Search

Weak correlation between gamma passing rates and dose differences in target volumes and organs at risk (OARs) has been reported in several studies. Evaluation on the differences between planned dose-volume histogram (DVH) and reconstructed DVH from measurement was adopted and incorporated into patient-specific quality assurance (PSQA). However, it is difficult to develop a methodology allowing the evaluation of errors on DVHs accurately and quickly.To develop a DVH-based pretreatment PSQA for volumetric modulated arc therapy (VMAT) with combined deep learning (DL) and machine learning models to overcome the limitation of conventional gamma index (GI) and improve the efficiency of DVH-based PSQA.A DL model with a three-dimensional squeeze-and-excitation residual blocks incorporated into a modified U-net was developed to predict the measured PSQA DVHs of 208 head-and-neck (HN) cancer patients underwent VMAT between 2018 and 2021 from two hospitals, in which 162 cases was randomly selected for training, 18 for validation, and 28 for testing. After evaluating the differences between treatment planning system (TPS) and PSQA DVHs predicted by DL model with multiple metrics, a pass or fail (PoF) classification model was developed using XGBoost algorithm. Evaluation of domain experts on dose errors between TPS and reconstructed PSQA DVHs was taken as ground truth for PoF classification model training.The prediction model was able to achieve a good agreement between predicted, measured, and TPS doses. Quantitative evaluation demonstrated no significant difference between predicted PSQA dose and measured dose for target and OARs, except for DThe successful prediction of PSQA doses and classification of PoF for HN VMAT PSQA indicating that this DVH-based PSQA method is promising to overcome the limitations of GI and to improve the efficiency and accuracy of VMAT delivery.

Objectives. To differentiate the primary site of brain metastases (BMs) is of high clinical value for the successful management of patients with BM. The purpose of this study is to investigate a combined radiomics model with computer tomography (CT) and magnetic resonance imaging (MRI) images in differentiating BMs originated from lung and breast cancer. Methods. Pretreatment cerebral contrast enhanced CT and T1-weighted MRI images of 78 patients with 179 BMs from primary lung and breast cancer were retrospectively analyzed. Radiomic features were extracted from contoured BM lesions and selected using the Mann–Whitney U test and the least absolute shrinkage and selection operator (LASSO) logistic regression. Binary logistic regression (BLR) and support vector machine (SVM) models were built and evaluated based on selected radiomic features from CT alone, MRI alone, and combined images to differentiate BMs originated from lung and breast cancer. Results. A total of 10 and 6 optimal radiomic features were screened out of 1288 CT and 1197 MRI features, respectively. The mean area under the curves (AUCs) of the BLR and SVM models using fivefolds cross-validation were 0.703 vs. 0.751, 0.718 vs. 0.754, and 0.781 vs. 0.803 in the training dataset and 0.708 vs. 0.763, 0.715 vs. 0.717, and 0.771 vs. 0.805 in the testing dataset for models with CT alone, MRI alone, and combined CT and MRI radiomic features, respectively. Conclusions. Radiomics model based on combined CT and MRI features is feasible and accurate in the differentiation of the primary site of BMs from lung and breast cancer.

Strategy for the evaluation of the early diagnostic accuracy of AC- and SCC-specific tumor-derived exosomal miRNAs in a symptomatic set

The raw miRNA-seq data

Preoperative prediction of cervical lymph node metastasis in papillary thyroid carcinoma provided a basis for tumor staging and treatment decision. This study aimed to investigate the utility of machine learning and develop different models to preoperative predict cervical lymph node metastasis based on ultrasonic radiomic features and clinical characteristics in papillary thyroid carcinoma nodules. Data from 400 papillary thyroid carcinoma nodules were included and divided into training and validation group. With help of machine learning, clinical characteristics and ultrasonic radiomic features were extracted and selected using randomforest and least absolute shrinkage and selection operator regression before classified by 5 classifiers. Finally, 10 models were built and their area under the receiver operating characteristic curve, accuracy, sensitivity, specificity, positive predictive value and negative predictive value were measured. Among the 10 models, RF-RF model revealed the highest area under curve (0.812) and accuracy (0.7542) in validation group. The top 10 variables of it included age, 7 textural features, 1 shape feature and 1 first-order feature, in which 8 were high-dimensional features. Thus, RF-RF model showed the best predictive performance for cervical lymph node metastasis. And the importance features selected by it highlighted the unique role of higher-dimensional statistical methods for radiomics analysis.

