p=0.031, respectively). RMS's V47.25 was only 1.52 compared to RML and UMO (p=0.001, p=0.001, respectively). For the Rectum, RMS had the lowest Dmean and V40, with only Dmean showing significant differences compared to RML (p=0.017). The V45 of RMS was 28.85, significantly lower than UMO and RML (p=0.033, p=0.005, respectively). Figure 2 illustrates changes in the Bladder and Bowelbag for each patient in the training and validation sets, focusing on the hotspot region (V105%) of protocol-specified DVH metrics in both reclassified and original plans. Table 4 Bladder, BowelBag and Rectum DVH metrics for UMO, RMS and RML with SVR≤0.40 Structure/Parameter UMO RMS RML P Bladder Dmean[Gy] Mean±SD 33.04±1.7 33.12±1.68 33.23±1.75 i=0.256;ii=0.002;iii=0.012 Median[range] 32.89[30.26-35.7] 33.01[30.25-35.74] 33[30.27-35.86] V20[%] Mean±SD 87.44±6.27 87.39±5.23 88.31±6.09 i=0.802;iii=0.016;iii=0.004 Median[range] 89.7[78-95.5] 90.6[78-93.9] 91.1[77.8-96] V30[%] Mean±SD 57.05±6.63 57.47±6.58 57.59±6.94 i=0.050;ii=0.008;iii=0.826 Median[range] 56.7[47.2-70.2] 57.4[47.3-69.4] 57.1[47.1-71] V45[%] Mean±SD 21.91±5.79 21.87±5.83 22.07±5.78 i=0.936;ii=0.088;iii=0.052 Median[range] 23.2[11.2-30.6] 23.2[11.3-30.6] 23.6[11.6-30.8] V47.25[%] Mean±SD 2.25±1.6 2.00±1.24 2.33±1.47 i=0.011;ii=0.243;iii=0.011 Median[range] 1.90[0-5.6] 1.90[0-4] 2.2[0-5.6] BowelBag Dmean Mean±SD 21.83±3.51 21.89±3.5 21.85±3.51 i=0.047;ii=0.842;iii=0.272 Median[range] 22.59[15.76-28.09] 22.6[15.7-28.07] 22.53[15.67-28.05] V40[%] Mean±SD 13.17±4.71 13.2±4.67 13.19±4.65 i=0.272;ii=0.546;iii=0.726 Median[range] 12[6.5-20.9] 12.1[6.5-20.9] 12.2[6.5-20.9] V45[%] Mean±SD 8.35±3.83 8.29±3.81 8.35±3.77 i=0.039;ii=0.763;iii=0.039 Median[range] 7.1 [3-15.4] 7.1[2.9-15.3] 7.2[3-15.3] V47.25[%] Mean±SD 0.66±0.45 0.51±0.38 0.70±0.44 i=0.003;ii=0.207;iii=0.001 Median[range] 0.7[0-1.7] 0.5[0-1.4] 0.7[0.1-1.6] Rectum Dmean[Gy] Mean±SD 36.42±3.54 36.09±3.93 36.45±3.49 i=0.036;ii=0.221;iii=0.011 Median[range] 36.68[25.36-41.84] 36.37[23.34-41.3] 36.78[25.3-41.39] V40[%] Mean±SD 53.35±12.78 53.45±13.36 53.63±12.65 i=0.865;ii=0.345;iii=0.609 Median[range] 55.1[17.6-73.6] 56.5[15.8-73.5] 56.5[18.8-73] V45[%] Mean±SD 20.43±8.63 19.77±9.74 22.59±9.5 i=0.155;ii=0.001;iii=0.004 Median[range] 22.1[1-36.5] 20.2[0.9-40.9] 23.4[1.7-43.4] Abbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P0.40 Structure/Parameter UMO RMS RML P Bladder Dmean[Gy] Mean±SD 39.06±1.73 38.98±1.79 39.09±1.7 i=0.139;ii=0.955;iii=0.094 Median[range] 39.78[36.96-42.25] 39.66[36.79-42.31] 39.63[36.92-42.23] V20[%] Mean±SD 97.26±3.24 97.13±3.48 97.31±3.3 i=0.099;ii=0.55;iii=0.133 Median[range] 98.3[89.7-100] 98.8[89.2-100] 98.9[89.9-100] V30[%] Mean±SD 80.64±6.85 80.61±7.21 80.8±6.95 i=0.95;ii=964;iii=0.66 Median[range] 81.3[70.9-91.4] 82.1[70.7-92.9] 81.3[70.6-91.5] V45 [%] Mean±SD 43.03±7.07 42.44±6.97 43.05±6.75 i=0.001;ii=0.752;iii=0.003 Median[range] 42.1[32.4-59.4] 41.7[31.3-58.4] 41.9[33.4-59] V47.25[%] Mean±SD 3.77±3.17 2.80±2.35 3.47±2.95 i=0.002;ii=0.004;iii=0.004 Median[range] 3.1[0-10.5] 2.3[0-7.6] 3[0-9.7] BowelBag Dmean[Gy] Mean±SD 24.42±3.81 24.40±3.79 24.47±3.83 i=1.000;ii=0.315;iii=0.649 Median[range] 24.28[16.22-31.68] 24.62[16.35-31.96] 24.53[16.2-31.54] V40[%] Mean±SD 21.28±6.84 21.19±6.69 21.27±6.8 i=0.449;ii=0.885;iii=0.429 Median[range] 20.2[11.1-33.4] 20.5[11.2-33.5] 20.4[11.1-33.3] V45[%] Mean±SD 15.28±5.45 15.13±5.32 15.23±5.42 i=0.007;ii=0.052;iii=0.031 Median[range] 14.8[7.5-25.9] 14.9[7.6-25.8] 14.8[7.6-25.8] V47.25 [%] Mean±SD 1.96±1.29 1.52±1.06 1.85±1.18 i=0.001;ii=0.041;iii=0.001 Median[range] 1.7[0.4-5] 1.3[0.3-4.5] 1.4[0.5-4.8] Rectum Dmean[Gy] Mean±SD 38.33±3.49 38.24±3.5 38.46±3.47 i=0.427;ii=0.363;iii=0.017 Median[range] 39.33[30.88-42.53] 39.11[31.34-42.32] 39.29[31.53-42.54] V40[%] Mean±SD 65.96±10.01 66.24±10.44 66.29±9.87 i=0.91;ii=0.865;iii=0.801 Median[range] 67.1[49.2-83.2] 66.9[47.6-82.9] 67.3[48.1-82.5] V45[%] Mean±SD 30.79±10.67 28.85±11.44 33.05±10.5 i=0.033;ii=0.003;iii=0.005 Median[range] 30.5[14.1-53] 27.5[12.9-54.5] 33[16.7-53.4] Abbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P<0.05 indicates a statistical significance.SD =Standard deviation. We also analyzed the dosimetry of the overlap regions between the Bladder and PTV (Bladder overlap ) and between the Bowelbag and PTV (Bowelbag overlap ). In the test case group with SVR ≤ 0.40 (Table 6), RMS's Bladder overlap and Bowelbag overlap showed significantly lower Dmean and V47.25 compared to UMO and RML. Additionally, although not statistically significant, RMS provided lower V45 in both Bladder overlap and Bowelbag overlap compared to UMO and RML. Only in V45 were RML's Bladder overlap and Bowelbag overlap significantly higher than UMO, with no significant differences in other parameters. In the test case group with SVR > 0.40 (Table 7), except for RML's significantly lower V47.25 in Bladder overlap compared to UMO, there were no significant differences in dose parameters for both Bladder overlap and Bowelbag overlap . However, in the hotspot region V47.25 (V105%), RML's Bowelbag overlap and Bladder overlap were respectively 0.53 and 0.65 percentage points lower than UMO. The hotspot region (V105%) of protocol-specified DVH metrics was compared, and Figure 3 showed the changes in Bladder overlap and Bowelbag overlap for each patient in the SVR ≤ 0.40 and SVR > 0.40 sets, with different verification plans plotted. Table 6 Bladder overlap and BowelBag overlap DVH metrics for UMO, RMS and RML with SVR≤0.40 Structure/Parameter UMO RMS RML P Bladder overlap Dmean[Gy] Mean±SD 46.16±0.22 46.14±0.23 46.21±0.19 i=0.019;ii=0.05;iii=0.004 Median[range] 46.14[45.67-46.5] 46.14[45.55-46.42] 46.22[45.74-46.5] V45[%] Mean±SD 93.02±2.46 92.81±2.71 93.53±1.93 i=0.813;ii=0.035;iii=0.028 Median[range] 93.1[90-97.4] 93.3[87.5-97] 93.2[90.8-96.8] V47.25 [%] Mean±SD 9.29±5.18 8.58±4.5 9.98±5.05 i=0.011;ii=0.055;iii=0.006 Median[range] 9.1[0-18.3] 8.4[0.1-15.2] 10.2[0.1-21.3] BowelBag overlap Dmean[Gy] Mean±SD 46.18±0.18 46.11±0.18 46.21±0.14 i=0.001;ii=0.053;iii=0.001 Median[range] 46.19[45.77-46.44] 46.1[45.72-46.38] 46.23[45.85-46.39] V45[%] Mean±SD 94.05±2.18 93.87±2.56 94.47±2.04 i=0.255;ii=0.048;iii=0.009 Median[range] 94.7[87.8-96.8] 93.9[86.8-97.7] 94.7[88.9-97] V47.25[%] Mean±SD 7.11±4.26 5.18±3.54 7.39±3.73 i=0.001;ii=0.182;iii=0.001 Median[range] 6.9[1-15.1] 5.4[0.6-11.8] 6.9[1.5-12] Abbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P0.40 Structure/Parameter UMO RMS RML P Bladder overlap Dmean[Gy] Mean±SD 46.15±0.3 46.05±0.3 46.15±0.28 i=0.001;ii=1;iii=0.001 Median[range] 46.21[45.67-46.52] 46.14[45.57-46.42] 46.22[45.61-46.5] V45[%] Mean±SD 93.81±3.91 92.75±4.54 94.04±3.41 i=0.001;ii=0.95;iii=0.002 Median[range] 96.2[87.3-98] 95.4[84.5-96.9] 95.2[86.3-97.8] V47.25[%] Mean±SD 8.15±6.43 6.05±4.76 7.5±5.89 i=0.002;ii=0.004;iii=0.003 Median[range] 7.9[0-19] 5.8[0-13.7] 7.5[0-17.5] BowelBag overlap Dmean[Gy] Mean±SD 46.33±0.17 46.25±0.14 46.33±0.15 i=0.002;ii=0.551;iii=0.001 Median[range] 46.26[46.08-46.61] 46.22[46.07-46.5] 46.31[46.11-46.58] V45[%] Mean±SD 95.49±1.72 95.05±1.35 95.37±1.5 i=0.018;ii=0.176;iii=0.016 Median[range] 96.2[91.5-97.3] 95.4[92.3-96.7] 95.9[92-97.1] V47.25[%] Mean±SD 11.72±5.94 9.04±4.73 11.19±5.38 i=0.001;ii=0.102;iii=0.001 Median[range] 11[4.3-21.5] 8.9[3.1-19.2] 11.5[4.2-20.8] Abbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P<0.05 indicates a statistical significance.SD =Standard deviation. As an illustrative case, Figure 4 depicts the dose distribution of the UMO, RMS, and RML plans with SVR ≤ 0.40. Figure 5 shows the DVHs of the example above, primarily including PTV, bladder, and bowelbag. In this case, the V100% for UMO, RMS, and RML was 96.7, 96.5, and 96.5, respectively; there were minimal differences in V47.25 (V105%) for the Bladder and Bowelbag. The values for Bladder overlap and Bowelbag overlap were 1.1 and 6.9 for UMO, and 1.0 and 6.0 for RML, respectively. RMS exhibited lower values in Bladder overlap and Bowelbag overlap , with V47.25 (V105%) being 0.6 and 3.8, respectively. 4. DISCUSSION This study demonstrates an effective approach for constructing overlap volumes-based training libraries for cervical cancer RapidPlan models and validates their predictive performance in DVH. Many studies have shown that establishing a general RapidPlan model for a specific disease to improve OAR dose is feasible. However, some researchers have opted to establish multiple RapidPlan models for certain cases based on specific selection criteria. Nan Li et al. [29] demonstrated that REFINED models, reconstructed for cervical cancer cases with specific dose constraints on pelvic bone marrow (PBM) and bowelbag, could better protect PBM and bowelbag. Gang Yu et al. [19] used a specific selection criterion where the distance from the PTV to the right kidney was less than three centimeters, resulting in a model that improved the PTV's conformity index (CI) and more effectively protected OARs during IMRT treatment for individual liver cancer patients. O'Toole J et al. [24] found that modeling paired OARs together yielded superior performance compared to separate modeling. The essence of the above reports is to construct diversified RapidPlan models based on specific selection criteria to enhance model performance. Therefore, when there are sufficient training case numbers, constructing specific models based on special criteria is a better way to achieve improved OAR sparing. This study utilized the volumes-based IMRT RapidPlan with SVR as a specific screening criterion to construct two models, Model-S and Model-L, which were then compared with Model-O. The findings suggest that volumes-based IMRT RapidPlan offers improved OAR sparing in scenarios with substantial overlap between PTV and OARs. This study pioneers a novel approach to diversified RapidPlan modeling, demonstrating strong generalization capability in complex clinical scenarios. For instance, in nasopharyngeal carcinoma (NPC) treatment, the anatomical overlap between target volumes and bilateral parotid glands varies significantly across different disease stages. Reducing high-dose exposure to the parotid glands (to preserve salivary function) remains a central challenge in radiotherapy planning, and the anatomy overlap volume-based RapidPlan models offer a promising solution for enhanced parotid sparing. In our study, we extended the strict classification of training cases based on specific criteria to the grouping of validation cases. While most studies construct a general model to predict all cases of a particular cancer type [30, 31], a minority utilize multiple specialized models to predict randomly selected cases. One study employs both a specialized and a general model to predict patients similar to those in the specialized model training set [33]. We reconstructed two specialized models based on the SVR of overlap regions between OARs and the PTV and applied this classification criterion to validation cases. Implementation of this approach was facilitated by the ease of obtaining overlap regions between OAR and PTV in our Eclipse system, contributing significantly to the diversification of such RapidPlan models. Additionally, we established a many-to-many mapping relationship between model construction for the same type of case and training cases. However, in our research, relying solely on patient anatomy (the ratio of OAR and PTV overlap volume) as a limiting factor raises other concerns. For instance, certain patients may have specific pathological conditions in certain organs at risk, requiring special dose sparing, potentially rendering our model predictions clinically inadequate. Therefore, constructing precise models based on a comprehensive selection criterion encompassing patient anatomy and pathological characteristics may be a future potential research direction. Another potential drawback lies in the classification criteria for patients. For instance, dividing patients into two categories based on case numbers results in a significant difference in the SVR between the two groups. In our training cases, the interval range for SVR ≤ 0.4 is [0.166–0.385], while for SVR > 0.4, it is [0.450–0.868]. The group with the larger interval range contains richer anatomical information, but it also implies a decrease in the average quality of the training plans. Currently, there are no reported studies on how to classify overlap regions between OAR and PTV. We have another idea: assuming a considerable number of cases, the SVR of cases should exhibit a normal distribution trend. We can use the 80/20 rule (also known as the Pareto principle) [34, 35] based on SVR values to divide cases into three categories: the middle 80% of cases are training into one model, while the remaining 10% at each end are using to construct two additional models. Models constructed in this manner contain more concentrated anatomical information, thereby enhancing the model's ability to predict DVHs for OARs. Alternatively, employing statistical methods (e.g., derived from ROC analysis) that balance sensitivity and specificity for clinical outcomes can help establish a more robust threshold. Extending this construction method to incorporate both patient anatomy and pathological characteristics as comprehensive selection criteria for building precise models is also highly appealing. RapidPlan is a knowledge-based, semi-automated planning system, and its predictive performance for DVHs is influenced by various factors. The quality of prior plans and anatomical information from regions of interest significantly impact the model's predictive capabilities. We classified anatomical structures based on the SVR of bowelbag and bladder overlap with the PTV, while not restricting other OARs, and established two specific RapidPlan models. Our findings may offer guidance for the clinical application of radiotherapy in cervical cancer. Future research will focus on exploring models using broader screening criteria to develop precise models tailored for individual patients, aiming to optimize clinical plans. 