Improvement of deep learning prediction model in patient-specific QA for VMAT with MLC leaf position map and patient's dose distribution.

PURPOSE: Deep learning-based virtual patient-specific quality assurance (QA) is a novel technique that enables patient QA without measurement. However, this method could be improved by further evaluating the optimal data to be used as input. Therefore, a deep learning-based model that uses multileaf collimator (MLC) information per control point and dose distribution in patient's CT as inputs was developed. METHODS: Overall, 96 volumetric-modulated arc therapy plans generated for prostate cancer treatment were used. We developed a model (Model 1) that can predict measurement-based gamma passing rate (GPR) for a treatment plan using data stored as a map reflecting the MLC leaf position at each control point (MLPM) and data of the dose distribution in patient's CT as inputs. The evaluation of the model was based on the mean absolute error (MAE) and Pearson's correlation coefficient (r) between the measured and predicted GPR. For comparison, we also analyzed models trained with the dose distribution in patient's CT alone (Model 2) and with dose distributions recalculated on a virtual phantom CT (Model 3). RESULTS: At the 2%/2 mm criterion, MAE[%] and r for Model 1, Model 2, and Model 3 were 2.32% ± 0.43% and 0.54 ± 0.03, 2.70% ± 0.26%, and 0.32 ± 0.08, and 2.96% ± 0.23% and 0.24 ± 0.22, respectively; at the 3%/3 mm criterion, these values were 1.25% ± 0.05% and 0.36 ± 0.18, 1.57% ± 0.35% and 0.19 ± 0.20, and 1.39% ± 0.32% and 0.17 ± 0.22, respectively. This result showed that Model 1 exhibited the lowest MAE and highest r at both criteria of 2%/2 mm and 3%3 mm. CONCLUSIONS: These findings showed that a model that combines the MLPM and dose distribution in patient's CT exhibited a better GPR prediction performance compared with the other two studied models.

Radiation pneumonitis prediction model with integrating multiple dose-function features on 4DCT ventilation images.

PURPOSE: Radiation pneumonitis (RP) is dose-limiting toxicity for non-small-cell cancer (NSCLC). This study developed an RP prediction model by integrating dose-function features from computed four-dimensional computed tomography (4DCT) ventilation using the least absolute shrinkage and selection operator (LASSO). METHODS: Between 2013 and 2020, 126 NSCLC patients were included in this study who underwent a 4DCT scan to calculate ventilation images. We computed two sets of candidate dose-function features from (1) the percentage volume receiving > 20 Gy or the mean dose on the functioning zones determined with the lower cutoff percentile ventilation value, (2) the functioning zones determined with lower and upper cutoff percentile ventilation value using 4DCT ventilation images. An RP prediction model was developed by LASSO while simultaneously determining the regression coefficient and feature selection through fivefold cross-validation. RESULTS: We found 39.3 % of our patients had a ≥ grade 2 RP. The mean area under the curve (AUC) values for the developed models using clinical, dose-volume, and dose-function features with a lower cutoff were 0.791, and the mean AUC values with lower and upper cutoffs were 0.814. The relative regression coefficient (RRC) on dose-function features with upper and lower cutoffs revealed a relative impact of dose to each functioning zone to RP. RRCs were 0.52 for the mean dose on the functioning zone, with top 20 % of all functioning zone was two times greater than that of 0.19 for these with 60 %-80 % and 0.17 with 40 %-60 % (P < 0.01). CONCLUSIONS: The introduction of dose-function features computed from functioning zones with lower and upper cutoffs in a machine learning framework can improve RP prediction. The RRC given by LASSO using dose-function features allows for the quantification of the RP impact of dose on each functioning zones and having the potential to support treatment planning on functional image-guided radiotherapy.

A deep learning method for translating 3DCT to SPECT ventilation imaging: First comparison with 81m Kr-gas SPECT ventilation imaging.