An accurate and reliable target volume delineation is critical for the safe and successful radiotherapy. The purpose of this study is to develop new 2D and 3D automatic segmentation models based on RefineNet for clinical target volume (CTV) and organs at risk (OARs) for postoperative cervical cancer based on computed tomography (CT) images.A 2D RefineNet and 3D RefineNetPlus3D were adapted and built to automatically segment CTVs and OARs on a total of 44 222 CT slices of 313 patients with stage I-III cervical cancer. Fully convolutional networks (FCNs), U-Net, context encoder network (CE-Net), UNet3D, and ResUNet3D were also trained and tested with randomly divided training and validation sets, respectively. The performances of these automatic segmentation models were evaluated by Dice similarity coefficient (DSC), Jaccard similarity coefficient, and average symmetric surface distance when comparing them with manual segmentations with the test data.The DSC for RefineNet, FCN, U-Net, CE-Net, UNet3D, ResUNet3D, and RefineNet3D were 0.82, 0.80, 0.82, 0.81, 0.80, 0.81, and 0.82 with a mean contouring time of 3.2, 3.4, 8.2, 3.9, 9.8, 11.4, and 6.4 s, respectively. The generated RefineNetPlus3D demonstrated a good performance in the automatic segmentation of bladder, small intestine, rectum, right and left femoral heads with a DSC of 0.97, 0.95, 091, 0.98, and 0.98, respectively, with a mean computation time of 6.6 s.The newly adapted RefineNet and developed RefineNetPlus3D were promising automatic segmentation models with accurate and clinically acceptable CTV and OARs for cervical cancer patients in postoperative radiotherapy.

Multi‐isocenter volumetric modulated arc therapy (VMAT) is recommended for craniospinal irradiation (CSI) to smooth the dose distribution in the junction regions relying solely on inverse optimization. However, few studies have measured the dosimetric impact of setup errors on this multi‐isocenter VMAT in the junction areas. The purpose of this study is to evaluate the impact of positional errors during VMAT CSI with two‐dimension (2D) and three‐dimension (3D) dosimetric measurements. A total of 20 patients treated by three‐isocenter VMAT CSI were retrospectively reviewed and analyzed. A 3D diode array ArcCHECK and radiochromic film EBT3 were applied to measure the percentage gamma passing rates (%GPs) and dose distributions in the junction areas between the cranial/upper‐spinal and the upper/lower‐spinal fields with intentionally introduced setup errors of ± 1 mm, ±2 mm, ±3 mm, ±5 mm, and ± 8 mm, respectively. The length and volume of planning target volume (PTV) for these CSI patients ranged from 50.14 to 80.8 cm, and 1572.3 to 2114.5 cm3, respectively. The %GPs for ±3 mm, ±5 mm, and ±8 mm positional errors were around 95%, 90%, and 85%, respectively, in the junction areas. The dosimetric verification results with EBT3 films indicated that cold and hot areas were observed with the increase of introduced setup errors. In conclusion, the dosimetric verification with intentionally introduced setup errors demonstrated that positional errors within 3 mm have a little impact for VMAT CSI, although setup errors should be minimized. Relying on the inverse optimization of VMAT to smooth the dose distribution in the junction areas is feasible for CSI.