5. CONCLUSION In conclusion, we have explored an efficient method to train and refine the RapidPlan model using the SVR of overlapping regions between OARs and the PTV. In the final validation samples, the reclassification models have demonstrated improved sparing of normal tissue doses. In the future, we may construct diversified models based on various anatomical and case characteristics to provide better OARs sparing for individual clinical cases of cervical. Declarations Funding statement and Acknowledgments ： This research was funded by the National Natural Science Foundation of China (grant number 82260604), the National Natural Science Foundation of Jiangxi Province (grant number 20192 BAB205053), both awarded to Jinghua Zhong, and was supported by an Jiangxi Cancer Hospital scientific research open fund project (KFJJ2023YB21), awarded to Changfei Gong. Author Contribution： Minfeng Huang: Data curation, methodology, project administration, writing – original draft. Changfei Gong: Data curation, methodology, project administration, writing – original draft. Biaoshui Liu: analysis, investigation Yili Wang: Investigation, project administration Jinghua Zhong: Data curation, formal analysis Min Wang:Data curation, and investigation. Jun Chi: Formal analysis, investigation Xinyuan Li: Formal analysis, investigation Ruilian Xie: Conceptualization, formal analysis, writing – review, and editing. Chunbo Tang: Conceptualization, formal analysis, writing – review, and editing. Minfeng Huang and Changfei Gong have contributed equally to this work Ruilian Xie and Chunbo Tang are co-corresponding authors. Conflict of Interest: Studies involving animal subjects Generated Statement: No animal studies are presented in this manuscript. Inclusion of identifiable human data Generated Statement: No potentially identifiable human images or data is presented in this study Ethics statement The use of patient data was approved by the institutional review board at the First Affiliated Hospital of Gannan Medical University with Ethics review number of LLSC-2024-133. REFERENCES Sung H, Ferlay J, Siegel R L, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries[J]. CA: A Cancer Journal for Clinicians, 2021, 71(3): 209-249. Dizon D S, Mackay H J, Thomas G M, et al. State of the science in cervical cancer: where we are today and where we need to go[J]. Cancer, 2014, 120(15): 2282-2288. Powell M E B. Modern radiotherapy and cervical cancer[J]. International Journal of Gynecological Cancer: Official Journal of the International Gynecological Cancer Society, 2010, 20(11 Suppl 2): S49-51. Chargari C, Peignaux K, Escande A, et al. Radiotherapy of cervical cancer[J]. Cancer/Radiothérapie, 2022, 26(1-2): 298-308. Miguel-Chumacero E, Currie G, Johnston A, et al. Effectiveness of Multi-Criteria Optimization-based Trade-Off exploration in combination with RapidPlan for head & neck radiotherapy planning[J]. Radiation Oncology (London, England), 2018, 13(1): 229. Citrin D E. Recent Developments in Radiotherapy[J]. The New England Journal of Medicine, 2017, 377(11): 1065-1075. Hussein M, South C P, Barry M A, et al. Clinical validation and benchmarking of knowledge-based IMRT and VMAT treatment planning in pelvic anatomy[J]. Radiotherapy and Oncology: Journal of the European Society for Therapeutic Radiology and Oncology, 2016, 120(3): 473-479. Loiselle C, Koh W-J. The emerging use of IMRT for treatment of cervical cancer[J]. Journal of the National Comprehensive Cancer Network: JNCCN, 2010, 8(12): 1425-1434. Anonymous. Cervical cancer[J]. The Lancet, 2019, 393(10167): 169-182. Yuan L, Ge Y, Lee W R, et al. Quantitative analysis of the factors which affect the interpatient organ-at-risk dose sparing variation in IMRT plans[J]. Medical Physics, 2012, 39(11): 6868-6878. Nelms B E, Robinson G, Markham J, et al. Variation in external beam treatment plan quality: An inter-institutional study of planners and planning systems[J]. Practical Radiation Oncology, 2012, 2(4): 296-305. Moore K L, Brame R S, Low D A, et al. Quantitative metrics for assessing plan quality[J]. Seminars in Radiation Oncology, 2012, 22(1): 62-69. Tinoco M, Waga E, Tran K, et al. RapidPlan development of VMAT plans for cervical cancer patients in low- and middle-income countries[J]. Medical Dosimetry, 2020, 45(2): 172-178. Chua L M H, Pang E P P, Master Z, et al. Dosimetric comparison of RapidPlan and manually optimised volumetric modulated arc therapy plans in prostate cancer[J]. Journal of Radiotherapy in Practice, 2021, 20(3): 257-264. Yu S, Xu H, Sinclair A, et al. Dosimetric and planning efficiency comparison for lung SBRT: CyberKnife vs VMAT vs knowledge-based VMAT[J]. Medical Dosimetry, 2020, 45(4): 346-351. D G, J L, Wr L, et al. A knowledge-based approach to improving and homogenizing intensity modulated radiation therapy planning quality among treatment centers: an example application to prostate cancer planning[J/OL]. International journal of radiation oncology, biology, Physics, 2013, 87(1)[2024-03-31]. Wu B, McNutt T, Zahurak M, et al. Fully automated simultaneous integrated boosted-intensity modulated radiation therapy treatment planning is feasible for head-and-neck cancer: a prospective clinical study[J]. International Journal of Radiation Oncology, Biology, Physics, 2012, 84(5): e647-653. Yuan L, Wu Q J, Yin F-F, et al. Incorporating single-side sparing in models for predicting parotid dose sparing in head and neck IMRT[J]. Medical Physics, 2014, 41(2): 021728. Yu G, Li Y, Feng Z, et al. Knowledge-based IMRT planning for individual liver cancer patients using a novel specific model[J]. Radiation Oncology (London, England), 2018, 13(1): 52. Xu Y, Cyriac J, De Ornelas M, et al. Knowledge-Based Planning for Robustly Optimized Intensity-Modulated Proton Therapy of Head and Neck Cancer Patients[J]. Frontiers in Oncology, 2021, 11: 737901. Gay H A, Barthold H J, O’Meara E, et al. Pelvic normal tissue contouring guidelines for radiation therapy: a Radiation Therapy Oncology Group consensus panel atlas[J]. International Journal of Radiation Oncology, Biology, Physics, 2012, 83(3): e353-362. Way H. Eclipse Photon and Electron Algorithms Reference[J]. [no date]. Mattes M D, Lee J C, Elnaiem S, et al. A predictive model to guide management of the overlap region between target volume and organs at risk in prostate cancer volumetric modulated arc therapy[J]. Radiation Oncology Journal, 2014, 32(1): 23-30. O’Toole J, Wu K, Bromley R, et al. Parotid sparing in RapidPlan Oropharynx models: To split or not to split[J]. Journal of Medical Radiation Sciences, 2020, 67(1): 80-86. Bratengeier K, Oechsner M, Gainey M, et al. Remarks on reporting and recording consistent with the ICRU reference dose[J]. Radiation Oncology (London, England), 2009, 4: 44. Shang H, Pu Y, Wang W, et al. Evaluation of plan quality and robustness of IMPT and helical IMRT for cervical cancer[J]. Radiation Oncology (London, England), 2020, 15(1): 34. Wang M, Gu H, Hu J, et al. Evaluation of a highly refined prediction model in knowledge-based volumetric modulated arc therapy planning for cervical cancer[J]. Radiation Oncology (London, England), 2021, 16(1): 58. Kaur H, Thakur N, Sharma R, et al. Dosimetric comparison between carotid-sparing IMRT and 3DCRT in early glottic cancer patients treated with definitive radiation therapy[J]. Journal of Cancer Research and Therapeutics, 2024, 20(1): 327-332. Li N, Carmona R, Sirak I, et al. Highly Efficient Training, Refinement, and Validation of a Knowledge-based Planning Quality-Control System for Radiation Therapy Clinical Trials[J]. International Journal of Radiation Oncology, Biology, Physics, 2017, 97(1): 164-172. Bi S, Sun X, Sohaimi W F B W, et al. Study on the transferability of the knowledge-based VMAT model to predict IMRT plans in prostate cancer radiotherapy[J]. European Journal of Medical Research, 2023, 28(1): 309. Franceschini D, Cozzi L, Fogliata A, et al. Training and validation of a knowledge-based dose-volume histogram predictive model in the optimisation of intensity-modulated proton and volumetric modulated arc photon plans for pleural mesothelioma patients[J]. Radiation Oncology (London, England), 2022, 17(1): 150. Hussein M, South C P, Barry M A, et al. Clinical validation and benchmarking of knowledge-based IMRT and VMAT treatment planning in pelvic anatomy[J]. Radiotherapy and Oncology: Journal of the European Society for Therapeutic Radiology and Oncology, 2016, 120(3): 473-479. Fogliata A, Reggiori G, Stravato A, et al. RapidPlan head and neck model: the objectives and possible clinical benefit[J]. Radiation Oncology (London, England), 2017, 12(1): 73. Thoma A, Farrokhyar F, McKnight L, Bhandari M. How to optimize patient recruitment. Can J Surg. 2010;53(3):205–10. François B, Clavel M, Vignon P, Laterre PF. Perspective on optimizing clinical trials in critical care: how to puzzle out recurrent failures. J Intensive Care. 2016;4(1):1–11. Cite Share Download PDF Status: Published Journal Publication published 29 Apr, 2026 Read the published version in Physical and Engineering Sciences in Medicine → Version 1 posted Editor assigned by journal 16 May, 2025 Reviewers agreed at journal 10 Apr, 2025 Reviewers invited by journal 10 Apr, 2025 Editor invited by journal 05 Apr, 2025 First submitted to journal 04 Apr, 2025 Editorial decision: Accept 31 Mar, 2025 Editorial decision: Minor revisions 31 Mar, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5971123","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":441097470,"identity":"19a075f6-d2b8-46bd-9674-9b3be9961ae0","order_by":0,"name":"Minfeng Huang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0ElEQVRIiWNgGAWjYBACNvnDxz8k/LBh5mdmPkCcFj4JtjSGjz1p7JLtbQnEaZGT4FFjnMF2iN/gzBkDIh0m3cP2mIfngLTkjJyPN94w2MnpNhDSInP2uDGPxR1jfonczZZzGJKNzQ4Q0sKQlyDNw/MsWXJG7jZpHoYDidsIa8kxkOZhO1y/4UbOMyK1SOSYSc5gO8wM9D4bkVp4jiUbAAOZGRjIxpZzDIjwi3x788EH0Kh8eONNhZ0cQS0oQIKHyKhB1kKqjlEwCkbBKBgRAACeg0ESq9+q9QAAAABJRU5ErkJggg==","orcid":"","institution":"First Clinical Medical College, Gannan Medical University, Ganzhou, Jiangxi Province, China; Department of Oncology, First Affiliated Hospital of Gannan Medical University","correspondingAuthor":true,"prefix":"","firstName":"Minfeng","middleName":"","lastName":"Huang","suffix":""},{"id":441097471,"identity":"885fe2a0-c755-4fe9-9aee-b2d43388c8e2","order_by":1,"name":"Changfei Gong","email":"","orcid":"","institution":"Department of Radiation Oncology, Jiangxi Province Cancer Hospital, Nanchang, Jiangxi, China","correspondingAuthor":false,"prefix":"","firstName":"Changfei","middleName":"","lastName":"Gong","suffix":""},{"id":441097472,"identity":"77f69314-9541-43ac-aa68-1436171dfea4","order_by":2,"name":"Biaoshui Liu","email":"","orcid":"","institution":"Department of Radiation Oncology, Sun Yat-sen University Cancer Center, Guangzhou, Guangdong, China","correspondingAuthor":false,"prefix":"","firstName":"Biaoshui","middleName":"","lastName":"Liu","suffix":""},{"id":441097473,"identity":"13824496-f058-4e51-a372-7a248b05d3c8","order_by":3,"name":"Yili Wang","email":"","orcid":"","institution":"Department of Oncology, First Affiliated Hospital of Gannan Medical University, Ganzhou, Jiangxi, China; Jiangxi Clinical Research Center for Cancer, Ganzhou, Jiangxi, China","correspondingAuthor":false,"prefix":"","firstName":"Yili","middleName":"","lastName":"Wang","suffix":""},{"id":441097474,"identity":"a2882e19-c08e-4a4c-b07e-f2457e658945","order_by":4,"name":"Jinghua Zhong","email":"","orcid":"","institution":"Department of Oncology, First Affiliated Hospital of Gannan Medical University, Ganzhou, Jiangxi, China; Jiangxi Clinical Research Center for Cancer, Ganzhou, Jiangxi, China","correspondingAuthor":false,"prefix":"","firstName":"Jinghua","middleName":"","lastName":"Zhong","suffix":""},{"id":441097475,"identity":"570510c6-8c8f-4c19-9152-98a269ae0cb0","order_by":5,"name":"Min Wang","email":"","orcid":"","institution":"First Clinical Medical College, Gannan Medical University, Ganzhou, Jiangxi Province, China; Department of Oncology, First Affiliated Hospital of Gannan Medical University, Ganzhou, Jiangxi, China","correspondingAuthor":false,"prefix":"","firstName":"Min","middleName":"","lastName":"Wang","suffix":""},{"id":441097476,"identity":"96570c81-a5c2-46a9-b044-013f91eb2730","order_by":6,"name":"Jun Chi","email":"","orcid":"","institution":"Department of Oncology, First Affiliated Hospital of Gannan Medical University, Ganzhou, Jiangxi, China; Jiangxi Clinical Research Center for Cancer, Ganzhou, Jiangxi, China","correspondingAuthor":false,"prefix":"","firstName":"Jun","middleName":"","lastName":"Chi","suffix":""},{"id":441097477,"identity":"f650f613-439f-41a3-955f-4f44dd327db1","order_by":7,"name":"Li Xinyuan","email":"","orcid":"","institution":"First Clinical Medical College, Gannan Medical University, Ganzhou, Jiangxi Province, China","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Xinyuan","suffix":""},{"id":441097478,"identity":"983e356c-f668-4667-87b8-75848e53b777","order_by":8,"name":"Ruilian Xie","email":"","orcid":"","institution":"Department of Oncology, First Affiliated Hospital of Gannan Medical University, Ganzhou, Jiangxi, China; Jiangxi Clinical Research Center for Cancer, Ganzhou, Jiangxi, China","correspondingAuthor":false,"prefix":"","firstName":"Ruilian","middleName":"","lastName":"Xie","suffix":""},{"id":441097479,"identity":"7f5d7d34-eab2-4a82-99ec-5346d129cd0a","order_by":9,"name":"Chunbo Tang","email":"","orcid":"","institution":"Jiangxi Clinical Research Center for Cancer, Ganzhou, Jiangxi, China; Department of Radiation Oncology, First Affiliated Hospital of Gannan Medical University, Ganzhou, Jiangxi, China","correspondingAuthor":false,"prefix":"","firstName":"Chunbo","middleName":"","lastName":"Tang","suffix":""}],"badges":[],"createdAt":"2025-02-06 07:47:43","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5971123/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5971123/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s13246-026-01732-4","type":"published","date":"2026-04-29T15:57:56+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":80580580,"identity":"24699e17-5d05-4738-a35b-55fef65160d5","added_by":"auto","created_at":"2025-04-14 23:16:04","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1013631,"visible":true,"origin":"","legend":"\u003cp\u003eThe process of RapidPlan model construction and the validation of automated plans.