OBJECTIVE: To identify the problems in clinical radiotherapy planning for cervical cancer through quantitative evaluation of the radiotherapy plans to improve the quality of the plans and the radiotherapy process. METHODS: We selected the clinically approved and administered radiotherapy plans for 227 cervical cancer patients undergoing external radiotherapy at Sun Yat-sen University Cancer Center from May, 2019 to January, 2022. These plans were transferred from the treatment planning system to the Plan IQ^TM workstation. The plan quality metrics were determined based on the guidelines of ICRU83 report, the GEC-ESTRO Working Group, and the clinical requirements of our center and were approved by a senior clinician. The problems in the radiotherapy plans were summarized and documented, and those with low scores were re-planned and the differences were analyzed. RESULTS: We identified several problems in the 277 plans by quantitative evaluation. Inappropriate target volume selection (with scores < 60) in terms of GTV, PGTV (CI) and PGTV (V_{66 Gy}) was found in 10.6%, 65.2%, and 1% of the plans, respectively; and the PGTV (CI), GTV, and PCTV (D_98%, HI) had a score of 0 in 0.4%, 10.1%, 0.4%, 0.4% of the plans, respectively. The problems in the organs at risk (OARs) involved mainly the intestines (the rectum, small intestine, and colon), found in 20.7% of the plans, and in occasional cases, the rectum, small intestine, colon, kidney, and the femoral head had a score of 0. Senior planners showed significantly better performance than junior planners in PGTV (V_{60 Gy}, D_98%), PCTV (CI), and CTV (D_98%) (P≤0.046) especially in terms of spinal cord and small intestine protection (P≤0.034). The bowel (the rectum, small intestine and colon) dose was significantly lower in the prone plans than supine plans (P < 0.05), and targets coverage all met clinical requirements. Twenty radiotherapy plans with low scores were selected for re-planning. The re-planned plans had significantly higher GTV (D_min) and PTV (V_{45 Gy}, D_98%) (P < 0.05) with significantly reduced doses of the small intestines (V_{40 Gy} vs V_{30 Gy}), the colon (V_{40 Gy} vs V_{30 Gy}), and the bladder (D_35%) (P < 0.05). CONCLUSION Quantitative evaluation of the radiotherapy plans can not only improve the quality of radiotherapy plan, but also facilitate risk management of the radiotherapy process.
Humans ; Female ; Uterine Cervical Neoplasms/radiotherapy* ; Rectum ; Colon ; Kidney ; Organs at Risk

To study an automatic plan(AP) method for radiotherapy after breast-conserving surgery based on TiGRT system and and compare with manual plan (MP). The dosimetry parameters of 10 patients and the evaluation of scoring table were analyzed, it was found that the targets dose of AP were better than that of MP, but there was no statistical difference except for CI, The V5, V20 and V30 of affected lungs and whole lungs in AP were lower than all that in MP, the Dmean of hearts was slightly higher than that of MP, but the difference was not statistically significant, the MU of AP was increase by 16.1% compared with MP, the score of AP evaluation was increase by 6.1% compared with MP. So the AP could be programmed and automated while ensuring the quality of the plan, and can be used to design the plans for radiotherapy after breast-conserving surgery.
Breast Neoplasms/surgery* ; Female ; Humans ; Mastectomy, Segmental ; Organs at Risk ; Radiotherapy Dosage ; Radiotherapy Planning, Computer-Assisted ; Radiotherapy, Intensity-Modulated

CT image based organ segmentation is essential for radiotherapy treatment planning, and it is laborious and time consuming to outline the endangered organs and target areas before making radiation treatment plans. This study proposes a fully automated segmentation method based on fusion convolutional neural network to improve the efficiency of physicians in outlining the endangered organs and target areas. The CT images of 170 postoperative cervical cancer stage IB and IIA patients were selected for network training and automatic outlining of bladder, rectum, femoral head and CTV, and the neural network was used to localize easily distinguishable vessels around the target area to achieve more accurate outlining of CTV.
Female ; Humans ; Image Processing, Computer-Assisted ; Neural Networks, Computer ; Organs at Risk ; Pelvis ; Tomography, X-Ray Computed ; Uterine Cervical Neoplasms/surgery*