Objective To research and assess automatic volumetric modulated arc therapy (VMAT) planning methods for patients with intraocular cancer. Methods The mdaccAutoPlan system was added to the Pinnacle3 treatment planning system as a plug‐in. Automatic VMAT plans were generated for 10 patients diagnosed with intraocular cancer, and evaluated according to standard dose‐volume histogram parameters. Results The planning target volume of enrolled patients ranges from 14.24 to 50.69 cm3. Both planning methods lead to acceptable planning target volume target coverage with a V95 of 97.9 ± 1.4% and 96.4 ± 1.5% for manual and automatic plans (P = 0.03), respectively. Automatic planning lowered the dose delivered to the ipsilateral lens and optical nerves, but increased the dose to the brainstem compared with manual planning. Automatic planning significantly prolonged VMAT planning time (3.25 ± 0.53 h vs. 1.02 ± 0.69 h, P

It is of high clinical importance to identify the primary lesion and its pathological types for patients with brain metastases (BM). The purpose of this study is to investigate the feasibility and accuracy of differentiating the primary adenocarcinoma (AD) and squamous cell carcinoma (SCC) of non-small-cell lung cancer (NSCLC) for patients with BM based on radiomics from brain contrast-enhanced computer tomography (CECT) images. A total of 144 BM patients (94 male, 50 female) were enrolled in this study with 102 with primary lung AD and 42 with SCC, respectively. Radiomics features from manually contoured tumors were extracted using python. Mann–Whitney U test and the least absolute shrinkage and selection operator (LASSO) logistic regression were applied to select relative radiomics features. Binary logistic regression and support vector machines (SVM) were applied to build models with radiomics features alone and with radiomics features plus age and sex. Fourteen features were selected from a total of 105 radiomics features for the final model building. The area under the curves (AUCs) and accuracy of SVM and binary logistic regression models were 0.765 vs. 0.769, 0.795 vs.0.828, and 0.716 vs. 0.726, 0.768 vs. 0.758, respectively, for models with radiomics features alone and models with radiomics features plus sex and age. Brain CECT radiomics are promising in differentiating primary AD and SCC to achieve optimal therapeutic management in patients with BM from NSCLC. • It is of high clinical importance to identify the primary lesion and its pathological types for patients with brain metastases (BM) to define the prognosis and treatment. • Few studies had investigated the feasibility and accuracy of differentiating the pathological subtypes of primary non-small-cell lung cancer between adenocarcinoma (AD) and squamous cell carcinoma (SCC) for patients with BM based on radiomics from brain contrast-enhanced CT (CECT) images, although CECT images are often the initial imaging modality to screen for metastases and are recommended on equal footing with MRI for the detection of cerebral metastases. • Brain CECT radiomics are promising in differentiating primary AD and SCC to achieve optimal therapeutic management in patients with BM from NSCLC with a highest area under the curve (AUC) of 0.828 and an accuracy of 0.758, respectively.

To investigate the feasibility and accuracy of radiomics models based on contrast-enhanced CT (CECT) in the prediction of perineural invasion (PNI), so as to stratify high-risk recurrence and improve the management of patients with gastric cancer (GC) preoperatively.Total of 154 GC patients underwent D2 lymph node dissection with pathologically confirmed GC and preoperative CECT from an open-label, investigator-sponsored trial (NCT01711242) were enrolled. Radiomics features were extracted from contoured images and selected using Mann-Whitney U test and the least absolute shrinkage and selection operator (LASSO) after inter-class correlation coefficient (ICC) analysis. Models based on radiomics features (R), clinical factors (C) and combined parameters (R + C) were built and evaluated using Support Vector Machine (SVM) and logistic regression to predict the PNI for patients with GC preoperatively.Total of 11 radiomics features were selected for final analysis, along with two clinical factors. The area under curve (AUC) of models based on R, C, and R + C with logistic regression and SVM were 0.77 vs. 0.83, 0.71 vs.0.70, 0.86 vs. 0.90, and 0.73 vs.0.80, 0.62 vs. 0.64, 0.77 vs. 0.82 in the training and testing cohorts, respectively. SVM(R + C) achieved a best AUC of 0.82(0.69-0.94) in the test cohorts with a sensitivity, specificity and accuracy of 0.63, 0.91, and 0.77, respectively.The performance of these models indicates that radiomics features alone or combined with clinical factors provide a feasible way to classify patients preoperatively and improve the management of patients with GC.