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5971123/v1/e9e7ae27e1958c83b7981e41.png"},{"id":80579333,"identity":"0caa7aaf-9c26-497d-b2d3-1ff9945b66f0","added_by":"auto","created_at":"2025-04-14 23:08:04","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":275299,"visible":true,"origin":"","legend":"\u003cp\u003eThe hot spot region (V105%) of the bladder and bowel for each patient in the SVR≤0.40 and SVR＞0.40 sets was plotted, showing changes with different verification plans generated by the RapidPlan model. Abbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5971123/v1/504e5809fe7d72437860948d.png"},{"id":80579332,"identity":"5da4b975-1e6b-45f1-8959-857180e63235","added_by":"auto","created_at":"2025-04-14 23:08:04","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":309874,"visible":true,"origin":"","legend":"\u003cp\u003eThe hot spot region (V105%) of the Bladder\u003csub\u003eoverlap\u003c/sub\u003e and Bowelbag\u003csub\u003eoverlap\u003c/sub\u003e for each patient in the SVR≤0.40 and SVR＞0.40 sets was plotted, showing changes with different verification plans generated by the RapidPlan model.\u0026nbsp; Abbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-5971123/v1/aaebb9b6cb98bcaea98db85a.png"},{"id":80579338,"identity":"86223851-72e5-41f0-bdd7-45c6c78c4283","added_by":"auto","created_at":"2025-04-14 23:08:04","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":717415,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of the dose distribution for an example case with SVR≤0.40. Abbreviations: SVR=Sum of the Volume Ratios.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-5971123/v1/2e9e2a457534fb50330dcb1d.png"},{"id":80581895,"identity":"c5be5a8f-1019-4e05-add0-367bdaac746f","added_by":"auto","created_at":"2025-04-14 23:24:04","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1071516,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of DVHs for an example case with the SVR≤0.40. \u0026nbsp;Abbreviations: SVR=Sum of the Volume Ratios.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-5971123/v1/bca5573bac7aff0ff6685e10.png"},{"id":108437685,"identity":"dd344f54-810b-42bd-8784-721455b9a6b6","added_by":"auto","created_at":"2026-05-04 16:02:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3646387,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5971123/v1/115daa3d-389c-4ff3-8e6b-94da43891b0d.pdf"}],"financialInterests":"","formattedTitle":"Exploration of overlap volumes-based IMRT Rapidplan for cervical cancer patients with the aim of OARs sparing","fulltext":[{"header":"1. INTRODUCTION","content":"\u003cp\u003eCervical cancer ranks as the fourth most common malignancy among women globally, posing significant challenges to global health [1]. Treatment for cervical cancer typically involves multimodal approaches, with curative options primarily encompassing surgery and radiation therapy [2]. Radiation therapy stands as a key modality in the management of cervical cancer [3,4]. the quality of radiation therapy heavily relies on treatment planning, with one challenge being how to ensure adequate target dose delivery while sparing healthy tissues [5].\u003c/p\u003e\n\u003cp\u003eIntensity-modulated radiation therapy (IMRT) employs inverse planning to adjust the dose distribution within the target area, aiming to enhance dose uniformity within the target while minimizing doses to organs at risk (OARs) [6,7]. Currently, this technique is widely used in treating cervical cancer [8], contributing to the reduction of treatment-related side effects in patients with locally advanced cervical cancer [9]. However, due to the substantial overlap between OARs and the planning target volume (PTV), not only does it increase the complexity of planning design, but it also significantly affects the quality of IMRT plans [10]. Despite physicians typically prioritizing the protection of these organs, variations in clinical practices among different oncologists regarding planning requirements and reviews, as well as differences among planning optimization personnel, may ultimately lead to discrepancies between plans.\u003c/p\u003e\n\u003cp\u003eTraditional IMRT techniques achieve optimal dose distribution through heuristic or iterative methods, heavily reliant on staff experience, resulting in lower inter-plan consistency [11,12]. Knowledge-based RapidPlan is a semi-automatic technique that extracts relevant knowledge from existing treatment plans to generate high-quality plans [13], improving both plan quality and workflow efficiency [14, 15]. Within the industry, creating RapidPlan models is typically categorized by disease site, constructing universal models to generate high-quality plans that pass clinical evaluation [16–18]. However, a minority of researchers guide model construction within a single disease site based on specific conditions, or use general models to predict doses for categorized cases. In the study by Yu et al. [19], a special model was constructed based on conditions where the distance from the PTV to the right kidney was less than 3cm, demonstrating that specific models compared to general models can improve the conformity of liver cancer target areas while reducing doses to OARs. In a study by Yihang Xu et al. [20], universal models were used to predict doses for unilateral and bilateral cases of advanced head and neck (HN) cancer patients, showing different capabilities in protecting OARs.\u003c/p\u003e\n\u003cp\u003eIn this study, we explored model diversification by developing three automatic planning models for cervical cancer. This involved generating the original Unclassified Model-O based on 80 cases. Simultaneously, the sum of ratio (SVR): \u0026nbsp;bladder volume overlapping with PTV to total bladder volume, and the ratio of bowelbag volume overlapping with PTV to total bowelbag volume, with a threshold value of 0.4 was used for classification. We then created two new Reclassification Models: Model-S (SVR ≤ 0.4) and Model-L (SVR \u0026gt; 0.4), using the remaining 30 cases for validation. Our aim is to analyze whether constructing different RapidPlan Models under specific conditions, especially in cases with significant overlap between the PTV and OARs, could better protect the overlapping region while ensuring high-quality target coverage.\u003c/p\u003e"},{"header":"2. METHODS","content":"\u003cp\u003e\u003cstrong\u003e2.1. Clinical planning\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe retrospectively selected a cohort of 110 patients with stage IB1 to IIIC2 cervical cancer who underwent dynamic IMRT treatment. All patients have signed an informed consent for radiotherapy, The radiotherapy data were anonymized and approved by our hospital\u0026apos;s ethics committee with the number LLSC-2024-133. Patients were prescribed a dose of 45 Gy, delivered at 1.8 Gy per day over 5 weeks. All plans were designed using the Eclipse 15.2 treatment planning system and utilized 7 coplanar fields (0\u0026deg;, 52\u0026deg;, 105\u0026deg;, 156\u0026deg;, 208\u0026deg;, 260\u0026deg;, 312\u0026deg;). Plan optimization and dose calculations were performed using the photon algorithm (PVO) version 15.5.12 and the anisotropic analytical algorithm (AAA) version 15.5.12, with a calculation grid of 2.5 mm. CT slices were 2.5 mm thick and obtained using the GE Discovery CT590. Target volumes were delineated by experienced senior oncologists following RTOG international guidelines and reviewed by another senior oncologist.\u0026nbsp;Organs at risk (OARs), including the bowel bag, bladder, and rectum, were automatically contoured using AccuContour (Manteia Technologies, China) to minimize inter-observer variability, and then reviewed and adjusted (if necessary) by two senior radiation oncologists to ensure accuracy.\u0026nbsp;The Contour of the Rectum: The lower border is the anal margin, which is marked with an opaque marker during simulated positioning. The rectum is no longer circular from the upper edge to the horizontal position and is connected to the sigmoid colon in the forward direction.The Contouring of the Bowel Bag: It starts from the bottom of the small intestine or large intestine loop, or from the anorectum, whichever is the lowest. If the rectum or anorectal portion appears in the axial section following the pattern of the intestinal loop, it should be included as part of the bag; otherwise, it should be excluded.Tip: The abdominal content outline excludes muscles and bones. [21].\u0026nbsp;All plans aimed to deliver effective doses to OARs overlapping with the PTV while minimizing dose to individual OARs, with at least 95% of the PTV receiving the prescribed dose. The original clinical plans were manually optimized by experienced physicists with ample experience in cervical cancer treatment, ensuring that each IMRT plan met clinical protocols.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2. Model Building, Training, and Reclassification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn this study, we utilized a PTV and three OARs\u0026mdash;bladder, bowelbag, and rectum\u0026mdash;to develop the RapidPlan DVH estimation model. Initially, we included 80 clinically approved IMRT plans for cervical cancer from our institution in the training dataset. The DVH estimation modeling engine processed the dataset for each patient, correlating dose-volume histogram information with the geometric characteristics of the respective OARs [22]. Subsequently, the RapidPlan modeling procedure was applied to these initial 80 plans, resulting in the creation of the original Unreclassification model (Model-O).\u003c/p\u003e\n\u003cp\u003ePrevious research has indicated that the volume of individual cases\u0026apos; OARs overlapping with the PTV is a crucial anatomical factor influencing OARs dose sparing [23]. Additionally, several scholars have demonstrated that the overall quality of training plans can significantly influence the outcomes produced by the model [10]. Therefore, selecting a case classification method based on the extent of overlap between different organs and the PTV is critical, especially in cervical cancer patients, where there may be substantial overlap between the bowelbag and bladder regions with the PTV. Research by O\u0026apos;Toole et al. suggests that combining the sparing of OARs yields greater advantages compared to independent optimization [24]. Consequently, we utilize the ratio of the bladder\u0026apos;s overlap volume with the PTV (Bladder\u003csub\u003eoverlap\u003c/sub\u003e) to its original volume (V\u003csub\u003ebladder\u003c/sub\u003e), and the ratio of the bowelbag\u0026apos;s overlap volume with the PTV (Bowelbag\u003csub\u003eoverlap\u003c/sub\u003e)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eto its original volume (V\u003csub\u003ebowelbag\u003c/sub\u003e), summed together as the Sum of these Volume Ratios (SVR), as the criteria for grouping. This is represented by Equation 1:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/p\u003e\n\u003cp\u003eConsidering the number of cases we trained with, our goal was to maintain consistency in the sample size for training the RapidPlan model. Therefore, we divided the 80 patients into two groups, using the median of the SVR from Equation 1, which is 0.4, as the threshold for grouping. Among these, 40 original plans with SVR \u0026le; 0.40 were used to construct Reclassification Model-S, while Reclassification Model-L was trained using the remaining 40 original plans with SVR \u0026gt; 0.40.One advantage of this approach is that by consolidating the modeling of several OAR DVHs within the training plans, the model can achieve a comprehensive balance in predicting dose protection among OARs, resulting in more balanced DVH estimates. However, a concern arises under specific patient requirements (such as mandatory targets for critical organs): the necessary trade-offs in combining OAR DVH modeling may lead to the prediction of unattainable DVH estimates. In such cases, manually setting fixed priorities for these special OARs during model construction may provide a more reliable solution\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.3.\u003c/strong\u003e \u003cstrong\u003eClassification of training samples and automatic generation of plans\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe performance of the DVH estimation models in automatically generating plans was validated under reclassification scenarios. The remaining 30 cases of cervical cancer were divided into two groups based on Equation 1: SVR \u0026le; 0.4 and SVR \u0026gt; 0.4, without any additional specific criteria applied. Both groups of test cases were assigned the three models mentioned earlier. Plans configured with the Unclassification Model-O were denoted as UMO, those configured with Reclassification Model-S as RMS, and those with Reclassification Model-L as RML. When using the RapidPlan model for automatic plan generation, efforts were made not only to ensure the model\u0026apos;s capability to generate automated plans but also to consider multiple criteria for OAR/PTV during the optimization process. However, determining the optimal combination of specific details such as OAR targets and priorities, especially in light of the patient\u0026apos;s pathological characteristics, is not straightforward. The process of constructing the RapidPlan model and validating automated plans is illustrated in Figure 1. The primary steps involved in the models used were as follows: (1) A model has been selected. (2) Manually aligned the structure of the validation plan with that of the model. (3) Utilizing a sophisticated regression model, we generated an estimation range for dose-volume histogram (DVH) constraints and automatically established the corresponding dose-volume limits. (4) Finally, we optimized the verification plan.