OBJECTIVE: To design a series of geometric indexes, which can improve the correlation between geometric parameters and dosimetric parameters. METHODS: 48 cases of upper abdomen were selected. Manual and automatic segmentation were performed for two organs at risk, which were stomach and duodenum. Three overlapping structures, which were the overlaps with target expanded by 5 mm, 10 mm and 20 mm, were generated for each organ at risk. The geometric parameters of overlapping structures were calculated. The relationship between these geometric parameters and the dosimetric parameters of organs was investigated. RESULTS: When the geometric parameters of overlapping structures related to the target expand 5 mm, 10 mm and 20 mm were larger than 0.4, 0.6 and 0.8 respectively, the maximum dose differences of manual and automatic segmentation were less than 3 Gy. For the case with no overlaps between the organs and the target expansions, the overlap structure corresponding to target expanding 20 mm were recommended for safety considerations. CONCLUSIONS For organs at risk in the upper abdomen, the overlapping geometric parameters were closely related to the maximum dose of organs. Overlapping geometric parameters could predict whether the difference of maximum dose caused by automaticsegmentation was clinically acceptable or not.
Organs at Risk ; Radiometry ; Radiotherapy Dosage ; Radiotherapy Planning, Computer-Assisted ; Radiotherapy, Intensity-Modulated

OBJECTIVES: The automatic delineation of organs at risk (OARs) can help doctors make radiotherapy plans efficiently and accurately, and effectively improve the accuracy of radiotherapy and the therapeutic effect. Therefore, this study aims to propose an automatic delineation method for OARs in cervical cancer scenarios of both after-loading and external irradiation. At the same time, the similarity of OARs structure between different scenes is used to improve the segmentation accuracy of OARs in difficult segmentations. METHODS: Our ensemble model adopted the strategy of ensemble learning. The model obtained from the pre-training based on the after-loading and external irradiation was introduced into the integrated model as a feature extraction module. The data in different scenes were trained alternately, and the personalized features of the OARs within the model and the common features of the OARs between scenes were introduced. Computer tomography (CT) images for 84 cases of after-loading and 46 cases of external irradiation were collected as the train data set. Five-fold cross-validation was adopted to split training sets and test sets. The five-fold average dice similarity coefficient (DSC) served as the figure-of-merit in evaluating the segmentation model. RESULTS: The DSCs of the OARs (the rectum and bladder in the after-loading images and the bladder in the external irradiation images) were higher than 0.7. Compared with using an independent residual U-net (convolutional networks for biomedical image segmentation) model [residual U-net (Res-Unet)] delineate OARs, the proposed model can effectively improve the segmentation performance of difficult OARs (the sigmoid in the after-loading CT images and the rectum in the external irradiation images), and the DSCs were increased by more than 3%. CONCLUSIONS Comparing to the dedicated models, our ensemble model achieves the comparable result in segmentation of OARs for different treatment options in cervical cancer radiotherapy, which may be shorten time for doctors to sketch OARs and improve doctor's work efficiency.
Female ; Humans ; Image Processing, Computer-Assisted/methods* ; Machine Learning ; Organs at Risk/radiation effects* ; Radiotherapy Planning, Computer-Assisted/methods* ; Uterine Cervical Neoplasms/radiotherapy*

OBJECTIVE: To establish an algorithm based on 3D convolution neural network to segment the organs at risk (OARs) in the head and neck on CT images. METHODS: We propose an automatic segmentation algorithm of head and neck OARs based on V-Net. To enhance the feature expression ability of the 3D neural network, we combined the squeeze and exception (SE) module with the residual convolution module in V-Net to increase the weight of the features that has greater contributions to the segmentation task. Using a multi-scale strategy, we completed organ segmentation using two cascade models for location and fine segmentation, and the input image was resampled to different resolutions during preprocessing to allow the two models to focus on the extraction of global location information and local detail features respectively. RESULTS: Our experiments on segmentation of 22 OARs in the head and neck indicated that compared with the existing methods, the proposed method achieved better segmentation accuracy and efficiency, and the average segmentation accuracy was improved by 9%. At the same time, the average test time was reduced from 33.82 s to 2.79 s. CONCLUSIONS The 3D convolution neural network based on multi-scale strategy can effectively and efficiently improve the accuracy of organ segmentation and can be potentially used in clinical setting for segmentation of other organs to improve the efficiency of clinical treatment.
Head ; Humans ; Image Processing, Computer-Assisted ; Neck ; Neural Networks, Computer ; Organs at Risk ; Tomography, X-Ray Computed