Globally, colorectal cancer (CRC) is one of the leading causes of cancer-related deaths. Oxaliplatin based treatments are frequently used as chemotherapeutic methods for CRC, however, associated side effects and drug resistance often limit their clinical application. Dihydroartemisinin (DHA) induces apoptosis in various cancer cells by increasing reactive oxygen species (ROS) production. However, the direct target of DHA and underlying molecular mechanisms in oxaliplatin-mediated anti-tumor activities against CRC are unclear.We used 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT), flow cytometry, and colony formation assays to investigate cell phenotype alterations and ROS generation. We also used quantitative Real-Time PCR (qRT-PCR) and western blotting to measure relative gene and protein expression. Finally, an in vivo mouse xenograft model was used to assess the anti-tumor activity of oxaliplatin and DHA alone, and combinations.DHA synergistically enhanced the anti-tumor activity of oxaliplatin in colon cancer cells by regulating ROS-mediated ER stress, signal transducer and activator of transcription 3 (STAT3), C-Jun-amino-terminal kinase (JNK), and p38 signaling pathways. Mechanistically, DHA increased ROS levels by inhibiting peroxiredoxin 2 (PRDX2) expression, and PRDX2 knockdown sensitized DHA-mediated cell growth inhibition and ROS production in CRC cells. A mouse xenograft model showed strong anti-tumor effects from combination treatments when compared with single agents.We demonstrated an improved therapeutic strategy for CRC patients by combining DHA and oxaliplatin treatments.

PURPOSE Thecurrent practice in intensity-modulated radiation therapy (IMRT) planning almost always includes different dose calculation strategies for plan optimization and final dose verification. The accurate Monte Carlo (MC) dose algorithm is considered to be time-consuming for the optimization. Thus a fast, simplified dose algorithm is used in the optimization. The significant differences between the optimized dose and the delivered dose lead to tediously planning loops and potentially suboptimal solutions. This work aims to develop an IMRT optimization algorithm to minimize the dose discrepancy so that the delivered dose can be optimized in a holistic, end-to-end manner. METHODS The proposed algorithm, namely NeuralDAO, integrates a neural dose network into the column generation (CG) direct aperture optimization (DAO) formulation for step-and-shoot IMRT planning. The neural dose network is designed and trained to produce doses of MC-level accuracy within few milliseconds. Its differentiability is fully exploited to compute gradients for identifying potential aperture shapes. A prototype of NeuralDAO was developed in PyTorch and available to the public. Five lung patient cases have been studied. Dosimetric accuracy was compared with the MC dose. Plan quality and time were compared with a state-of-the-art (SoA) dose-correct algorithm. Statistical analysis was performed by Wilcoxon signed-rank test. RESULTS The average gamma passing rate at 2 mm/2% is 99.7% between the optimized and delivered doses. The convergence process produced by NeuralDAO is virtually identical to that produced by an MC-based DAO. The average dose calculation time is 12.1 ms for an aperture on GPU. One session of optimization took 10-36 min. Compared with the SoA, better conformity index and homogeneity index were observed for the target. The esophagus was significantly spared. Significant reductions were observed for the replanning number and the planning time. CONCLUSIONS A new DAO algorithm based on the neural dose network has been developed. The results suggest that this algorithm minimizes the discrepancy between the optimized and delivered doses, which offers a promising approach to reduce the time and effort required in IMRT planning. This work demonstrates the possibility of applying the neural network in IMRT optimization. It is of great potential to extend this algorithm to other treatment modalities.

While convolutional neural networks (CNNs) have become the first choice for the medical image segmentation, they still lack the critical ingredient of incorporating priors, such as smoothness and boundary shapes. The authors tackle the limitation by developing a novel prior that is boundary-aware in two ways: promoting smoothness without blurring object boundaries and punishing prediction errors according to boundary shapes. They bring the boundary-aware property into effect by weighting the prediction gradients and errors with the distance map. Their prior differs from previous approaches that either over-smooth boundaries or tend to produce rough boundaries. They evaluate their prior alongside the cross-entropy (CE) on a cardiac MRI dataset. Compared to CE alone, their prior improves the Dice score by 1.5% and Hausdorff distance by 53%. It also yielded a faster and more stable learning process.