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.4. Analysis and validation of plans\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe validation of the evaluation set were completed by comparing the PTV and OAR DVH results among RMS, RML, and the original plan UMO. Criteria for assessing the quality of radiation therapy plans for cervical cancer were established based on dose constraints outlined in the International Commission on Radiation Units and Measurements (ICRU) Report 83, as well as recommendations from the GEC-ESTRO Gynecological Working Group [25\u0026ndash;27], taking into account specific dose requirements for cervical cancer at our center. Key dosimetric evaluation indices for the target volume (PTV) include minimum dose (Dmin), maximum dose (Dmax), mean dose (Dmean), D2%, D50%, D98%, V45, and V47.25. Conformity Index (CI) and Homogeneity Index (HI) are important metrics used to assess target dose. The calculation method for CI is as follows [28]:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eA Homogeneity Index (HI) closer to 0 indicates better homogeneity within the target volume.\u003c/p\u003e\n\u003cp\u003eThe irradiation dose to organs at risk should be kept as low as reasonably achievable, considering their maximum tolerated doses, while ensuring that the overlap regions between organs at risk and target volumes meet clinical dosimetric criteria. The primary dosimetric evaluation parameters for organs at risk are as follows: bowel V40, V45 (V100%), and V47.25 (V105%); bladder V40, V45, and V47.25. Additionally, for organs at risk overlapping with targets, the primary dosimetric evaluation parameters include bowelbag overlap of V40, V45, and V47.25; bladder overlap of V40, V45, and V47.25.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5. Statistical Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePlan data generated by the three automatic planning models were compared between groups using SPSS 26.0 software. The Wilcoxon signed-rank test was conducted on the datasets, with statistical significance defined as P \u0026lt; 0.05.\u003c/p\u003e"},{"header":"3. RESULTS","content":"\u003cp\u003eIn the study of 110 cases, the SVR ranged from [0.148\u0026ndash;0.961] in 80 training cases, with a mean and standard deviation of 0.442 \u0026plusmn; 0.16. For 30 training cases, the SVR ranged from [0.166\u0026ndash;0.827], with a mean and standard deviation of 0.438 \u0026plusmn; 0.17. Detailed patient characteristics are provided in Table 1.\u003c/p\u003e\n\u003cp\u003eWe compared automatic plan quality differences using models with different configurations across two sets of validation cases. Table 2 shows PTV dose metrics for 15 patients with SVR \u0026le; 0.40, revealing no significant differences among the three plans for Dmin, Dmax, and D2% (p \u0026gt; 0.05). However, the Conformity Index (CI) of RMS differed significantly from UMO (p = 0.015). In terms of Dmean, RMS closely aligned with the prescription dose compared to UMO and RML (p = 0.006, p = 0.044, respectively). The D98% of RMS was significantly lower than RML (p = 0.033). RMS exhibited a smaller V45 compared to UMO by 0.22 (p = 0.007) and compared to RML by 0.24 (p = 0.038), remaining at 95.53%. The high-dose region V47.25 (V105%) of RMS was significantly lower than UMO (p = 0.007), showing a decrease of 1.19 percentage points. Although not significant (p = 0.0965), RMS\u0026apos;s V47.25 was 1.08 percentage points lower than RML. \u0026nbsp;Table 3 presents PTV dose metrics for automated plans in patients with SVR \u0026gt; 0.40. For V45, RML was significantly higher than UMO and RMS (p = 0.006, p = 0.001, respectively), with RMS averaging only 94.9%, indicating that many RMS plans fail to meet clinical requirements. RML\u0026apos;s D98% and CI were slightly inferior to UMO, these differences were not statistically significant. Meanwhile, RML outperformed UMO significantly in Dmax, Dmean, D2%, D50%, V45, V47.25 (V105%), and Homogeneity Index (HI) (p = 0.047, p = 0.013, p = 0.003, p = 0.019, p = 0.06, p = 0.002, p = 0.002, respectively). Notably, RML\u0026apos;s V47.25 was 1.58 percentage points lower than UMO.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1 Characteristics of the 110 patients enrolled in this study\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"418\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eCharacteristic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003en\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNumber of patients\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e110\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge (years)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e55.65\u0026plusmn;10.63\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e56[25-86]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePathological Type and Staging\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSquamous Cell Carcinoma\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e85\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAdenocarcinoma\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e17\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAdenosquamous Carcinoma\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eOther\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eStaging\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eIB1~IIIC2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrescription dose (Gy)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e45\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSum of the volume ratios\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e80 Initial Patients\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.442\u0026plusmn;0.16\u003c/p\u003e\n \u003cp\u003e0.400[0.148-0.961]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e30 Validation Patients\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.438\u0026plusmn;0.17\u003c/p\u003e\n \u003cp\u003e0.400[0.166-0.827]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eAbbreviations: SD =Standard deviation.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2 The PTV\u0026rsquo;s DVH metrics for UMO, RMS and RML with SVR\u0026le;0.40\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"568\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eParameter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003eUMO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eRMS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eRML\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 567px;\"\u003e\n \u003cp\u003eDmin[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e39.19\u0026plusmn;0.71\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e38.5\u0026plusmn;2.14\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e38.51\u0026plusmn;2.14\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.307;ii=0.156;iii=0.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e39.21[37.43\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e-40.56]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.35[31.63-40.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.06[31.31-40.49]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 567px;\"\u003e\n \u003cp\u003eDmax[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e49.32\u0026plusmn;0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e49.34\u0026plusmn;0.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e49.36\u0026plusmn;0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=1;ii=0.379;iii=0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e49.26[48.71-50.84]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e49.31[48.79-50.57]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e49.25[48.6-50.57]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 567px;\"\u003e\n \u003cp\u003eDmean[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e46.47\u0026plusmn;0.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.45\u0026plusmn;0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.48\u0026plusmn;0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.006;Ii=0.533;iii=0.044\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e46.44[46.33-46.63]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.42[46.32-46.62]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.44[46.34-46.75]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 567px;\"\u003e\n \u003cp\u003eD2%[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e47.83\u0026plusmn;0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e47.81\u0026plusmn;0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e47.82\u0026plusmn;0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.131;ii=0.561;iii=0.925\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e47.83[47.67-48.03]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e47.84[47.66-47.93]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e47.8[47.67-48.08]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 567px;\"\u003e\n \u003cp\u003eD50%[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e46.5\u0026plusmn;0.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.47\u0026plusmn;0.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.51\u0026plusmn;0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.004;ii=0.329;iii=0.012\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e46.46[46.39-46.69]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.46[46.37-46.67]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.48[46.4-46.82]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 567px;\"\u003e\n \u003cp\u003eD98%[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e44.54\u0026plusmn;0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e44.52\u0026plusmn;0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e44.55\u0026plusmn;0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.052;ii=0.381;iii=0.033\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e44.51[44.32-44.76]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e44.53[44.3-44.68]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e44.55[44.34-44.74]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 567px;\"\u003e\n \u003cp\u003eV45[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e95.75\u0026plusmn;0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.53\u0026plusmn;0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.77\u0026plusmn;0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.007;ii=0.505;iii=0.038\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e95.7[94.1-96.9]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.7[93.7-96.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.7[94.2-97]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 1px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" valign=\"top\" style=\"width: 568px;\"\u003e\n \u003cp\u003eV47.25[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e17.25\u0026plusmn;3.87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e16.06\u0026plusmn;3.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e17.14\u0026plusmn;4.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" rowspan=\"2\" valign=\"top\" style=\"width: 114px;\"\u003e\n \u003cp\u003ei=0.007;ii=0.798;iii=0.096\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e16[12.4-24.2]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e15.2[10.9-21.9]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e16.1[11.5-30.3]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" valign=\"top\" style=\"width: 568px;\"\u003e\n \u003cp\u003eCI\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.997\u0026plusmn;0.016\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.993\u0026plusmn;0.014\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.996\u0026plusmn;0.016\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" rowspan=\"2\" valign=\"top\" style=\"width: 114px;\"\u003e\n \u003cp\u003ei=0.015;ii=0.57;iii=0.069\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.994[0.972-1.023]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.989[0.968-1.019]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.994[0.973-1.028]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" valign=\"top\" style=\"width: 568px;\"\u003e\n \u003cp\u003eHI\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.071\u0026plusmn;0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.071\u0026plusmn;0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.070\u0026plusmn;0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" rowspan=\"2\" valign=\"top\" style=\"width: 114px;\"\u003e\n \u003cp\u003ei=0.733;ii=0.069;iii=0.011\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 112px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.07[0.063-0.076]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.071[0.064-0.075]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.07[0.064-0.075]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" valign=\"top\" style=\"width: 568px;\"\u003e\n \u003cp\u003eAbbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P\u0026lt;0.05 indicates a statistical significance.SD =Standard deviation.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3 The PTV\u0026rsquo;s DVH metrics for UMO, RMS and RML with SVR\u0026gt;0.40\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"571\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eParameter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eUMO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eRMS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003eRML\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 571px;\"\u003e\n \u003cp\u003eDmin[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.27\u0026plusmn;0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.04\u0026plusmn;0.