ABSTRACT
BACKGROUND: Alopecia causes significant distress for patients and negatively impacts quality of life for low-grade glioma (LGG) patients. We aimed to compare and evaluate variations in dose distribution for scalp-sparing in LGG patients with proton therapy and photon therapy, namely intensity-modulated proton therapy (IMPT), intensity-modulated radiotherapy (IMRT), volumetric modulated arc therapy (VMAT), and helical tomotherapy (HT). METHODS: This retrospective study utilized a dataset comprising imaging data from 22 patients with LGG who underwent postoperative radiotherapy. Treatment plans were generated for each patient with scalp-optimized (SO) approaches and scalp-non-optimized (SNO) approaches using proton techniques and photons techniques; all plans adhered to the same dose constraint of delivering a total radiation dose of 54.04â¯Gy to the target volume. All treatment plans were subsequently analyzed. RESULTS: All the plans generated in this study met the dose constraints for the target volume and OARs. The SO plans resulted in reduced maximum scalp dose (Dmax), mean scalp dose (Dmean), and volume of the scalp receiving 30â¯Gy (V30) and 40â¯Gy (V40) compared with SNO plans in all radiation techniques. Among all radiation techniques, the IMPT plans exhibited superior performance compared to other plans for dose homogeneity as for SO plans. Also, IMPT showed lower values for Dmean and Dmax than all photon radiation techniques. CONCLUSION: Our study provides evidence that the SO approach is a feasible technique for reducing scalp radiation dose. However, it is imperative to conduct prospective trials to assess the benefits associated with this approach.
ABSTRACT
BACKGROUND: Correct delineation of organs at risk (OARs) is an important step for radiotherapy and it is also a time-consuming process that depends on many factors. OBJECTIVE: An automatic quality assurance (QA) method based on deep learning (DL) was proposed to improve efficiency for detecting contouring errors of OARs. MATERIALS AND METHODS: A total of 180 planning CT scan sets at the pelvic site and the corresponding OARs contours from clinics were enrolled in this study. Among them, 140 cases were randomly chosen as the training datasets, 20 cases were used as the validation datasets, and the remaining 20 cases were used as the test datasets. DL-based models were trained through data curation for data cleaning based on the Dice similarity coefficient and the 95th percentile Hausdorff distance between the original contours and the predicted contours. All contouring errors could be classified into two types; minor modification required and major modification required. The pass criteria were established using Bias- Corrected and Accelerated bootstrap on 20 manually reviewed validation datasets. The performance of the QA method was evaluated with the metrics of sensitivity, specificity, the area under the receiving operator characteristic curve (AUC), and detection rate sensitivity on the 20 test datasets. RESULTS: For all OARs, segmentation results after data curation were superior to those without. The sensitivity of the QA method was greater than 0.890 and the specificity was higher than 0.975. The AUCs were 0.948, 0.966, 0.965, and 0.932 for the bladder, right femoral head, left femoral head, and rectum, respectively. Almost all major errors could be detected by the automatic QA method, and the lowest detection rate sensitivity of minor errors was 0.863 for the rectum. CONCLUSIONS: QA of OARs is an important step for the correct implementation of radiotherapy. The DL-based QA method proposed in this study showed a high potential to automatically detect contouring errors with high precision. The method can be integrated into the existing radiotherapy procedures to improve the efficiency of delineating the OARs.
Subject(s)
Deep Learning , Organs at Risk , Humans , Tomography, X-Ray ComputedABSTRACT
BACKGROUND: Manual segment target volumes were time-consuming and inter-observer variability couldn't be avoided. With the development of computer science, auto-segmentation had the potential to solve this problem. OBJECTIVE: To evaluate the accuracy and stability of Atlas-based and deep-learning-based auto-segmentation of the intermediate risk clinical target volume, composed of CTV2 and CTVnd, for nasopharyngeal carcinoma quantitatively. METHODS AND MATERIALS: A cascade-deep-residual neural network was constructed to automatically segment CTV2 and CTVnd by deep learning method. Meanwhile, a commercially available software was used to automatically segment the same regions by Atlas-based method. The datasets included contrast computed tomography scans from 102 patients. For each patient, the two regions were manually delineated by one experienced physician. The similarity between the two auto-segmentation methods was quantitatively evaluated by Dice similarity coefficient, the 95th Hausdorff distance, volume overlap error and relative volume difference, respectively. Statistical analyses were performed using the ranked Wilcoxon test. RESULTS: The average Dice similarity coefficient (±standard deviation) given by the deep-learning- based and Atlas-based auto-segmentation were 0.84 (±0.03) and 0.74 (±0.04) for CTV2, 0.79 (±0.02) and 0.68 (±0.03) for CTVnd, respectively. For the 95th Hausdorff distance, the corresponding values were 6.30±3.55 mm and 9.34±3.39 mm for CTV2, 7.09±2.27 mm and 14.33±3.98 mm for CTVnd. Besides, volume overlap error and relative volume difference could also predict the same situations. Statistical analyses showed significant difference between the two auto-segmentation methods (p<0.01). CONCLUSION: Compared with the Atlas-based segmentation approach, the deep-learning-based segmentation method performed better both in accuracy and stability for meaningful anatomical areas other than organs at risk.