Purpose As affordable equipment, electronic portal imaging devices (EPIDs) are wildly used in radiation therapy departments to verify patients' positions for accurate radiotherapy. However, these devices tend to produce visually ambiguous and low-contrast planar digital radiographs under megavoltage x ray (MV-DRs), which poses a tremendous challenge for clinicians to perform multimodal registration between the MV-DRs and the kilovoltage digital reconstructed radiographs (KV-DRRs) developed from the planning computed tomography. Furthermore, the existent of strong appearance variations also makes accurate registration beyond the reach of current automatic algorithms. Methods We propose a novel modality conversion approach to this task that first synthesizes KV images from MV-DRs, and then registers the synthesized and real KV-DRRs. We focus on the synthesis technique and develop a conditional generative adversarial network with information bottleneck extension (IB-cGAN) that takes MV-DRs and nonaligned KV-DRRs as inputs and outputs synthesized KV images. IB-cGAN is designed to address two main challenges in deep-learning-based synthesis: (a) training with a roughly aligned dataset suffering from noisy correspondences; (b) making synthesized images have real clinical meanings that faithfully reflects MV-DRs rather than nonaligned KV-DRRs. Accordingly, IB-cGAN employs (a) an adversarial loss to provide training supervision at semantic level rather than the imprecise pixel level; (b) an IB to constrain the information from the nonaligned KV-DRRs. Results We collected 2698 patient scans to train the model and 208 scans to test its performance. The qualitative results demonstrate realistic KV images can be synthesized allowing clinicians to perform the visual registration. The quantitative results show it significantly outperforms current nonmodality conversion methods by 22.37% (P = 0.0401) in terms of registration accuracy. Conclusions The modality conversion approach facilitates the downstream MV-KV registration for both clinicians and off-the-shelf registration algorithms. With this approach, it is possible to benefit the developing countries where inexpensive EPIDs are widely used for the image-guided radiation therapy.

Background Controversial conclusions had been reported in studies trying to confirm the impact of heart dose on overall survival (OS) reported in RTOG 0167 for non-small cell lung cancer (NSCLC) patients who underwent radiotherapy (RT). The purpose of this study is to investigate the association of lung and heart dosimetric parameters with OS in NSCLC patients treated by volumetric modulated arc therapy (VMAT). Methods Inoperable NSCLC patients treated by VMAT from March 2012 to December 2015 were retrospectively reviewed. OS and progression-free survival (PFS) were estimated with the Kaplan-Meier method. Univariate and multivariate analyses were conducted with Cox proportional hazards model. Multivariate model building was conducted using stepwise regression for variables with p-value smaller than 0.2 in the univariate analysis. Results There were 130 NSCLC patients enrolled in this study with a median age of 63 years (range from 34 to 82 y). The median prescription dose for these patients was 56 Gy (range 40-70 Gy) with a mean heart and lung dose of 14.8±8.5 Gy and 13.6±4.4 Gy, respectively. The rates of patients with above grade III radiation pneumonitis (RP) and fibrosis were 8.5% and 8.5%, respectively. The 2-year PFS and OS of these patients were 15.2% and 39.8%, with a median PFS and OS of 7.2 and 18.8 months, respectively. RP was correlated with OS (p=0.048) and lung V20 was associated with PFS (p=0.04) according to the univariate analysis. Multivariate analysis demonstrated that RP (HR 1.39, 95%CI 1.010-1.909, p=0.043) and heart V15 (HR 1.02, 95%CI 1.006-1.025 p=0.002) were progression factors of OS, and no factor was associated with PFS. Conclusions RP and heart V15 were associated with OS for patients with stage III NSCLC who underwent VMAT. Heart and lung dosimetric parameters were highly correlated with each other, sparing of heart and lung should be considered equally during the treatment planning.