PURPOSE: This study aimed to evaluate the accuracy of deep learning (DL)-based computed tomography (CT) ventilation imaging (CTVI). METHODS: A total of 71 cases that underwent single-photon emission CT 81m Kr-gas ventilation (SPECT V) and CT imaging were included. Sixty cases were assigned to the training and validation sets, and the remaining 11 cases were assigned to the test set. To directly transform three-dimensional (3D) CT (free-breathing CT) images to SPECT V images, a DL-based model was implemented based on the U-Net architecture. The input and output data were 3DCT- and SPECT V-masked, respectively, except for whole-lung volumes. These data were rearranged in voxel size, registered rigidly, cropped, and normalized in preprocessing. In addition to a standard estimation method (i.e., without dropout during the estimation process), a Monte Carlo dropout (MCD) method (i.e., with dropout during the estimation process) was used to calculate prediction uncertainty. To evaluate the two models' (CTVIMCD U-Net , CTVIU-Net ) performance, we used fivefold cross-validation for the training and validation sets. To test the final model performances for both approaches, we applied the test set to each trained model and averaged the test prediction results from the five trained models to acquire the mean test result (bagging) for each approach. For the MCD method, the models were predicted repeatedly (sample size = 200), and the average and standard deviation (SD) maps were calculated in each voxel from the predicted results: The average maps were defined as test prediction results in each fold. As an evaluation index, the voxel-wise Spearman rank correlation coefficient (Spearman rs ) and Dice similarity coefficient (DSC) were calculated. The DSC was calculated for three functional regions (high, moderate, and low) separated by an almost equal volume. The coefficient of variation was defined as prediction uncertainty, and these average values were calculated within three functional regions. The Wilcoxon signed-rank test was used to test for a significant difference between the two DL-based approaches. RESULTS: The average indexes with one SD (1SD) between CTVIMCD U-Net and SPECT V were 0.76 ± 0.06, 0.69 ± 0.07, 0.51 ± 0.06, and 0.75 ± 0.04 for Spearman rs , DSChigh , DSCmoderate , and DSClow , respectively. The average indexes with 1SD between CTVIU-Net and SPECT V were 0.72 ± 0.05, 0.66 ± 0.04, 0.48 ± 0.04, and 0.74 ± 0.06 for Spearman rs , DSChigh , DSCmoderate , and DSClow , respectively. These indexes between CTVIMCD U-Net and CTVIU-Net showed no significance difference (Spearman rs , p = 0.175; DSChigh , p = 0.123; DSCmoderate , p = 0.278; DSClow , p = 0.520). The average coefficient of variations with 1SD were 0.27 ± 0.00, 0.27 ± 0.01, and 0.36 ± 0.03 for the high-, moderate-, and low-functional regions, respectively, and the low-functional region showed a tendency to exhibit larger uncertainties than the others. CONCLUSION: We evaluated DL-based framework for estimating lung-functional ventilation images only from CT images. The results indicated that the DL-based approach could potentially be used for lung-ventilation estimation.

Prediction of radiation pneumonitis with machine learning using 4D-CT based dose-function features.

In this article, we highlight the fundamental importance of the simultaneous use of dose-volume histogram (DVH) and dose-function histogram (DFH) features based on functional images calculated from 4-dimensional computed tomography (4D-CT) and deformable image registration (DIR) in developing a multivariate radiation pneumonitis (RP) prediction model. The patient characteristics, DVH features and DFH features were calculated from functional images by Hounsfield unit (HU) and Jacobian metrics, for an RP grade ≥ 2 multivariate prediction models were computed from 85 non-small cell lung cancer patients. The prediction model is developed using machine learning via a kernel-based support vector machine (SVM) machine. In the patient cohort, 21 of the 85 patients (24.7%) presented with RP grade ≥ 2. The median area under curve (AUC) was 0.58 for the generated 50 prediction models with patient clinical features and DVH features. When HU metric and Jacobian metric DFH features were added, the AUC improved to 0.73 and 0.68, respectively. We conclude that predictive RP models that incorporate DFH features were successfully developed via kernel-based SVM. These results demonstrate that effectiveness of the simultaneous use of DVH features and DFH features calculated from 4D-CT and DIR on functional image-guided radiotherapy.