Subject(s)
Deep Learning , Nasopharyngeal Neoplasms , Humans , Nasopharyngeal Carcinoma/diagnostic imaging , Nasopharyngeal Neoplasms/diagnostic imaging , Organs at Risk , Radiotherapy Planning, Computer-Assisted/methodsABSTRACT
OBJECTIVE: To explore the feasibility of using the bidirectional local distance based medical similarity index (MSI) to evaluate automatic segmentation on medical images. METHODS: Taking the intermediate risk clinical target volume for nasopharyngeal carcinoma manually segmented by an experience radiation oncologist as region of interest, using Atlas-based and deep-learning-based methods to obtain automatic segmentation respectively, and calculated multiple MSI and Dice similarity coefficient (DSC) between manual segmentation and automatic segmentation. Then the difference between MSI and DSC was comparatively analyzed. RESULTS: DSC values for Atlas-based and deep-learning-based automatic segmentation were 0.73 and 0.84 respectively. MSI values for them varied between 0.29~0.78 and 0.44~0.91 under different inside-outside-level. CONCLUSIONS: It is feasible to use MSI to evaluate the results of automatic segmentation. By setting the penalty coefficient, it can reflect phenomena such as under-delineation and over-delineation, and improve the sensitivity of medical image contour similarity evaluation.
Subject(s)
Radiotherapy Planning, Computer-Assisted , Feasibility StudiesABSTRACT
BACKGROUND: Surface-guided radiation therapy can be used to continuously monitor a patient's surface motions during radiotherapy by a non-irradiating, noninvasive optical surface imaging technique. In this study, machine learning methods were applied to predict external respiratory motion signals and predict internal liver motion in this therapeutic context. METHODS: Seven groups of interrelated external/internal respiratory liver motion samples lasting from 5 to 6 min collected simultaneously were used as a dataset, Dv. Long short-term memory (LSTM) and support vector regression (SVR) networks were then used to establish external respiratory signal prediction models (LSTMpred/SVRpred) and external/internal respiratory motion correlation models (LSTMcorr/SVRcorr). These external prediction and external/internal correlation models were then combined into an integrated model. Finally, the LSTMcorr model was used to perform five groups of model updating experiments to confirm the necessity of continuously updating the external/internal correlation model. The root-mean-square error (RMSE), mean absolute error (MAE), and maximum absolute error (MAX_AE) were used to evaluate the performance of each model. RESULTS: The models established using the LSTM neural network performed better than those established using the SVR network in the tasks of predicting external respiratory signals for latency-compensation (RMSE < 0.5 mm at a latency of 450 ms) and predicting internal liver motion using external signals (RMSE < 0.6 mm). The prediction errors of the integrated model (RMSE ≤ 1.0 mm) were slightly higher than those of the external prediction and external/internal correlation models. The RMSE/MAE of the fifth model update was approximately ten times smaller than that of the first model update. CONCLUSIONS: The LSTM networks outperform SVR networks at predicting external respiratory signals and internal liver motion because of LSTM's strong ability to deal with time-dependencies. The LSTM-based integrated model performs well at predicting liver motion from external respiratory signals with system latencies of up to 450 ms. It is necessary to update the external/internal correlation model continuously.