Background: Published data on the effects and toxicities of volumetric modulated arc therapy (VMAT) in the management of inoperable lung cancer are scarce. Materials and methods: The clinical outcomes and pulmonary toxicities of 134 patients with consecutive inoperable lung cancer who underwent VMAT from March 2011 to September 2016 were retrospectively reviewed. The dosimetric and characteristic factors associated with acute radiation pneumonitis (RP) and pulmonary fibrosis were evaluated with univariate and multivariate analysis. Results: The average prescription doses to these 134 patients were 57.07±6.27 Gy (range 52-64 Gy). The overall median follow-up time was 18.6 months (range, 2-45 mo), with a median follow-up time for the surviving patients of 20 months (range, 7-45 mo). The 2-year progression-free survival (PFS) and overall survival (OS) for all patients were 18.2% and 38.4%, with a median PFS and OS of 7.6 months and 18.6 months, respectively. The percent of patients with grade III/higher RP and pulmonary fibrosis were 10.5% and 9.0%, respectively. V13 (p=0.02) and age (p=0.02) were independently associated with acute RP according to multivariate analysis. The constraints for lung dosimetric metrics V10,V13,V20 and V30 were approximately 49%,41%,26% and 17% in VMAT treatment of lung cancer to limit the RP rate < 10%. Conclusion: VMAT can be delivered safely with acceptable acute and late toxicities for lung cancer patients. Lung dosimetric metrics were valuable in predicting acute RP. A lung V13 constraint of 40% was helpful to limit the RP rate < 10% in VMAT treatment of lung cancer patients.

Medical physicists are a tiny part of the medical staff group. Medical physics activities are categorized into administrative, clinical services, education, quality assurance (QA), and radiation safety. Medical physicists in radiation oncology departments perform the acceptance and commissioning of new equipment, dosimetry services, treatment planning, patient chart review, patient-specific QA, etc. Due to their professional characteristics, medical physicists cannot go to the front line in the battle of COVID-19. The outbreak of the pandemic in late 2019 is a global disaster and its great impact in Asia-Pacific has been unmasked in economic and health crises. COVID-19 can create devastating social, economic, and political crises that will leave deep scars. Asia-Pacific’s governments and societies should seize the historic opportunity to strengthen public health systems, make online education safer and more accessible, address poverty and social exclusion, tackle violence against children, and create greener, more sustainable economies.

Prognostic parameters and models were believed to be helpful in improving the treatment outcome for patients with brain metastasis (BM). The purpose of this study was to investigate the feasibility of computer tomography (CT) radiomics based nomogram to predict the survival of patients with BM from non-small cell lung cancer (NSCLC) treated with whole brain radiotherapy (WBRT). A total of 195 patients with BM from NSCLC who underwent WBRT from January 2012 to December 2016 were retrospectively reviewed. Radiomics features were extracted and selected from pretherapeutic CT images with least absolute shrinkage and selection operator (LASSO) regression. A nomogram was developed and evaluated by integrating radiomics features and clinical factors to predict the survival of individual patient. Five radiomics features were screened out from 105 radiomics features according to the LASSO Cox regression. According to the optimal cutoff value of radiomics score (Rad-score), patients were stratified into low-risk (Rad-score −0.14) groups. Multivariable analysis indicated that sex, karnofsky performance score (KPS) and Rad-score were independent predictors for overall survival (OS). The concordance index (C-index) of the nomogram in the training cohort and validation cohort was 0.726 and 0.660, respectively. An area under curve (AUC) of 0.786 and 0.788 was achieved for the short-term and long-term survival prediction, respectively. In conclusion, the nomogram based on radiomics features from CT images and clinical factors was feasible to predict the OS of BM patients from NSCLC who underwent WBRT.

