Purpose: To develop an automated treatment planning strategy for external beam intensity-modulated radiation therapy (IMRT), including a deep learning-based three-dimensional (3D) dose prediction and a dose distribution-based plan generation algorithm. Methods and Materials: A residual neural network-based deep learning model is trained to predict a dose distribution based on patient-specific geometry and prescription dose. A total of 270 headand-neck cancer cases were enrolled in this study, including 195 cases in the training dataset, 25 cases in the validation dataset, and 50 cases in the testing dataset. All patients were treated with IMRT with a variety of different prescription patterns. The model input consists of CT images and contours delineating the organs at risk (OARs) and planning target volumes (PTVs). The algorithm output is trained to predict the dose distribution on the CT image slices. The obtained prediction model is used to predict dose distributions for new patients. Then, an optimization objective function based on these predicted dose distributions is created for automatic plan generation. Results: Our results demonstrate that the deep learning method can predict clinically acceptable dose distributions. There is no statistically significant difference between prediction and real clinical plan for all clinically relevant dose-volume histogram (DVH) indices, except brainstem, right and left lens. However, the predicted plans were still clinically acceptable. The results of plan generation show no statistically significant differences between the automatic generated plan and the predicted plan except PTV 70.4 , but the difference is only 0.5% which is still clinically acceptable. Conclusion: This study developed a new automated radiotherapy treatment planning system based on 3D dose prediction and 3D dose distribution-based optimization. It is a promising approach for realizing automated treatment planning in the future.

BackgroundKnowledge-based planning (KBP) is a promising technique that can improve plan quality and increase planning efficiency. However, no attempts have been made to extend the domain of KBP for planners with different planning experiences so far. The purpose of this study was to quantify the potential gains for planners with different planning experiences after implementing KBP in intensity modulated radiation therapy (IMRT) plans for left-sided breast cancer patients.MethodsThe model libraries were populated with 80 expert clinical plans from treated patients who previously received left-sided breast-conserving surgery and IMRT with simultaneously integrated boost. The libraries were created on the RapidPlanTM. 6 planners with different planning experiences (2 beginner planners, 2 junior planners and 2 senior planners) generated manual and KBP optimized plans for additional 10 patients, similar to those included in the model libraries. The plan qualities were compared between manual and KBP plans.ResultsAll plans were capable of achieving the prescription requirement. There were almost no statistically significant differences in terms of the planning target volume (PTV) coverage and dose conformality. It was demonstrated that the doses for most of organs-at-risk (OARs) were on average lower or equal in KBP plans compared to manual plans except for the senior planners, where the very small differences were not statistically significant. KBP data showed a systematic trend to have superior dose sparing at most parameters for the heart and ipsilateral lung. The observed decrease in the doses to these OARs could be achieved, particularly for the beginner and junior planners. Many differences were statistically significant.ConclusionsIt is feasible to generate acceptable IMRT plans after implementing KBP for left-sided breast cancer. KBP helps to effectively improve the quality of IMRT plans against the benchmark of manual plans for less experienced planners without any manual intervention. KBP showed promise for homogenizing the plan quality by transferring planning expertise from more experienced to less experienced planners.

BackgroundThe aim of this study was to analyze the influence of volumetric changes of bladder and rectum filling on the 3D dose distribution in prostate cancer radiotherapy.MethodsA total of 314 cone-beam CT (CBCT) image data sets from 19 patients were enrolled in this study. For each CBCT, the bladder and rectum were contoured and volume sizes were normalized to those on their original CT. The daily delivered dose was recalculated on the CBCT images and the doses to bladder and rectum were investigated. Linear regression analysis was performed to identify the mean dose change of the volume change using SPSS 19.ResultsThe data show that the variances of the normalized volume of the bladder and the rectum are 0.13–0.58 and 0.12–0.50 respectively. The variances of V70Gy, V60Gy, V50Gy, V40Gy and V30Gy of bladder are bigger than those of rectum for 17 patients. The linear regression analysis indicates a 10 % increase in bladder volume will cause a 5.6 % (±4.9 %) reduction in mean dose (p <0.05).ConclusionsThe bladder’s volume change is more significant than that of the rectum for the prostate cancer patient. The rectum volume variations are not significant except for air bubbles, which change the shape and the position of the rectum. The bladder volume variations may cause dose changes proportionately. Monitoring the bladder’s volume before fractional treatment delivery will be crucial for accurate dose delivery.

PurposeTo evaluate the reproducibility of radiomics features by repeating computed tomographic (CT) scans in rectal cancer. To choose stable radiomics features for rectal cancer.ResultsVolume normalized features are much more reproducible than unnormalized features. The average value of all slices is the most reproducible feature type in rectal cancer. Different filters have little effect for the reproducibility of radiomics features. For the average type features, 496 out of 775 features showed high reproducibility (ICC ≥ 0.8), 225 out of 775 features showed medium reproducibility (0.8 > ICC ≥ 0.5) and 54 out of 775 features showed low reproducibility (ICC < 0.5).Methods40 rectal cancer patients with stage II were enrolled in this study, each of whom underwent two CT scans within average 8.7 days. 775 radiomics features were defined in this study. For each features, five different values (value from the largest slice, maximum value, minimum value, average value of all slices and value from superposed intermediate matrix) were extracted. Meanwhile a LOG filter with different parameters was applied to these images to find stable filter value. Concordance correlation coefficients (CCC) and inter-class correlation coefficients (ICC) of two CT scans were calculated to assess the reproducibility, based on original features and volume normalized features.ConclusionsFeatures are recommended to be normalized to volume in radiomics analysis. The average type radiomics features are the most stable features in rectal cancer. Further analysis of these features of rectal cancer can be warranted for treatment monitoring and prognosis prediction.