Subject(s)
Liver Neoplasms/radiotherapy , Neural Networks, Computer , Radiotherapy, Image-Guided/methods , Algorithms , Humans , Liver , Motion , RespirationABSTRACT
CDATA[Purpose: The aim of this study is to evaluate the accuracy and dosimetric effects for auto- segmentation of the CTV for GO in CT images based on FCN. METHODS: An FCN-8s network architecture for auto-segmentation was built based on Caffe. CT images of 121 patients with GO who have received radiotherapy at the West China Hospital of Sichuan University were randomly selected for training and testing. Two methods were used to segment the CTV of GO: treating the two-part CTV as a whole anatomical region or considering the two parts of CTV as two independent regions. Dice Similarity Coefficient (DSC) and Hausdorff Distance (HD) were used as evaluation criteria. The auto-segmented contours were imported into the original treatment plan to analyse the dosimetric characteristics. RESULTS: The similarity comparison between manual contours and auto-segmental contours showed an average DSC value of up to 0.83. The max HD values for segmenting two parts of CTV separately was a little bit smaller than treating CTV with one label (8.23±2.80 vs. 9.03±2.78). The dosimetric comparison between manual contours and auto-segmental contours showed there was a significant difference (p<0.05) with the lack of dose for auto-segmental CTV. CONCLUSION: Based on deep learning architecture, the automatic segmentation model for small target areas can carry out auto contouring tasks well. Treating separate parts of one target as different anatomic regions can help to improve the auto-contouring quality. The dosimetric evaluation can provide us with different perspectives for further exploration of automatic sketching tools.
Subject(s)
Radiometry , Radiotherapy Planning, Computer-Assisted , China , Humans , Tomography, X-Ray ComputedABSTRACT
Compared with the previous automatic segmentation neural network for the target area which considered the target area as an independent area, a stacked neural network which uses the position and shape information of the organs around the target area to regulate the shape and position of the target area through the superposition of multiple networks and fusion of spatial position information to improve the segmentation accuracy on medical images was proposed in this paper. Taking the Graves' ophthalmopathy disease as an example, the left and right radiotherapy target areas were segmented by the stacked neural network based on the fully convolutional neural network. The volume Dice similarity coefficient (DSC) and bidirectional Hausdorff distance (HD) were calculated based on the target area manually drawn by the doctor. Compared with the full convolutional neural network, the stacked neural network segmentation results can increase the volume DSC on the left and right sides by 1.7% and 3.4% respectively, while the two-way HD on the left and right sides decrease by 0.6. The results show that the stacked neural network improves the degree of coincidence between the automatic segmentation result and the doctor's delineation of the target area, while reducing the segmentation error of small areas. The stacked neural network can effectively improve the accuracy of the automatic delineation of the radiotherapy target area of Graves' ophthalmopathy.
Subject(s)
Neural Networks, Computer , Algorithms , Image Processing, Computer-Assisted , Tomography, X-Ray ComputedABSTRACT
OBJECTIVE: To locate CT images by using the deep learning model based on convolutional neural network. METHODS: The AlexNet network was used as a deep learning model, which was preset by the transfer learning approach. Training samples were divided into 4 categories according to the vertebral body parts and labeled, and the data augmentation was used to improve the classification accuracy. RESULTS: The accuracy of image classification after augmentation increased from 94.95% to 97.72%, and the testing time increased from 2.05 s to 3.03 s. CONCLUSIONS: It is feasible to use the convolutional neural network to locate CT images. The data augmentation approach can increase the classification accuracy but also increase the training and testing time.
Subject(s)
Neural Networks, Computer , Tomography, X-Ray Computed , Feasibility StudiesABSTRACT
A fully biobased and supertough thermoplastic vulcanizate (TPV) consisting of polylactide (PLA) and a biobased vulcanized unsaturated aliphatic polyester elastomer (UPE) was fabricated via peroxide-induced dynamic vulcanization. Interfacial compatibilization between PLA and UPE took place during dynamic vulcanization, which was confirmed by gel measurement and NMR analysis. After vulcanization, the TPV exhibited a quasi cocontinuous morphology with vulcanized UPE compactly dispersed in PLA matrix, which was different from the pristine PLA/UPE blend, exhibiting typically phase-separated morphology with unvulcanized UPE droplets discretely dispersed in matrix. The TPV showed significantly improved tensile and impact toughness with values up to about 99.3 MJ/m(3) and 586.6 J/m, respectively, compared to those of 3.2 MJ/m(3) and 16.8 J/m for neat PLA, respectively. The toughening mechanisms under tensile and impact tests were investigated and deduced as massive shear yielding of the PLA matrix triggered by internal cavitation of VUPE. The fully biobased supertough PLA vulcanizate could serve as a promising alternative to traditional commodity plastics.