The increase of proton beam number might provide higher degrees of freedom in the optimization of intensity-modulated proton therapy planning. In this study, we aimed to quantitatively explore the potential benefits of the increased beam number, including dose volume histogram (DVH), linear energy transfer volume histogram, and DVH bandwidth metrics. Twelve patients with lung cancer are retrospectively selected. Four plans were created based on internal target volume (ITV) robust optimization for each patient using the RayStation treatment planning system. Four plans were generated using different numbers (three, five, seven, and nine) of evenly separated coplanar beams. The three-beam plan was considered as the reference plan. Biologically equivalent doses were calculated using both constant relative biological effectiveness (RBE) and variable RBE models, respectively. To evaluate plan quality, DVH metrics in the target [ITV: D2%, CI, HI] and organs-at-risk [Lung: V5Gy[RBE], V20Gy[RBE], V30Gy[RBE]; Heart D2%; Spinal cord D2%] were calculated using both RBE models. To evaluate LET distributions, LET volume histogram metrics [ITV LETmean and LET2%; Lung LETmean and LET2%; Heart LET2%; Spinal cord LET2%] were quantified. To evaluate plan robustness, the metrics using DVH bandwidth [ITV: D2%, D99%; Lung: V5Gy[RBE], V20Gy[RBE], V30Gy[RBE]; Heart D2%; Spinal cord D2%] were also reported. For plan quality, the increase of proton beam number resulted in fewer target hot spots, improved target dose conformity, improved target dose homogeneity, lower median-dose lung volume, and fewer hot spots in spinal cord. As to LET distributions, target mean LET increased significantly as the beam number increased to seven or more. Lung LET hot spots were significantly reduced with the increase of proton beams. With respect to plan robustness, the robustness of target dose coverage, target hot spots, and low-dose lung volume were improved, while the robustness of heart hot spots became worse as the beam number increased to nine. The robustness of cord hot spots became worse using five and seven beams compared to that using three beams. As the proton beam number increased, plan quality and LET distributions were comparable or significantly improved. The robustness of target dose coverage, target dose hot spots, and low-dose lung volume were significantly improved.

It is of high clinical importance to identify the primary lesion and its pathological types for patients with brain metastases (BM). The purpose of this study is to investigate the feasibility and accuracy of differentiating the primary adenocarcinoma (AD) and squamous cell carcinoma (SCC) of non-small-cell lung cancer (NSCLC) for patients with BM based on radiomics from brain contrast-enhanced computer tomography (CECT) images.A total of 144 BM patients (94 male, 50 female) were enrolled in this study with 102 with primary lung AD and 42 with SCC, respectively. Radiomics features from manually contoured tumors were extracted using python. Mann-Whitney U test and the least absolute shrinkage and selection operator (LASSO) logistic regression were applied to select relative radiomics features. Binary logistic regression and support vector machines (SVM) were applied to build models with radiomics features alone and with radiomics features plus age and sex.Fourteen features were selected from a total of 105 radiomics features for the final model building. The area under the curves (AUCs) and accuracy of SVM and binary logistic regression models were 0.765 vs. 0.769, 0.795 vs.0.828, and 0.716 vs. 0.726, 0.768 vs. 0.758, respectively, for models with radiomics features alone and models with radiomics features plus sex and age.Brain CECT radiomics are promising in differentiating primary AD and SCC to achieve optimal therapeutic management in patients with BM from NSCLC.• It is of high clinical importance to identify the primary lesion and its pathological types for patients with brain metastases (BM) to define the prognosis and treatment. • Few studies had investigated the feasibility and accuracy of differentiating the pathological subtypes of primary non-small-cell lung cancer between adenocarcinoma (AD) and squamous cell carcinoma (SCC) for patients with BM based on radiomics from brain contrast-enhanced CT (CECT) images, although CECT images are often the initial imaging modality to screen for metastases and are recommended on equal footing with MRI for the detection of cerebral metastases. • Brain CECT radiomics are promising in differentiating primary AD and SCC to achieve optimal therapeutic management in patients with BM from NSCLC with a highest area under the curve (AUC) of 0.828 and an accuracy of 0.758, respectively.