Error detection model developed using a multi-task convolutional neural network in patient-specific quality assurance for volumetric-modulated arc therapy.

PURPOSE: In patient-specific quality assurance (QA) for static beam intensity-modulated radiation therapy (IMRT), machine-learning-based dose analysis methods have been developed to identify the cause of an error as an alternative to gamma analysis. Although these new methods have revealed that the cause of the error can be identified by analyzing the dose distribution obtained from the two-dimensional detector, they have not been extended to the analysis of volumetric-modulated arc therapy (VMAT) QA. In this study, we propose a deep learning approach to detect various types of errors in patient-specific VMAT QA. METHODS: A total of 161 beams from 104 prostate VMAT plans were analyzed. All beams were measured using a cylindrical detector (Delta4; ScandiDos, Uppsala, Sweden), and predicted dose distributions in a cylindrical phantom were calculated using a treatment planning system (TPS). In addition to the error-free plan, we simulated 12 types of errors: two types of multileaf collimator positional errors (systematic or random leaf offset of 2 mm), two types of monitor unit (MU) scaling errors (±3%), two types of gantry rotation errors (±2° in clockwise and counterclockwise direction), and six types of phantom setup errors (±1 mm in lateral, longitudinal, and vertical directions). The error-introduced predicted dose distributions were created by editing the calculated dose distributions using a TPS with in-house software. Those 13 types of dose difference maps, consisting of an error-free map and 12 error maps, were created from the measured and predicted dose distributions and were used to train the convolutional neural network (CNN) model. Our model was a multi-task model that individually detected each of the 12 types of errors. Two datasets, Test sets 1 and 2, were prepared to evaluate the performance of the model. Test set 1 consisted of 13 types of dose maps used for training, whereas Test set 2 included the dose maps with 25 types of errors in addition to the error-free dose map. The dose map, which introduced 25 types of errors, was generated by combining two of the 12 types of simulated errors. For comparison with the performance of our model, gamma analysis was performed for Test sets 1 and 2 with the criteria set to 3%/2 mm and 2%/1 mm (dose difference/distance to agreement). RESULTS: For Test set 1, the overall accuracy of our CNN model, gamma analysis with the criteria set to 3%/2 mm, and gamma analysis with the criteria set to 2%/1 mm was 0.92, 0.19, and 0.81, respectively. Similarly, for Test set 2, the overall accuracy was 0.44, 0.42, and 0.95, respectively. Our model outperformed gamma analysis in the classification of dose maps containing a single type error, and the performance of our model was inferior in the classification of dose maps containing compound errors. CONCLUSIONS: A multi-task CNN model for detecting errors in patient-specific VMAT QA using a cylindrical measuring device was constructed, and its performance was evaluated. Our results demonstrate that our model was effective in identifying the error type in the dose map for VMAT QA.

Systematic method for a deep learning-based prediction model for gamma evaluation in patient-specific quality assurance of volumetric modulated arc therapy.

PURPOSE: This study aimed to develop and evaluate a novel strategy for establishing a deep learning-based gamma passing rate (GPR) prediction model for volumetric modulated arc therapy (VMAT) using dummy target plan data, one measurement process, and a multicriteria prediction method. METHODS: A total of 147 VMAT plans were used for the training set (two sets of 48 dummy target plans) and test set (51 clinical target plans). The dummy plans were measured using a diode array detector. We developed an original convolutional neural network that accepts coronal and sagittal dose distributions to predict the GPRs of 36 pairs of gamma criteria from 0.5%/0.5 mm to 3%/3 mm. Sixfold cross-validation and model averaging were performed, and the mean training result and mean test result were derived from six trained models that were produced during cross-validation. RESULTS: Strong or moderate correlations were observed between the measured and predicted GPRs in all criteria. The mean absolute errors and root mean squared errors of the test set (clinical target plan) were 0.63 and 1.11 in 3%/3 mm, 1.16 and 1.73 in 3%/2 mm, 1.96 and 2.66 in 2%/2 mm, 5.00 and 6.35 in 1%/1 mm, and 5.42 and 6.78 in 0.5%/1 mm, respectively. The Pearson correlation coefficients were 0.80 in the training set and 0.68 in the test set at the 0.5%/1 mm criterion. CONCLUSION: Our results suggest that the training of the deep learning-based quality assurance model can be performed using a dummy target plan.