When applying deep learning to the automatic segmentation of organs at risk in medical images, we combine two network models of Dense Net and V-Net to develop a Dense V-network for automatic segmentation of three-dimensional computed tomography (CT) images, in order to solve the problems of degradation and gradient disappearance of three-dimensional convolutional neural networks optimization as training samples are insufficient. This algorithm is applied to the delineation of pelvic endangered organs and we take three representative evaluation parameters to quantitatively evaluate the segmentation effect. The clinical result showed that the Dice similarity coefficient values of the bladder, small intestine, rectum, femoral head and spinal cord were all above 0.87 (average was 0.9); Jaccard distance of these were within 2.3 (average was 0.18). Except for the small intestine, the Hausdorff distance of other organs were less than 0.9 cm (average was 0.62 cm). The Dense V-Network has been proven to achieve the accurate segmentation of pelvic endangered organs.
Algorithms ; Humans ; Image Processing, Computer-Assisted ; Imaging, Three-Dimensional ; Neural Networks, Computer ; Organs at Risk ; Pelvis ; Tomography, X-Ray Computed

OBJECTIVE: To investigate the accuracy of automatic segmentation of organs at risk (OARs) in radiotherapy for nasopharyngeal carcinoma (NPC). METHODS: The CT image data of 147 NPC patients with manual segmentation of the OARs were randomized into the training set (115 cases), validation set (12 cases), and the test set (20 cases). An improved network based on three-dimensional (3D) Unet was established (named as AUnet) and its efficiency was improved through end-to-end training. Organ size was introduced as a priori knowledge to improve the performance of the model in convolution kernel size design, which enabled the network to better extract the features of different organs of different sizes. The adaptive histogram equalization algorithm was used to preprocess the input CT images to facilitate contour recognition. The similarity evaluation indexes, including Dice Similarity Coefficient (DSC) and Hausdorff Distance (HD), were calculated to verify the validity of segmentation. RESULTS: DSC and HD of the test dataset were 0.86±0.02 and 4.0±2.0 mm, respectively. No significant difference was found between the results of AUnet and manual segmentation of the OARs ( CONCLUSIONS AUnet, an improved deep learning neural network, is capable of automatic segmentation of the OARs in radiotherapy for NPC based on CT images, and for most organs, the results are comparable to those of manual segmentation.
Databases, Factual ; Humans ; Image Processing, Computer-Assisted ; Nasopharyngeal Carcinoma/radiotherapy* ; Nasopharyngeal Neoplasms/radiotherapy* ; Organs at Risk ; Tomography, X-Ray Computed

PURPOSE: The aim is to study the dependence of deformable based auto-segmentation of head and neck organs-at-risks (OAR) on anatomy matching for a single atlas based system and generate an acceptable set of contours. METHODS: A sample of ten patients in neutral neck position and three atlas sets consisting of ten patients each in different head and neck positions were utilized to generate three scenarios representing poor, average and perfect anatomy matching respectively and auto-segmentation was carried out for each scenario. Brainstem, larynx, mandible, cervical oesophagus, oral cavity, pharyngeal muscles, parotids, spinal cord, and trachea were the structures selected for the study. Automatic and oncologist reference contours were compared using the dice similarity index (DSI), Hausdroff distance and variation in the centre of mass (COM). RESULTS: The mean DSI scores for brainstem was good irrespective of the anatomy matching scenarios. The scores for mandible, oral cavity, larynx, parotids, spinal cord, and trachea were unacceptable with poor matching but improved with enhanced bony matching whereas cervical oesophagus and pharyngeal muscles had less than acceptable scores for even perfect matching scenario. HD value and variation in COM decreased with better matching for all the structures. CONCLUSION: Improved anatomy matching resulted in better segmentation. At least a similar setup can help generate an acceptable set of automatic contours in systems employing single atlas method. Automatic contours from average matching scenario were acceptable for most structures. Importance should be given to head and neck position during atlas generation for a single atlas based system.
Brain Stem ; Head and Neck Neoplasms ; Head ; Humans ; Larynx ; Mandible ; Methods ; Mouth ; Neck ; Organs at Risk ; Pharyngeal Muscles ; Radiotherapy ; Spinal Cord ; Trachea

PURPOSE: The aim of this study was to identify volume changes and dose variations of rectum and bladder during radiation therapy in prostate cancer (PC) patients. MATERIALS AND METHODS: We analyzed 20 patients with PC treated with helical tomotherapy. Daily image guidance was performed. We re-contoured the entire bladder and rectum including its contents as well as the organ walls on megavoltage computed tomography once a week. Dose variations were analyzed by means of Dmedian, Dmean, Dmax, V₁₀ to V₇₅, as well as the organs at risk (OAR) volume. Further, we investigated the correlation between volume changes and changes in Dmean of OAR.
Humans ; Organs at Risk ; Prostate ; Prostatic Neoplasms ; Radiotherapy, Image-Guided ; Radiotherapy, Intensity-Modulated ; Rectum ; Urinary Bladder