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e39.33\u0026plusmn;0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.017;ii=0.733;iii=0.011\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.28[38-40.81]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e38.95[37.29-40.96]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e39.37[37.83-40.77]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 571px;\"\u003e\n \u003cp\u003eDmax[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e49.25\u0026plusmn;0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e49.15\u0026plusmn;0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e49.17\u0026plusmn;0.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.055;ii=0.047;iii=0.851\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e49.26[48.58-50.49]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e49.07[48.55-50.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e49.08[48.62-50.31]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 571px;\"\u003e\n \u003cp\u003eDmean[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.48\u0026plusmn;0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.43\u0026plusmn;0.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e46.46\u0026plusmn;0.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.003;ii=0.013;iii=0.014\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.47[46.32-46.71]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.42[46.31-46.67]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e46.45[46.29-46.72]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 571px;\"\u003e\n \u003cp\u003eD2%[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e47.87\u0026plusmn;0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e47.82\u0026plusmn;0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e47.82\u0026plusmn;0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.017;ii=0.003;iii=0.889\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e47.81[47.68-48.35]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e47.76[47.65-48.31]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e47.76[47.64-48.39]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 571px;\"\u003e\n \u003cp\u003eD50%[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.5\u0026plusmn;0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.44\u0026plusmn;0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e46.48\u0026plusmn;0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.002;ii=0.019;iii=0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.48[46.33-46.73]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.44[46.31-46.68]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e46.47[46.3-46.73]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 571px;\"\u003e\n \u003cp\u003eD98%[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e44.54\u0026plusmn;0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e44.49\u0026plusmn;0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e44.54\u0026plusmn;0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.011;ii=0.899;iii=0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e44.54[44.38-44.73]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e44.5[44.35-44.61]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e44.54[44.37-44.75]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 571px;\"\u003e\n \u003cp\u003eV45[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.02\u0026plusmn;2.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e94.9\u0026plusmn;2.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e95.33\u0026plusmn;2.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.137;ii=0.006;iii=0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.5[86.7-96.8]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.3 [87-96.2]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e95.7 [88.4-97.1]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 571px;\"\u003e\n \u003cp\u003eV47.25[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e17.97\u0026plusmn;4.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e15.97\u0026plusmn;4.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e16.39\u0026plusmn;4.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.004;ii=0.002;iii=0.293\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e17[11.8-29.1]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e15.1[10.1-27.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e15.6[10.8-29.1]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 571px;\"\u003e\n \u003cp\u003eCI\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.99\u0026plusmn;0.011\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.986\u0026plusmn;0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e0.989\u0026plusmn;0.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.011;ii=0.069;iii=0.006\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.99[0.976-1.008]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.985[0.971-1.002]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e0.988[0.975-1.006]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 571px;\"\u003e\n \u003cp\u003eHI\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.071\u0026plusmn;0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.072\u0026plusmn;0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e0.071\u0026plusmn;0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003ei=0.307;ii=0.002;iii=0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.071[0.065-0.081]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.072[0.066-0.081]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 120px;\"\u003e\n \u003cp\u003e0.07[0.064-0.082]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 571px;\"\u003e\n \u003cp\u003eAbbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P\u0026lt;0.05 indicates a statistical significance.SD =Standard deviation.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eWhen the SVR was \u0026le;0.40 in the test group, dose variations in the Bladder and Bowelbag are shown in Table 4. For the Bladder, RML exhibited significantly higher Dmean (p=0.012) and V20 (p=0.004) compared to RMS, while V30 (p=0.826) and V45 (p=0.052) showed no significant differences between RMS and RML. There were no significant differences in Bladder dose between RMS and UMO (p \u0026gt; 0.05). RMS showed a decrease of 0.25 percentage points in the high-dose region V47.25 compared to UMO (p=0.011) and 0.33 percentage points compared to RML (p=0.011). For the Bowelbag, there were no significant differences in Dmean and V40Gy among UMO, RMS, and RML. However, V45% of RMS was significantly lower than UMO and RML (p=0.039, p=0.039, respectively). In terms of V47.25, RMS decreased by 0.15% compared to UMO (p=0.003) and by 0.19% compared to RML (p=0.001). For the Rectum, except for significantly lower Dmean in RMS compared to UMO and RML, and significantly lower V45% in UMO and RMS compared to RML, there were no significant differences in other parameters. Table 5 presents dose variations in the Bladder and Bowelbag when SVR was greater than 0.40. For the Bladder, there were no significant differences in Dmean, V20, and V30 among UMO, RMS, and RML. However, RMS had the lowest V45 (p=0.001, p=0.003, respectively) and also the lowest V47.25 (V105%), being only 2.80% (p=0.002, p=0.004, respectively). In the Bowelbag, although there were no significant differences in Dmean and V40Gy among UMO, RMS, and RML, RMS provided the lowest values. RMS had significantly lower V45 compared to RML and UMO (p=0.007, p=0.031, respectively). RMS\u0026apos;s V47.25 was only 1.52 compared to RML and UMO (p=0.001, p=0.001, respectively). For the Rectum, RMS had the lowest Dmean and V40, with only Dmean showing significant differences compared to RML (p=0.017). The V45 of RMS was 28.85, significantly lower than UMO and RML (p=0.033, p=0.005, respectively). Figure 2 illustrates changes in the Bladder and Bowelbag for each patient in the training and validation sets, focusing on the hotspot region (V105%) of protocol-specified DVH metrics in both reclassified and original plans.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 4 Bladder, BowelBag and Rectum DVH metrics for UMO, RMS and RML with SVR\u0026le;0.40\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"555\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eStructure/Parameter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eUMO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003eRMS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eRML\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBladder\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eDmean[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e33.04\u0026plusmn;1.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e33.12\u0026plusmn;1.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e33.23\u0026plusmn;1.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.256;ii=0.002;iii=0.012\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e32.89[30.26-35.7]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e33.01[30.25-35.74]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e33[30.27-35.86]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV20[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e87.44\u0026plusmn;6.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e87.39\u0026plusmn;5.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e88.31\u0026plusmn;6.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.802;iii=0.016;iii=0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e89.7[78-95.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e90.6[78-93.9]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e91.1[77.8-96]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV30[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e57.05\u0026plusmn;6.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e57.47\u0026plusmn;6.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e57.59\u0026plusmn;6.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.050;ii=0.008;iii=0.826\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e56.7[47.2-70.2]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e57.4[47.3-69.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e57.1[47.1-71]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV45[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e21.91\u0026plusmn;5.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e21.87\u0026plusmn;5.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e22.07\u0026plusmn;5.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.936;ii=0.088;iii=0.052\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e23.2[11.2-30.6]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e23.2[11.3-30.6]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e23.6[11.6-30.8]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV47.25[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.25\u0026plusmn;1.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e2.00\u0026plusmn;1.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e2.33\u0026plusmn;1.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.011;ii=0.243;iii=0.011\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.90[0-5.6]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e1.90[0-4]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e2.2[0-5.6]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBowelBag\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eDmean\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e21.83\u0026plusmn;3.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e21.89\u0026plusmn;3.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e21.85\u0026plusmn;3.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.047;ii=0.842;iii=0.272\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e22.59[15.76-28.09]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e22.6[15.7-28.07]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e22.53[15.67-28.05]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV40[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e13.17\u0026plusmn;4.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e13.2\u0026plusmn;4.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e13.19\u0026plusmn;4.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.272;ii=0.546;iii=0.726\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e12[6.5-20.9]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e12.1[6.5-20.9]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e12.2[6.5-20.9]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV45[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e8.35\u0026plusmn;3.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e8.29\u0026plusmn;3.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e8.35\u0026plusmn;3.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.039;ii=0.763;iii=0.039\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e7.1 [3-15.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e7.1[2.9-15.3]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e7.2[3-15.3]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV47.25[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.66\u0026plusmn;0.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e0.51\u0026plusmn;0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.70\u0026plusmn;0.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.003;ii=0.207;iii=0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e0.7[0-1.7]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e0.5[0-1.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.7[0.1-1.6]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRectum\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eDmean[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e36.42\u0026plusmn;3.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e36.09\u0026plusmn;3.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e36.45\u0026plusmn;3.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.036;ii=0.221;iii=0.011\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e36.68[25.36-41.84]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e36.37[23.34-41.3]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e36.78[25.3-41.39]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV40[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e53.35\u0026plusmn;12.