Dose distribution correction for the influence of magnetic field using a deep convolutional neural network for online MR-guided adaptive radiotherapy.

PURPOSE: This study aimed to develop a deep convolutional neural network (CNN)-based dose distribution conversion approach for the correction of the influence of a magnetic field for online MR-guided adaptive radiotherapy. METHODS: Our model is based on DenseNet and consists of two 2D input channels and one 2D output channel. These three types of data comprise dose distributions without a magnetic field (uncorrected), electron density (ED) maps, and dose distributions with a magnetic field. These data were generated as follows: both types of dose distributions were created using 15-field IMRT in the same conditions except for the presence or absence of a magnetic field with the GPU Monte Carlo dose in Monaco version 5.4; ED maps were acquired with planning CT images using a clinical CT-to-ED table at our institution. Data for 50 prostate cancer patients were used; 30 patients were allocated for training, 10 for validation, and 10 for testing using 4-fold cross-validation based on rectum gas volume. The accuracy of the model was evaluated by comparing 2D gamma-indexes against the dose distributions in each irradiation field with a magnetic field (true). RESULTS: The gamma indexes in the body for CNN-corrected uncorrected dose against the true dose were 94.95% ± 4.69% and 63.19% ± 3.63%, respectively. The gamma indexes with 2%/2-mm criteria were improved by 10% in most test cases (99.36%). CONCLUSIONS: Our results suggest that the CNN-based approach can be used to correct the dose-distribution influences with a magnetic field in prostate cancer treatment.

Evaluation of four-dimensional cone beam computed tomography ventilation images acquired with two different linear accelerators at various gantry speeds using a deformable lung phantom.

We evaluated four-dimensional cone beam computed tomography (4D-CBCT) ventilation images (VICBCT) acquired with two different linear accelerator systems at various gantry speeds using a deformable lung phantom. The 4D-CT and 4D-CBCT scans were performed using a computed tomography (CT) scanner, an X-ray volume imaging system (Elekta XVI) mounted in Versa HD, and an On-Board Imager (OBI) system mounted in TrueBeam. Intensity-based deformable image registration (DIR) was performed between peak-exhale and peak-inhale images. VICBCT- and 4D-CT-based ventilation images (VICT) were derived by DIR using two metrics: one based on the Jacobian determinant and one on changes in the Hounsfield unit (HU). Three different DIR regularization values (λ) were used for VICBCT. Correlations between the VICBCT and VICT values were evaluated using voxel-wise Spearman's rank correlation coefficient (r). In case of both metrics, the Jacobian-based VICBCT with a gantry speed of 0.6 deg/sec in Versa HD showed the highest correlation for all the gantry speeds (e.g., λ = 0.05 and r = 0.68). Thus, the r value of the Jacobian-based VICBCT was greater or equal to that of the HU-based VICBCT. In addition, the ventilation accuracy of VICBCT increased at low gantry speeds. Thus, the image quality of VICBCT was affected by the change in gantry speed in both the imaging systems. Additionally, DIR regularization considerably influenced VICBCT in both the imaging systems. Our results have the potential to assist in designing CBCT protocols, incorporating VICBCT imaging into the functional avoidance planning process.

Multi-atlas-based auto-segmentation for prostatic urethra using novel prediction of deformable image registration accuracy.