Radiation therapy is an effective method in the management of esophageal cancer. MicroRNAs (miRNAs) have been reported to play an important role in tumorigenesis. However, the roles of specific miRNAs in radioresistant esophageal cancer remain to be investigated. In present study, the relative expression level of miR-20b-5p and miR-125a-5p were evaluated by quantitative Real-time polymerase chain reaction. Cell counting Kit-8 assay, wound-healing assay, transwell assay were used to assess cell proliferation, cell migration and cell invasion. TUNEL and Annexin V-FITC assays were applied to evaluate cell apoptosis. Dual-luciferase reporter gene assay was conducted to identify direct targets of miRNAs. The protein expression level was assessed by Western blot. The results indicated that miR-20b-5p was increased in radioresistant KYSE-150R cells compared with KYSE-150 cells, whereas miR-125a-5p was downregulated. MiR-20b-5p upregulation promoted cell proliferation, migration, invasion, and the EMT process, and decreased apoptosis by negatively regulating PTEN. MiR-125a-5p inhibited cell proliferation, migration, invasion, the EMT process and it induced apoptosis by negatively regulating IL6R. These data indicate that miR-20b-5p and miR-125a-5p promote tumorigenesis in radioresistant KYSE-150R cells and have the potential to be used as novel therapeutic targets for the treatment of esophageal cancer.

The aim of this study was to investigate the feasibility and sensitivity of using individual volume-based 3D gamma indices for composite dose-volume histogram (DVH)-based intensity-modulated radiation therapy (IMRT) quality assurance (QA). Composite IMRT QA for 15 cervical cancer patients was performed with ArcCHECK. The percentage dosimetric errors (%DEs) of DVH metrics when comparing treatment planning system and QA-reconstructed dose distribution, percentage gamma passing rates (%GPs) with different criteria for individual volumes and global gamma indices were evaluated, as well as their correlations. Receiver operating characteristic (ROC) curves were applied in order to study the sensitivities of the global and individual volume gamma indices. Most %DEs of the DVH metrics were within 3%. The γPTV and γrectum were

Purpose: To investigate the feasibility of partial arc volumetric modulated arc therapy (VMAT) in lung cancer stereotactic body radiotherapy (SBRT), as well the volumetric and dosimetric effects of different internal target volume (ITV) definitions with 4D CT. Methods: Fourteen patients with primary and metastatic lung cancer underwent SBRT were enrolled. Full and partial arc VMAT plans were generated with four different ITVs: ITVall, ITVMIP, ITVAIP and ITV2phases, representing ITVs generated from all 10 respiratory phases, maximum intensity projection (MIP), average intensity projection (AIP), and 2 extreme respiratory phases. Volumetric and dosimetric differences, as well as MU and delivery time were investigated. Results: Partial arc VMAT irradiated more dose at 2 cm away from planning target volume (PTV) (P = 0.002), however, it achieved better protection on mean lung dose , lung V5, spinal cord, heart and esophagus compared with full arc VMAT. The average MU and delivery time of partial arc VMAT were 240 and 1.6 min less than those of full arc VMAT. There were no significant differences on target coverage and organ at risks (OARs) sparing among four ITVs. The average percent volume differences of ITVMIP, ITVAIP and ITV2phases to ITVall were 8.6%, 13.4%, and 25.2%, respectively. Conclusions: Although partial arc VMAT delivered more dose 2 cm out of PTV, it decreases the dose to lung, spinal cord, and esophagus, as well decreased the total MU and delivery time compared with full arc VMAT without sacrificing target coverage. Partial arc VMAT was feasible and more efficient for lung SBRT.

Multimodal medical image registration remains a challenging problem when strong appearance variations and imprecise alignment exist in images. Previous deep network approaches cannot handle such a high degree of variability and prohibit the use of strong geometric constraints. The authors introduce a novel deep architecture that not only produces image representations that are well-suited for this challenging task, but also leverages knowledge of the geometry constraints for robust registration. By enforcing the representations of different modalities living in a common semantic space, they obtain convolutional features tending to respond to object parts consistently across modality. This yields a unique description for all object fragments and allows the end user to know the model's decision process, whereas most existing models remain unclear and difficult to explain. By using the differentiable spatial transformer to compensate transform, they integrate geometric consensus into the cost function to enable end-to-end model optimisation which has not yet been exploited before. The authors evaluate their method on a very challenging medical image dataset. Experiments demonstrate that the proposed method provides a plausible representation and outperforms state-of-the-art approaches by a significant margin.