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e53.45\u0026plusmn;13.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e53.63\u0026plusmn;12.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.865;ii=0.345;iii=0.609\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e55.1[17.6-73.6]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e56.5[15.8-73.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e56.5[18.8-73]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV45[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e20.43\u0026plusmn;8.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e19.77\u0026plusmn;9.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e22.59\u0026plusmn;9.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 101px;\"\u003e\n \u003cp\u003ei=0.155;ii=0.001;iii=0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e22.1[1-36.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e20.2[0.9-40.9]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e23.4[1.7-43.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eAbbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P\u0026lt;0.05 indicates a statistical significance.SD =Standard deviation.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003eTable 5 Bladder, BowelBag and Rectum DVH metrics for UMO, RMS and RML with SVR\u0026gt;0.40\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"555\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eStructure/Parameter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eUMO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eRMS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eRML\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBladder\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eDmean[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.06\u0026plusmn;1.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e38.98\u0026plusmn;1.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.09\u0026plusmn;1.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=0.139;ii=0.955;iii=0.094\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.78[36.96-42.25]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.66[36.79-42.31]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.63[36.92-42.23]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV20[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e97.26\u0026plusmn;3.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e97.13\u0026plusmn;3.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e97.31\u0026plusmn;3.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=0.099;ii=0.55;iii=0.133\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e98.3[89.7-100]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e98.8[89.2-100]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e98.9[89.9-100]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV30[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e80.64\u0026plusmn;6.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e80.61\u0026plusmn;7.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e80.8\u0026plusmn;6.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=0.95;ii=964;iii=0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e81.3[70.9-91.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e82.1[70.7-92.9]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e81.3[70.6-91.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV45 [%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e43.03\u0026plusmn;7.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e42.44\u0026plusmn;6.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e43.05\u0026plusmn;6.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=0.001;ii=0.752;iii=0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e42.1[32.4-59.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e41.7[31.3-58.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e41.9[33.4-59]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV47.25[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e3.77\u0026plusmn;3.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.80\u0026plusmn;2.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e3.47\u0026plusmn;2.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=0.002;ii=0.004;iii=0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e3.1[0-10.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.3[0-7.6]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e3[0-9.7]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBowelBag\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eDmean[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e24.42\u0026plusmn;3.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e24.40\u0026plusmn;3.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e24.47\u0026plusmn;3.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=1.000;ii=0.315;iii=0.649\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e24.28[16.22-31.68]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e24.62[16.35-31.96]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e24.53[16.2-31.54]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV40[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e21.28\u0026plusmn;6.84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e21.19\u0026plusmn;6.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e21.27\u0026plusmn;6.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=0.449;ii=0.885;iii=0.429\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e20.2[11.1-33.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e20.5[11.2-33.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e20.4[11.1-33.3]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV45[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e15.28\u0026plusmn;5.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e15.13\u0026plusmn;5.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e15.23\u0026plusmn;5.42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=0.007;ii=0.052;iii=0.031\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e14.8[7.5-25.9]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e14.9[7.6-25.8]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e14.8[7.6-25.8]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV47.25 [%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.96\u0026plusmn;1.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.52\u0026plusmn;1.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.85\u0026plusmn;1.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=0.001;ii=0.041;iii=0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.7[0.4-5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.3[0.3-4.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.4[0.5-4.8]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRectum\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eDmean[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e38.33\u0026plusmn;3.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e38.24\u0026plusmn;3.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e38.46\u0026plusmn;3.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=0.427;ii=0.363;iii=0.017\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.33[30.88-42.53]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.11[31.34-42.32]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e39.29[31.53-42.54]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV40[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e65.96\u0026plusmn;10.01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e66.24\u0026plusmn;10.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e66.29\u0026plusmn;9.87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=0.91;ii=0.865;iii=0.801\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e67.1[49.2-83.2]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e66.9[47.6-82.9]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e67.3[48.1-82.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eV45[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e30.79\u0026plusmn;10.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e28.85\u0026plusmn;11.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e33.05\u0026plusmn;10.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003ei=0.033;ii=0.003;iii=0.005\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e30.5[14.1-53]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e27.5[12.9-54.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e33[16.7-53.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 555px;\"\u003e\n \u003cp\u003eAbbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P\u0026lt;0.05 indicates a statistical significance.SD =Standard deviation.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eWe also analyzed the dosimetry of the overlap regions between the Bladder and PTV (Bladder\u003csub\u003eoverlap\u003c/sub\u003e) and between the Bowelbag and PTV (Bowelbag\u003csub\u003eoverlap\u003c/sub\u003e). In the test case group with SVR \u0026le; 0.40 (Table 6), RMS\u0026apos;s Bladder\u003csub\u003eoverlap\u003c/sub\u003e and Bowelbag\u003csub\u003eoverlap\u003c/sub\u003e showed significantly lower Dmean and V47.25 compared to UMO and RML. Additionally, although not statistically significant, RMS provided lower V45 in both Bladder\u003csub\u003eoverlap\u003c/sub\u003e and Bowelbag\u003csub\u003eoverlap\u003c/sub\u003e compared to UMO and RML. Only in V45 were RML\u0026apos;s Bladder\u003csub\u003eoverlap\u003c/sub\u003e and Bowelbag\u003csub\u003eoverlap\u003c/sub\u003e significantly higher than UMO, with no significant differences in other parameters. In the test case group with SVR \u0026gt; 0.40 (Table 7), except for RML\u0026apos;s significantly lower V47.25 in Bladder\u003csub\u003eoverlap\u003c/sub\u003e compared to UMO, there were no significant differences in dose parameters for both Bladder\u003csub\u003eoverlap\u003c/sub\u003e and Bowelbag\u003csub\u003eoverlap\u003c/sub\u003e. However, in the hotspot region V47.25 (V105%), RML\u0026apos;s Bowelbag\u003csub\u003eoverlap\u003c/sub\u003e and Bladder\u003csub\u003eoverlap\u003c/sub\u003e were respectively 0.53 and 0.65 percentage points lower than UMO. The hotspot region (V105%) of protocol-specified DVH metrics was compared,\u0026nbsp;and Figure 3 showed the changes in Bladder\u003csub\u003eoverlap\u003c/sub\u003e and Bowelbag\u003csub\u003eoverlap\u003c/sub\u003e for each patient in the SVR\u0026nbsp;\u0026le;\u0026nbsp;0.40 and SVR \u0026gt; 0.40 sets, with different verification plans plotted.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 6 Bladder\u003csub\u003eoverlap\u003c/sub\u003e and BowelBag\u003csub\u003eoverlap\u003c/sub\u003e DVH metrics for UMO, RMS and RML with SVR\u0026le;0.40\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"565\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eStructure/Parameter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eUMO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eRMS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eRML\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBladder\u003csub\u003eoverlap\u003c/sub\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eDmean[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.16\u0026plusmn;0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.14\u0026plusmn;0.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.21\u0026plusmn;0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.019;ii=0.05;iii=0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.14[45.67-46.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.14[45.55-46.42]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.22[45.74-46.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eV45[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e93.02\u0026plusmn;2.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e92.81\u0026plusmn;2.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e93.53\u0026plusmn;1.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.813;ii=0.035;iii=0.028\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e93.1[90-97.4]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e93.3[87.5-97]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e93.2[90.8-96.8]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eV47.25 [%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e9.29\u0026plusmn;5.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e8.58\u0026plusmn;4.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e9.98\u0026plusmn;5.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.011;ii=0.055;iii=0.006\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e9.1[0-18.3]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e8.4[0.1-15.2]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e10.2[0.1-21.3]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBowelBag\u003csub\u003eoverlap\u003c/sub\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eDmean[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.18\u0026plusmn;0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.11\u0026plusmn;0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.21\u0026plusmn;0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.001;ii=0.053;iii=0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.19[45.77-46.44]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.1[45.72-46.38]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.23[45.85-46.39]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eV45[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e94.05\u0026plusmn;2.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e93.87\u0026plusmn;2.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e94.47\u0026plusmn;2.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.255;ii=0.048;iii=0.009\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e94.7[87.8-96.8]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e93.9[86.8-97.7]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e94.7[88.9-97]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eV47.25[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e7.11\u0026plusmn;4.