PURPOSE: Accurate identification of the prostatic urethra and bladder can help determine dosing and evaluate urinary toxicity during intensity-modulated radiation therapy (IMRT) planning in patients with localized prostate cancer. However, it is challenging to locate the prostatic urethra in planning computed tomography (pCT). In the present study, we developed a multiatlas-based auto-segmentation method for prostatic urethra identification using deformable image registration accuracy prediction with machine learning (ML) and assessed its feasibility. METHODS: We examined 120 patients with prostate cancer treated with IMRT. All patients underwent temporary urinary catheter placement for identification and contouring of the prostatic urethra in pCT images (ground truth). Our method comprises the following three steps: (a) select four atlas datasets from the atlas datasets using the deformable image registration (DIR) accuracy prediction model, (b) deform them by structure-based DIR, (3) and propagate urethra contour using displacement vector field calculated by the DIR. In (a), for identifying suitable datasets, we used the trained support vector machine regression (SVR) model and five feature descriptors (e.g., prostate volume) to increase DIR accuracy. This method was trained/validated using 100 patients and performance was evaluated within an independent test set of 20 patients. Fivefold cross-validation was used to optimize the hype parameters of the DIR accuracy prediction model. We assessed the accuracy of our method by comparing it with those of two others: Acostas method-based patient selection (previous study method, by Acosta et al.), and the Waterman's method (defines the prostatic urethra based on the center of the prostate, by Waterman et al.). We used the centerlines distance (CLD) between the ground truth and the predicted prostatic urethra as the evaluation index. RESULTS: The CLD in the entire prostatic urethra was 2.09 ± 0.89 mm (our proposed method), 2.77 ± 0.99 mm (Acosta et al., P = 0.022), and 3.47 ± 1.19 mm (Waterman et al., P < 0.001); our proposed method showed the highest accuracy. In segmented CLD, CLD in the top 1/3 segment was highly improved from that of Waterman et.al. and was slightly improved from that of Acosta et.al., with results of 2.49 ± 1.78 mm (our proposed method), 2.95 ± 1.75 mm (Acosta et al., P = 0.42), and 5.76 ± 3.09 mm (Waterman et al., P < 0.001). CONCLUSIONS: We developed a DIR accuracy prediction model-based multiatlas-based auto-segmentation method for prostatic urethra identification. Our method identified prostatic urethra with mean error of 2.09 mm, likely due to combined effects of SVR model employment in patient selection, modified atlas dataset characteristics and DIR algorithm. Our method has potential utility in prostate cancer IMRT and can replace use of temporary indwelling urinary catheters.

Homology-based radiomic features for prediction of the prognosis of lung cancer based on CT-based radiomics.

PURPOSE: Radiomics is a new technique that enables noninvasive prognostic prediction by extracting features from medical images. Homology is a concept used in many branches of algebra and topology that can quantify the contact degree. In the present study, we developed homology-based radiomic features to predict the prognosis of non-small-cell lung cancer (NSCLC) patients and then evaluated the accuracy of this prediction method. METHODS: Four datasets were used: two to provide training and test data and two for the selection of robust radiomic features. All the datasets were downloaded from The Cancer Imaging Archive (TCIA). In two-dimensional cases, the Betti numbers consist of two values: b0 (zero-dimensional Betti number), which is the number of isolated components, and b1 (one-dimensional Betti number), which is the number of one-dimensional or "circular" holes. For homology-based evaluation, computed tomography (CT) images must be converted to binarized images in which each pixel has two possible values: 0 or 1. All CT slices of the gross tumor volume were used for calculating the homology histogram. First, by changing the threshold of the CT value (range: -150 to 300 HU) for all its slices, we developed homology-based histograms for b0 , b1 , and b1 /b0 using binarized images. All histograms were then summed, and the summed histogram was normalized by the number of slices. 144 homology-based radiomic features were defined from the histogram. To compare the standard radiomic features, 107 radiomic features were calculated using the standard radiomics technique. To clarify the prognostic power, the relationship between the values of the homology-based radiomic features and overall survival was evaluated using LASSO Cox regression model and the Kaplan-Meier method. The retained features with nonzero coefficients calculated by the LASSO Cox regression model were used for fitting the regression model. Moreover, these features were then integrated into a radiomics signature. An individualized rad score was calculated from a linear combination of the selected features, which were weighted by their respective coefficients. RESULTS: When the patients in the training and test datasets were stratified into high-risk and low-risk groups according to the rad scores, the overall survival of the groups was significantly different. The C-index values for the homology-based features (rad score), standard features (rad score), and tumor size were 0.625, 0.603, and 0.607, respectively, for the training datasets and 0.689, 0.668, and 0.667 for the test datasets. This result showed that homology-based radiomic features had slightly higher prediction power than the standard radiomic features. CONCLUSIONS: Prediction performance using homology-based radiomic features had a comparable or slightly higher prediction power than standard radiomic features. These findings suggest that homology-based radiomic features may have great potential for improving the prognostic prediction accuracy of CT-based radiomics. In this result, it is noteworthy that there are some limitations.