Subject(s)
Biocompatible Materials/chemistry , Biodegradable Plastics/chemistry , Peroxides/chemistry , Polyesters/chemistry , Elastomers/chemistryABSTRACT
Novel urethane ionic groups were incorporated into biodegradable poly(ethylene succinate) (PES) by chain extension reaction of PES diol (HO-PES-OH) and diethanolamine hydrochloride (DEAH) using hexamethylene diisocyanate (HDI) as a chain extender. The synthesized polymer was a novel segmented poly(ester urethane) ionomer (PESI) in which the soft segments were formed by reaction of HO-PES-OH with HDI and the hard segments that contained ionic groups were derived from reaction of DEAH with HDI. The crystallization rate of PESI was dramatically accelerated when 3 mol % urethane ionic groups were incorporated. However, the crystallization mechanism did not change. The significant acceleration in crystallization rate was attributed to the improved nucleation efficiency by incorporation of the urethane ionic group, because PESI showed significantly enhanced nucleation density but slightly slowed spherurlitic growth rate in comparison with PES which was synthesized by chain extension reaction of HO-PES-OH with HDI. The increased nucleation efficiency was ascribed to the aggregation of hard segments of PESI induced by the ionic interactions.
ABSTRACT
Chitin is the second most abundant semicrystalline polysaccharide. Like cellulose, the amorphous domains of chitin can also be removed under certain conditions such as acidolysis to give rise to crystallites in nanoscale, which are the so-called chitin nanocrystals or chitin whiskers (CHWs). CHW together with other organic nanoparticles such as cellulose whisker (CW) and starch nanocrystal show many advantages over traditional inorganic nanoparticles such as easy availability, nontoxicity, biodegradability, low density, and easy modification. They have been widely used as substitutes for inorganic nanoparticles in reinforcing polymer nanocomposites. The research and development of CHW related areas are much slower than those of CW. However, CHWs are still of strategic importance in the resource scarcity periods because of their abundant availability and special properties. During the past decade, increasing studies have been done on preparation of CHWs and their application in reinforcing polymer nanocomposites. Some other applications such as being used as feedstock to prepare chitosan nanoscaffolds have also been investigated. This Article is to review the recent development on CHW related studies.
Subject(s)
Chitin/chemistry , Nanocomposites/chemistry , Nanoparticles/chemistryABSTRACT
AIM: To study the therapeutic value of combination of cryosurgery and (125)iodine seed implantation for locally advanced pancreatic cancer. METHODS: Forty-nine patients with locally advanced pancreatic cancer (males 36, females 13), with a median age of 59 years, were enrolled in the study. Twelve patients had liver metastases. In all cases the tumors were considered unresectable after a comprehensive evaluation. Patients were treated with cryosurgery, which was performed intraoperatively or percutaneously under guidance of ultrasound and/or computed tomography (CT), and (125)iodine seed implantation, which was performed during cryosurgery or post-cryosurgery under guidance of ultrasound and/or CT. A few patients received regional celiac artery chemotherapy. RESULTS: Thirteen patients received intraoperative cryosurgery and 36 received percutaneous cryosurgery. Some patients underwent repeat cryosurgery. (125)Iodine seed implantation was performed during freezing procedure in 35 patients and 3-9 d after cryosurgery in 14 cases. Twenty patients, 10 of whom had hepatic metastases received regional chemotherapy. At 3 mo after therapy, CT was repeated to estimate tumor response to therapy. Most patients showed varying degrees of tumor necrosis. Complete response (CR) of tumor was seen in 20.4% patients, partial response (PR), in 38.8%, stable disease (SD), in 30.6%, and progressive disease (PD), in 10.2%. Adverse effects associated with cryosurgery included upper abdomen pain and increased serum amylase. Acute pancreatitis was seen in 6 patients one of whom developed severe pancreatitis. All adverse effects were controlled by medical management with no poor outcome. There was no therapy-related mortality. During a median follow-up of 18 mo (range of 5-40), the median survival was 16.2 mo, with 26 patients (53.1%) surviving for 12 mo or more. Overall, the 6-, 12-, 24- and 36-mo survival rates were 94.9%, 63.1%, 22.8% and 9.5%, respectively. Eight patients had survival of 24 mo or more. The patient with the longest survival (40 mo) is still living without evidence of tumor recurrence. CONCLUSION: Cryosurgery, which is far less invasive than conventional pancreatic resection, and is associated with a low rate of adverse effects, should be the treatment of choice for patients with locally advanced pancreatic cancer. (125)Iodine seed implantation can destroy the residual surviving cancer cells after cryosurgery. Hence, a combination of both modalities has a complementary effect.