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e5.18\u0026plusmn;3.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e7.39\u0026plusmn;3.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.001;ii=0.182;iii=0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e6.9[1-15.1]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e5.4[0.6-11.8]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e6.9[1.5-12]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eAbbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P\u0026lt;0.05 indicates a statistical significance.SD =Standard deviation.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003eTable 7 Bladder\u003csub\u003eoverlap\u003c/sub\u003e and BowelBag\u003csub\u003eoverlap\u003c/sub\u003e DVH metrics for UMO, RMS and RML with SVR\u0026gt;0.40\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"565\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eStructure/Parameter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eUMO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eRMS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003eRML\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBladder\u003csub\u003eoverlap\u003c/sub\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eDmean[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.15\u0026plusmn;0.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.05\u0026plusmn;0.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.15\u0026plusmn;0.28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.001;ii=1;iii=0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.21[45.67-46.52]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.14[45.57-46.42]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.22[45.61-46.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eV45[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e93.81\u0026plusmn;3.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e92.75\u0026plusmn;4.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e94.04\u0026plusmn;3.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.001;ii=0.95;iii=0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e96.2[87.3-98]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.4[84.5-96.9]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.2[86.3-97.8]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eV47.25[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e8.15\u0026plusmn;6.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e6.05\u0026plusmn;4.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e7.5\u0026plusmn;5.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.002;ii=0.004;iii=0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e7.9[0-19]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e5.8[0-13.7]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e7.5[0-17.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBowelBag\u003csub\u003eoverlap\u003c/sub\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eDmean[Gy]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.33\u0026plusmn;0.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.25\u0026plusmn;0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.33\u0026plusmn;0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.002;ii=0.551;iii=0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.26[46.08-46.61]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.22[46.07-46.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e46.31[46.11-46.58]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eV45[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.49\u0026plusmn;1.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.05\u0026plusmn;1.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.37\u0026plusmn;1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.018;ii=0.176;iii=0.016\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e96.2[91.5-97.3]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.4[92.3-96.7]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e95.9[92-97.1]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eV47.25[%]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMean\u0026plusmn;SD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e11.72\u0026plusmn;5.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e9.04\u0026plusmn;4.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e11.19\u0026plusmn;5.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 121px;\"\u003e\n \u003cp\u003ei=0.001;ii=0.102;iii=0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMedian[range]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e11[4.3-21.5]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e8.9[3.1-19.2]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e11.5[4.2-20.8]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003eAbbreviations: UMO= Plan configured with Unclassification Model-O; RMS= plan configured with reclassification Model-S were named RMS; RML= plan configured with reclassification Model-L were named RML. i, UMO vs RMS; ii, UMO vs. RML; iii, RMS vs. RML P\u0026lt;0.05 indicates a statistical significance.SD =Standard deviation.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eAs an illustrative case, Figure 4 depicts the dose distribution of the UMO, RMS, and RML plans with SVR \u0026le; 0.40. Figure 5 shows the DVHs of the example above, primarily including PTV, bladder, and bowelbag. In this case, the V100% for UMO, RMS, and RML was 96.7, 96.5, and 96.5, respectively; there were minimal differences in V47.25 (V105%) for the Bladder and Bowelbag. The values for Bladder\u003csub\u003eoverlap\u003c/sub\u003e and Bowelbag\u003csub\u003eoverlap\u003c/sub\u003e were 1.1 and 6.9 for UMO, and 1.0 and 6.0 for RML, respectively. RMS exhibited lower values in Bladder\u003csub\u003eoverlap\u003c/sub\u003e and Bowelbag\u003csub\u003eoverlap\u003c/sub\u003e, with V47.25 (V105%) being 0.6 and 3.8, respectively.\u003c/p\u003e"},{"header":"4. DISCUSSION","content":"\u003cp\u003eThis study demonstrates an effective approach for constructing overlap volumes-based training libraries for cervical cancer RapidPlan models and validates their predictive performance in DVH. Many studies have shown that establishing a general RapidPlan model for a specific disease to improve OAR dose is feasible. However, some researchers have opted to establish multiple RapidPlan models for certain cases based on specific selection criteria. Nan Li et al. [29] demonstrated that REFINED models, reconstructed for cervical cancer cases with specific dose constraints on pelvic bone marrow (PBM) and bowelbag, could better protect PBM and bowelbag. Gang Yu et al. [19] used a specific selection criterion where the distance from the PTV to the right kidney was less than three centimeters, resulting in a model that improved the PTV\u0026apos;s conformity index (CI) and more effectively protected OARs during IMRT treatment for individual liver cancer patients. O\u0026apos;Toole J et al. [24] found that modeling paired OARs together yielded superior performance compared to separate modeling. The essence of the above reports is to construct diversified RapidPlan models based on specific selection criteria to enhance model performance. Therefore, when there are sufficient training case numbers, constructing specific models based on special criteria is a better way to achieve improved OAR sparing. This study utilized the volumes-based IMRT RapidPlan with SVR as a specific screening criterion to construct two models, Model-S and Model-L, which were then compared with Model-O. The findings suggest that volumes-based IMRT RapidPlan offers improved OAR sparing in scenarios with substantial overlap between PTV and OARs. This study pioneers a novel approach to diversified RapidPlan modeling, demonstrating strong generalization capability in complex clinical scenarios. For instance, in nasopharyngeal carcinoma (NPC) treatment, the anatomical overlap between target volumes and bilateral parotid glands varies significantly across different disease stages. Reducing high-dose exposure to the parotid glands (to preserve salivary function) remains a central challenge in radiotherapy planning, and the anatomy overlap volume-based RapidPlan models offer a promising solution for enhanced parotid sparing.\u003c/p\u003e\n\u003cp\u003eIn our study,\u0026nbsp;we extended the strict classification of training cases based on specific criteria to the grouping of validation cases. While most studies construct a general model to predict all cases of a particular cancer type [30, 31], a minority utilize multiple specialized models to predict randomly selected cases. One study employs both a specialized and a general model to predict patients similar to those in the specialized model training set [33]. We reconstructed two specialized models based on the SVR of overlap regions between OARs and the PTV and applied this classification criterion to validation cases. Implementation of this approach was facilitated by the ease of obtaining overlap regions between OAR and PTV in our Eclipse system, contributing significantly to the diversification of such RapidPlan models. Additionally, we established a many-to-many mapping relationship between model construction for the same type of case and training cases. However, in our research, relying solely on patient anatomy (the ratio of OAR and PTV overlap volume) as a limiting factor raises other concerns. For instance, certain patients may have specific pathological conditions in certain organs at risk, requiring special dose sparing, potentially rendering our model predictions clinically inadequate. Therefore, constructing precise models based on a comprehensive selection criterion encompassing patient anatomy and pathological characteristics may be a future potential research direction.\u003c/p\u003e\n\u003cp\u003eAnother potential drawback lies in the classification criteria for patients. For instance, dividing patients into two categories based on case numbers results in a significant difference in the SVR between the two groups. In our training cases, the interval range for SVR \u0026le; 0.4 is [0.166\u0026ndash;0.385], while for SVR \u0026gt; 0.4, it is [0.450\u0026ndash;0.868]. The group with the larger interval range contains richer anatomical information, but it also implies a decrease in the average quality of the training plans. Currently, there are no reported studies on how to classify overlap regions between OAR and PTV. We have another idea: assuming a considerable number of cases, the SVR of cases should exhibit a normal distribution trend. We can use the 80/20 rule (also known as the Pareto principle) [34, 35] based on SVR values to divide cases into three categories: the middle 80% of cases are training into one model, while the remaining 10% at each end are using to construct two additional models. Models constructed in this manner contain more concentrated anatomical information, thereby enhancing the model\u0026apos;s ability to predict DVHs for OARs. Alternatively, employing statistical methods (e.g., derived from ROC analysis) that balance sensitivity and specificity for clinical outcomes can help establish a more robust threshold. Extending this construction method to incorporate both patient anatomy and pathological characteristics as comprehensive selection criteria for building precise models is also highly appealing.\u003c/p\u003e\n\u003cp\u003eRapidPlan is a knowledge-based, semi-automated planning system, and its predictive performance for DVHs is influenced by various factors. The quality of prior plans and anatomical information from regions of interest significantly impact the model\u0026apos;s predictive capabilities. We classified anatomical structures based on the SVR of bowelbag and bladder overlap with the PTV, while not restricting other OARs, and established two specific RapidPlan models. Our findings may offer guidance for the clinical application of radiotherapy in cervical cancer. Future research will focus on exploring models using broader screening criteria to develop precise models tailored for individual patients, aiming to optimize clinical plans.\u003c/p\u003e"},{"header":"5. CONCLUSION","content":"\u003cp\u003eIn conclusion, we have explored an efficient method to train and refine the RapidPlan model using the SVR of overlapping regions between OARs and the PTV. In the final validation samples, the reclassification models have demonstrated improved sparing of normal tissue doses. In the future, we may construct diversified models based on various anatomical and case characteristics to provide better OARs sparing for individual clinical cases of cervical.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding statement and\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003cstrong\u003e：\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was funded by the National Natural Science Foundation of China (grant number 82260604), the National Natural Science Foundation of Jiangxi Province (grant number 20192 BAB205053), both awarded to Jinghua Zhong, and was supported by an Jiangxi Cancer Hospital scientific research open fund project (KFJJ2023YB21), awarded to Changfei Gong.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contribution：\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMinfeng Huang: Data curation, methodology, project administration, writing \u0026ndash; original draft.\u003c/p\u003e\n\u003cp\u003eChangfei Gong: Data curation, methodology, project administration, writing \u0026ndash; original draft.\u003c/p\u003e\n\u003cp\u003eBiaoshui Liu: analysis, investigation\u003c/p\u003e\n\u003cp\u003eYili Wang: Investigation, project administration\u003c/p\u003e\n\u003cp\u003eJinghua Zhong: Data curation, formal analysis\u003c/p\u003e\n\u003cp\u003eMin Wang:Data curation, and investigation.