A convolutional neural network approach for IMRT dose distribution prediction in prostate cancer patients.

The purpose of the study was to compare a 3D convolutional neural network (CNN) with the conventional machine learning method for predicting intensity-modulated radiation therapy (IMRT) dose distribution using only contours in prostate cancer. In this study, which included 95 IMRT-treated prostate cancer patients with available dose distributions and contours for planning target volume (PTVs) and organs at risk (OARs), a supervised-learning approach was used for training, where the dose for a voxel set in the dataset was defined as the label. The adaptive moment estimation algorithm was employed for optimizing a 3D U-net similar network. Eighty cases were used for the training and validation set in 5-fold cross-validation, and the remaining 15 cases were used as the test set. The predicted dose distributions were compared with the clinical dose distributions, and the model performance was evaluated by comparison with RapidPlan™. Dose-volume histogram (DVH) parameters were calculated for each contour as evaluation indexes. The mean absolute errors (MAE) with one standard deviation (1SD) between the clinical and CNN-predicted doses were 1.10% ± 0.64%, 2.50% ± 1.17%, 2.04% ± 1.40%, and 2.08% ± 1.99% for D2, D98 in PTV-1 and V65 in rectum and V65 in bladder, respectively, whereas the MAEs with 1SD between the clinical and the RapidPlan™-generated doses were 1.01% ± 0.66%, 2.15% ± 1.25%, 5.34% ± 2.13% and 3.04% ± 1.79%, respectively. Our CNN model could predict dose distributions that were superior or comparable with that generated by RapidPlan™, suggesting the potential of CNN in dose distribution prediction.

Automated noncoplanar treatment planning strategy in stereotactic radiosurgery of multiple cranial metastases: HyperArc and CyberKnife dose distributions.

The purpose of this study was to evaluate and compare the dosimetric effects of HyperArc-based stereotactic radiosurgery (SRS) and a robotic radiosurgery system-based planning using CyberKnife for multiple cranial metastases. In total, 11 cancer patients with multiple cranial metastases (3 to 5 tumors) treated with CyberKnife were examined. These patients were replanned using HyperArc (Varian Medical Systems, Palo Alto, USA). HyperArc plan were designed using 4 noncoplanar arc single-isocenter VMAT in 6 MV flattening filter free mode for simulated delivery with the True beam STx (Varian). The prescription dose was 23 Gy at single fraction. Dosimetric differences and blinded clinician scoring differences were evaluated. Conformity index (CI) and gradient index (GI) were 0.60 ± 0.11 and 3.94 ± 0.74, respectively, for the CyberKnife plan and 0.87 ± 0.08 and 5.31 ± 1.42, respectively, for the HyperArc plan (p < 0.05). Total brain V12-gross tumor volumes (GTVs) for the CyberKnife and HyperArc plans were 5.26 ± 2.83 and 4.02 ± 1.71 cm3, respectively. These results indicate that HyperArc plan showed better CI and total brain V12-GTV, while CyberKnife plan showed better GI. A blinded physician scoring evaluation did not show significant differences between CyberKnife and HyperArc plans. The HyperArc-based SRS plan is comparable with the CyberKnife plan, suggesting a greater potential to emerge as a suitable tool for SRS of multiple brain metastases.

Investigation of thoracic four-dimensional CT-based dimension reduction technique for extracting the robust radiomic features.

Robust feature selection in radiomic analysis is often implemented using the RIDER test-retest datasets. However, the CT Protocol between the facility and test-retest datasets are different. Therefore, we investigated possibility to select robust features using thoracic four-dimensional CT (4D-CT) scans that are available from patients receiving radiation therapy. In 4D-CT datasets of 14 lung cancer patients who underwent stereotactic body radiotherapy (SBRT) and 14 test-retest datasets of non-small cell lung cancer (NSCLC), 1170 radiomic features (shape: n = 16, statistics: n = 32, texture: n = 1122) were extracted. A concordance correlation coefficient (CCC) > 0.85 was used to select robust features. We compared the robust features in various 4D-CT group with those in test-retest. The total number of robust features was a range between 846/1170 (72%) and 970/1170 (83%) in all 4D-CT groups with three breathing phases (40%-60%); however, that was a range between 44/1170 (4%) and 476/1170 (41%) in all 4D-CT groups with 10 breathing phases. In test-retest, the total number of robust features was 967/1170 (83%); thus, the number of robust features in 4D-CT was almost equal to that in test-retest by using 40-60% breathing phases. In 4D-CT, respiratory motion is a factor that greatly affects the robustness of features, thus by using only 40-60% breathing phases, excessive dimension reduction will be able to be prevented in any 4D-CT datasets, and select robust features suitable for CT protocol of your own facility.

A deep learning-based prediction model for gamma evaluation in patient-specific quality assurance.

PURPOSE: Patient-specific quality assurance (QA) measurement is conducted to confirm the accuracy of dose delivery. However, measurement is time-consuming and places a heavy workload on the medical physicists and radiological technologists. In this study, we proposed a prediction model for gamma evaluation, based on deep learning. We applied the model to a QA measurement dataset of prostate cancer cases to evaluate its practicality. METHODS: Sixty pretreatment verification plans from prostate cancer patients treated using intensity modulated radiation therapy were collected. Fifteen-layer convolutional neural networks (CNN) were developed to learn the sagittal planar dose distributions from a RT-3000 QA phantom (R-TECH.INC., Tokyo, Japan). The percentage gamma passing rate (GPR) was measured using GAFCHROMIC EBT3 film (Ashland Specialty Ingredients, Covington, USA). The input training data also included the volume of the PTV (planning target volume), rectum, and overlapping region, measured in cm3 , and the monitor unit values for each field. The network produced predicted GPR values at four criteria: 2%(global)/2 mm, 3%(global)/2 mm, 2%(global)/3 mm, and 3%(global)/3 mm. Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, was used for learning and for optimizing the CNN-based model. Fivefold cross-validation was applied to validate the performance of the proposed method. Forty cases were used for training and validation set in fivefold cross-validation, and the remaining 20 cases were used for the test set. The predicted and measured GPR values were compared. RESULTS: A linear relationship was found between the measured and predicted values, for each of the four criteria. Spearman rank correlation coefficients in validation set between measured and predicted GPR values at four criteria were 0.73 at 2%/2 mm, 0.72 at 3%/2 mm, 0.74 at 2%/3 mm, and 0.65 at 3%/3 mm, respectively (P < 0.01). The Spearman rank correlation coefficients in the test set were 0.62 (P < 0.01) at 2%/2 mm, 0.56 (P < 0.01) at 3%/2 mm, 0.51 (P = 0.02) at 2%/3 mm, and 0.32 (P = 0.16) at 3%/3 mm. These results demonstrated a strong or moderate correlation between the predicted and measured values. CONCLUSIONS: We developed a CNN-based prediction model for patient-specific QA of dose distribution in prostate treatment. Our results suggest that deep learning may provide a useful prediction model for gamma evaluation of patient-specific QA in prostate treatment planning.