\u003c/p\u003e\n\u003cp\u003eJun Chi: Formal analysis, investigation\u003c/p\u003e\n\u003cp\u003eXinyuan Li: Formal analysis, investigation\u003c/p\u003e\n\u003cp\u003eRuilian Xie: Conceptualization, formal analysis, writing \u0026ndash; review, and editing.\u003c/p\u003e\n\u003cp\u003eChunbo Tang: Conceptualization, formal analysis, writing \u0026ndash; review, and editing.\u003c/p\u003e\n\u003cp\u003eMinfeng Huang and Changfei Gong have contributed equally to this work\u003c/p\u003e\n\u003cp\u003eRuilian Xie and Chunbo Tang are co-corresponding authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eConflict of Interest:\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudies involving animal subjects\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGenerated Statement: No animal studies are presented in this manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInclusion of identifiable human data\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGenerated Statement: No potentially identifiable human images or data is presented in this study\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe use of patient data was approved by the institutional review board at the First Affiliated Hospital of Gannan Medical University with Ethics review number of LLSC-2024-133.\u003c/p\u003e"},{"header":"REFERENCES","content":"\u003col\u003e\n\u003cli\u003eSung H, Ferlay J, Siegel R L, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries[J]. CA: A Cancer Journal for Clinicians, 2021, 71(3): 209-249.\u003c/li\u003e\n\u003cli\u003eDizon D S, Mackay H J, Thomas G M, et al. State of the science in cervical cancer: where we are today and where we need to go[J]. Cancer, 2014, 120(15): 2282-2288.\u003c/li\u003e\n\u003cli\u003ePowell M E B. Modern radiotherapy and cervical cancer[J]. International Journal of Gynecological Cancer: Official Journal of the International Gynecological Cancer Society, 2010, 20(11 Suppl 2): S49-51.\u003c/li\u003e\n\u003cli\u003eChargari C, Peignaux K, Escande A, et al. Radiotherapy of cervical cancer[J]. Cancer/Radioth\u0026eacute;rapie, 2022, 26(1-2): 298-308.\u003c/li\u003e\n\u003cli\u003eMiguel-Chumacero E, Currie G, Johnston A, et al. Effectiveness of Multi-Criteria Optimization-based Trade-Off exploration in combination with RapidPlan for head \u0026amp; neck radiotherapy planning[J]. Radiation Oncology (London, England), 2018, 13(1): 229.\u003c/li\u003e\n\u003cli\u003eCitrin D E. Recent Developments in Radiotherapy[J]. The New England Journal of Medicine, 2017, 377(11): 1065-1075.\u003c/li\u003e\n\u003cli\u003eHussein M, South C P, Barry M A, et al. Clinical validation and benchmarking of knowledge-based IMRT and VMAT treatment planning in pelvic anatomy[J]. Radiotherapy and Oncology: Journal of the European Society for Therapeutic Radiology and Oncology, 2016, 120(3): 473-479.\u003c/li\u003e\n\u003cli\u003eLoiselle C, Koh W-J. The emerging use of IMRT for treatment of cervical cancer[J]. Journal of the National Comprehensive Cancer Network: JNCCN, 2010, 8(12): 1425-1434.\u003c/li\u003e\n\u003cli\u003eAnonymous. Cervical cancer[J]. The Lancet, 2019, 393(10167): 169-182.\u003c/li\u003e\n\u003cli\u003eYuan L, Ge Y, Lee W R, et al. Quantitative analysis of the factors which affect the interpatient organ-at-risk dose sparing variation in IMRT plans[J]. Medical Physics, 2012, 39(11): 6868-6878.\u003c/li\u003e\n\u003cli\u003eNelms B E, Robinson G, Markham J, et al. Variation in external beam treatment plan quality: An inter-institutional study of planners and planning systems[J]. Practical Radiation Oncology, 2012, 2(4): 296-305.\u003c/li\u003e\n\u003cli\u003eMoore K L, Brame R S, Low D A, et al. Quantitative metrics for assessing plan quality[J]. Seminars in Radiation Oncology, 2012, 22(1): 62-69.\u003c/li\u003e\n\u003cli\u003eTinoco M, Waga E, Tran K, et al. RapidPlan development of VMAT plans for cervical cancer patients in low- and middle-income countries[J]. Medical Dosimetry, 2020, 45(2): 172-178.\u003c/li\u003e\n\u003cli\u003eChua L M H, Pang E P P, Master Z, et al. Dosimetric comparison of RapidPlan and manually optimised volumetric modulated arc therapy plans in prostate cancer[J]. Journal of Radiotherapy in Practice, 2021, 20(3): 257-264.\u003c/li\u003e\n\u003cli\u003eYu S, Xu H, Sinclair A, et al. Dosimetric and planning efficiency comparison for lung SBRT: CyberKnife vs VMAT vs knowledge-based VMAT[J]. Medical Dosimetry, 2020, 45(4): 346-351.\u003c/li\u003e\n\u003cli\u003eD G, J L, Wr L, et al. A knowledge-based approach to improving and homogenizing intensity modulated radiation therapy planning quality among treatment centers: an example application to prostate cancer planning[J/OL]. International journal of radiation oncology, biology, Physics, 2013, 87(1)[2024-03-31].\u003c/li\u003e\n\u003cli\u003eWu B, McNutt T, Zahurak M, et al. Fully automated simultaneous integrated boosted-intensity modulated radiation therapy treatment planning is feasible for head-and-neck cancer: a prospective clinical study[J]. International Journal of Radiation Oncology, Biology, Physics, 2012, 84(5): e647-653.\u003c/li\u003e\n\u003cli\u003eYuan L, Wu Q J, Yin F-F, et al. Incorporating single-side sparing in models for predicting parotid dose sparing in head and neck IMRT[J]. Medical Physics, 2014, 41(2): 021728.\u003c/li\u003e\n\u003cli\u003eYu G, Li Y, Feng Z, et al. Knowledge-based IMRT planning for individual liver cancer patients using a novel specific model[J]. Radiation Oncology (London, England), 2018, 13(1): 52.\u003c/li\u003e\n\u003cli\u003eXu Y, Cyriac J, De Ornelas M, et al. Knowledge-Based Planning for Robustly Optimized Intensity-Modulated Proton Therapy of Head and Neck Cancer Patients[J]. Frontiers in Oncology, 2021, 11: 737901.\u003c/li\u003e\n\u003cli\u003eGay H A, Barthold H J, O\u0026rsquo;Meara E, et al. Pelvic normal tissue contouring guidelines for radiation therapy: a Radiation Therapy Oncology Group consensus panel atlas[J]. International Journal of Radiation Oncology, Biology, Physics, 2012, 83(3): e353-362.\u003c/li\u003e\n\u003cli\u003eWay H. Eclipse Photon and Electron Algorithms Reference[J]. [no date]. \u003c/li\u003e\n\u003cli\u003eMattes M D, Lee J C, Elnaiem S, et al. A predictive model to guide management of the overlap region between target volume and organs at risk in prostate cancer volumetric modulated arc therapy[J]. Radiation Oncology Journal, 2014, 32(1): 23-30.\u003c/li\u003e\n\u003cli\u003eO\u0026rsquo;Toole J, Wu K, Bromley R, et al. Parotid sparing in RapidPlan Oropharynx models: To split or not to split[J]. Journal of Medical Radiation Sciences, 2020, 67(1): 80-86.\u003c/li\u003e\n\u003cli\u003eBratengeier K, Oechsner M, Gainey M, et al. Remarks on reporting and recording consistent with the ICRU reference dose[J]. Radiation Oncology (London, England), 2009, 4: 44.\u003c/li\u003e\n\u003cli\u003eShang H, Pu Y, Wang W, et al. Evaluation of plan quality and robustness of IMPT and helical IMRT for cervical cancer[J]. Radiation Oncology (London, England), 2020, 15(1): 34.\u003c/li\u003e\n\u003cli\u003eWang M, Gu H, Hu J, et al. Evaluation of a highly refined prediction model in knowledge-based volumetric modulated arc therapy planning for cervical cancer[J]. Radiation Oncology (London, England), 2021, 16(1): 58.\u003c/li\u003e\n\u003cli\u003eKaur H, Thakur N, Sharma R, et al. Dosimetric comparison between carotid-sparing IMRT and 3DCRT in early glottic cancer patients treated with definitive radiation therapy[J]. Journal of Cancer Research and Therapeutics, 2024, 20(1): 327-332.\u003c/li\u003e\n\u003cli\u003eLi N, Carmona R, Sirak I, et al. Highly Efficient Training, Refinement, and Validation of a Knowledge-based Planning Quality-Control System for Radiation Therapy Clinical Trials[J]. International Journal of Radiation Oncology, Biology, Physics, 2017, 97(1): 164-172.\u003c/li\u003e\n\u003cli\u003eBi S, Sun X, Sohaimi W F B W, et al. Study on the transferability of the knowledge-based VMAT model to predict IMRT plans in prostate cancer radiotherapy[J]. European Journal of Medical Research, 2023, 28(1): 309.\u003c/li\u003e\n\u003cli\u003eFranceschini D, Cozzi L, Fogliata A, et al. Training and validation of a knowledge-based dose-volume histogram predictive model in the optimisation of intensity-modulated proton and volumetric modulated arc photon plans for pleural mesothelioma patients[J]. Radiation Oncology (London, England), 2022, 17(1): 150.\u003c/li\u003e\n\u003cli\u003eHussein M, South C P, Barry M A, et al. Clinical validation and benchmarking of knowledge-based IMRT and VMAT treatment planning in pelvic anatomy[J]. Radiotherapy and Oncology: Journal of the European Society for Therapeutic Radiology and Oncology, 2016, 120(3): 473-479.\u003c/li\u003e\n\u003cli\u003eFogliata A, Reggiori G, Stravato A, et al. RapidPlan head and neck model: the objectives and possible clinical benefit[J]. Radiation Oncology (London, England), 2017, 12(1): 73.\u003c/li\u003e\n\u003cli\u003eThoma A, Farrokhyar F, McKnight L, Bhandari M. How to optimize patient recruitment. Can J Surg. 2010;53(3):205\u0026ndash;10. \u003c/li\u003e\n\u003cli\u003eFran\u0026ccedil;ois B, Clavel M, Vignon P, Laterre PF. Perspective on optimizing clinical trials in critical care: how to puzzle out recurrent failures. J Intensive Care. 2016;4(1):1\u0026ndash;11.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"physical-and-engineering-sciences-in-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"apes","sideBox":"Learn more about [Physical and Engineering Sciences in Medicine](http://link.springer.com/journal/13246)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/apes/default.aspx","title":"Physical and Engineering Sciences in Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"overlap volumes-based, Rapidplan model diversification, OARs sparing","lastPublishedDoi":"10.21203/rs.3.rs-5971123/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5971123/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003ePurpose: \u003c/strong\u003eDeveloping a novel overlap volumes-basedRapidPlan model for cervical cancer radiotherapy, aiming to spare organs at risk (OARs).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e A retrospective analysis was conducted on 110 cases of cervical cancer patients. We utilized the planning target volume (PTV) and two organs at risk (OARs): bladder and bowelbag, as the primary protocol standards for planning. Original and unclassified dose-volume histogram (DVH) estimation models (Model-O) were trained using 80 initial plans, there are 40 cases in RMS and RML respectively. The classification criteria used the ratio of bladder volume overlapping with PTV to total bladder volume, and the ratio of bowelbag volume overlapping with PTV to total bowelbag volume. Based on the sum of these volume ratios (SVR), with thresholds of ≤0.4 and \u0026gt;0.4, new training sets and DVH estimation models were generated, named Model-S and Model-L respectively. The plans created using Model-O were named UMO, those configured with Model-S were referred to as RMS, and the plans configured with Model-L were called RML. An analysis of dosimetric parameters for the PTV and OARs was conducted for the three automated plans.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e For 15 patients with SVR ≤ 0.40, UMO, RMS, and RML automatically produced clinically acceptable plans. Comparing UMO and RML, RMS reduced the high-dose region V47.25 (V105%) for the PTV and provided greater dose sparing in the Bladder and Bowelbag. In cases where SVR exceeded 0.40, the mean PTV coverage with RMS only 94.9%. Comparing to UMO, RML showed greater dose sparing for the Bladder, Bowelbag, Bladder overlap, and Bowelbag overlap, as well as improved V100% of the PTV.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eThe results indicate that employing overlap volumes-based IMRT Rapidplan enhances sparing of OARs when dealing with significant overlap between PTV and OARs.\u003c/p\u003e","manuscriptTitle":"Exploration of overlap volumes-based IMRT Rapidplan for cervical cancer patients with the aim of OARs sparing","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-14 23:07:59","doi":"10.21203/rs.3.rs-5971123/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorAssigned","content":"","date":"2025-05-16T14:44:30+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"","date":"2025-04-10T18:36:16+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-10T10:10:40+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"Physical and Engineering Sciences in Medicine","date":"2025-04-05T14:03:03+00:00","index":"","fulltext":""},{"type":"submitted","content":"Physical and Engineering Sciences in Medicine","date":"2025-04-05T02:16:30+00:00","index":"","fulltext":""},{"type":"decision","content":"Accept","date":"2025-03-31T04:51:45+00:00","index":"","fulltext":""},{"type":"decision","content":"Minor revisions","date":"2025-03-31T04:51:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"physical-and-engineering-sciences-in-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"apes","sideBox":"Learn more about [Physical and Engineering Sciences in Medicine](http://link.springer.com/journal/13246)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/apes/default.aspx","title":"Physical and Engineering Sciences in Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"36792d47-79f2-47f6-bd97-a8f0da35cfe0","owner":[],"postedDate":"April 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-05-04T16:00:46+00:00","versionOfRecord":{"articleIdentity":"rs-5971123","link":"https://doi.org/10.1007/s13246-026-01732-4","journal":{"identity":"physical-and-engineering-sciences-in-medicine","isVorOnly":false,"title":"Physical and Engineering Sciences in Medicine"},"publishedOn":"2026-04-29 15:57:56","publishedOnDateReadable":"April 29th, 2026"},"versionCreatedAt":"2025-04-14 23:07:59","video":"","vorDoi":"10.1007/s13246-026-01732-4","vorDoiUrl":"https://doi.org/10.1007/s13246-026-01732-4","workflowStages":[]},"version":"v1","identity":"rs-5971123","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5971123","identity":"rs-5971123","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}