Automated prediction of dosimetric eligibility of patients with prostate cancer undergoing intensity-modulated radiation therapy using a convolutional neural network.

The quality of radiotherapy has greatly improved due to the high precision achieved by intensity-modulated radiation therapy (IMRT). Studies have been conducted to increase the quality of planning and reduce the costs associated with planning through automated planning method; however, few studies have used the deep learning method for optimization of planning. The purpose of this study was to propose an automated method based on a convolutional neural network (CNN) for predicting the dosimetric eligibility of patients with prostate cancer undergoing IMRT. Sixty patients with prostate cancer who underwent IMRT were included in the study. Treatment strategy involved division of the patients into two groups, namely, meeting all dose constraints and not meeting all dose constraints, by experienced medical physicists. We used AlexNet (i.e., one of common CNN architectures) for CNN-based methods to predict the two groups. An AlexNet CNN pre-trained on ImageNet was fine-tuned. Two dataset formats were used as input data: planning computed tomography (CT) images and structure labels. Five-fold cross-validation was used, and performance metrics included sensitivity, specificity, and prediction accuracy. Class activation mapping was used to visualize the internal representation learned by the CNN. Prediction accuracies of the model with the planning CT image dataset and that with the structure label dataset were 56.7 ± 9.7% and 70.0 ± 11.3%, respectively. Moreover, the model with structure labels focused on areas associated with dose constraints. These results revealed the potential applicability of deep learning to the treatment planning of patients with prostate cancer undergoing IMRT.

Evaluation of functionally weighted dose-volume parameters for thoracic stereotactic ablative radiotherapy (SABR) using CT ventilation.

For the purpose of reducing radiation pneumontisis (RP), four-dimensional CT (4DCT)-based ventilation can be used to reduce functionally weighted lung dose. This study aimed to evaluate the functionally weighted dose-volume parameters and to investigate an optimal weighting method to realize effective planning optimization in thoracic stereotactic ablative radiotherapy (SABR). Forty patients treated with SABR were analyzed. Ventilation images were obtained from 4DCT using deformable registration and Hounsfield unit-based calculation. Functionally-weighted mean lung dose (fMLD) and functional lung fraction receiving at least x Gy (fVx) were calculated by two weighting methods: thresholding and linear weighting. Various ventilation thresholds (5th-95th, every 5th percentile) were tested. The predictive accuracy for CTCAE grade ≧ 2 pneumonitis was evaluated by area under the curve (AUC) of receiver operating characteristic analysis. AUC values varied from 0.459 to 0.570 in accordance with threshold and dose-volume metrics. A combination of 25th percentile threshold and fV30 showed the best result (AUC: 0.570). AUC values with fMLD, fV10, fV20, and fV40 were 0.541, 0.487, 0.548 and 0.563 using a 25th percentile threshold. Although conventional MLD, V10, V20, V30 and V40 showed lower AUC values (0.516, 0.477, 0.534, 0.552 and 0.527), the differences were not statistically significant. fV30 with 25th percentile threshold was the best predictor of RP. Our results suggested that the appropriate weighting should be used for better treatment outcomes in thoracic SABR.

First-order superconducting transition of Sr2RuO4.

By means of the magnetocaloric effect, we examine the nature of the superconducting-normal (S-N) transition of Sr(2)RuO(4), a most promising candidate for a spin-triplet superconductor. We provide thermodynamic evidence that the S-N transition of this oxide is of first order below approximately 0.8 K and only for magnetic field directions very close to the conducting plane, in clear contrast to the ordinary type-II superconductors exhibiting second-order S-N transitions. The entropy release across the transition at 0.2 K is 10% of the normal-state entropy. Our result urges an introduction of a new mechanism to break superconductivity by magnetic field.