Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “PCA”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 739 records · Page 41
All sourcesNASA NTRSDOE OSTI
DOE OSTI2013

A technique for estimating 4D-CBCT using prior knowledge and limited-angle projections

Purpose: To develop a technique to estimate onboard 4D-CBCT using prior information and limited-angle projections for potential 4D target verification of lung radiotherapy.Methods: Each phase of onboard 4D-CBCT is considered as a deformation from one selected phase (prior volume) of the planning 4D-CT. The deformation field maps (DFMs) are solved using a motion modeling and free-form deformation (MM-FD) technique. In the MM-FD technique, the DFMs are estimated using a motion model which is extracted from planning 4D-CT based on principal component analysis (PCA). The motion model parameters are optimized by matching the digitally reconstructed radiographs of the deformed volumes to the limited-angle onboard projections (data fidelity constraint). Afterward, the estimated DFMs are fine-tuned using a FD model based on data fidelity constraint and deformation energy minimization. The 4D digital extended-cardiac-torso phantom was used to evaluate the MM-FD technique. A lung patient with a 30 mm diameter lesion was simulated with various anatomical and respirational changes from planning 4D-CT to onboard volume, including changes of respiration amplitude, lesion size and lesion average-position, and phase shift between lesion and body respiratory cycle. The lesions were contoured in both the estimated and “ground-truth” onboard 4D-CBCT for comparison. 3D volume percentage-difference (VPD) and center-of-mass shift (COMS) were calculated to evaluate the estimation accuracy of three techniques: MM-FD, MM-only, and FD-only. Different onboard projection acquisition scenarios and projection noise levels were simulated to investigate their effects on the estimation accuracy.Results: For all simulated patient and projection acquisition scenarios, the mean VPD (±S.D.)/COMS (±S.D.) between lesions in prior images and “ground-truth” onboard images were 136.11% (±42.76%)/15.5 mm (±3.9 mm). Using orthogonal-view 15°-each scan angle, the mean VPD/COMS between the lesion in estimated and “ground-truth” onboard images for MM-only, FD-only, and MM-FD techniques were 60.10% (±27.17%)/4.9 mm (±3.0 mm), 96.07% (±31.48%)/12.1 mm (±3.9 mm) and 11.45% (±9.37%)/1.3 mm (±1.3 mm), respectively. For orthogonal-view 30°-each scan angle, the corresponding results were 59.16% (±26.66%)/4.9 mm (±3.0 mm), 75.98% (±27.21%)/9.9 mm (±4.0 mm), and 5.22% (±2.12%)/0.5 mm (±0.4 mm). For single-view scan angles of 3°, 30°, and 60°, the results for MM-FD technique were 32.77% (±17.87%)/3.2 mm (±2.2 mm), 24.57% (±18.18%)/2.9 mm (±2.0 mm), and 10.48% (±9.50%)/1.1 mm (±1.3 mm), respectively. For projection angular-sampling-intervals of 0.6°, 1.2°, and 2.5° with the orthogonal-view 30°-each scan angle, the MM-FD technique generated similar VPD (maximum deviation 2.91%) and COMS (maximum deviation 0.6 mm), while sparser sampling yielded larger VPD/COMS. With equal number of projections, the estimation results using scattered 360° scan angle were slightly better than those using orthogonal-view 30°-each scan angle. The estimation accuracy of MM-FD technique declined as noise level increased.Conclusions: The MM-FD technique substantially improves the estimation accuracy for onboard 4D-CBCT using prior planning 4D-CT and limited-angle projections, compared to the MM-only and FD-only techniques. It can potentially be used for the inter/intrafractional 4D-localization verification.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2014

SU-D-18C-02: Feasibility of Using a Short ASL Scan for Calibrating Cerebral Blood Flow Obtained From DSC-MRI

Purpose: This study aimed to evaluate the feasibility of using a short arterial spin labeling (ASL) scan for calibrating the dynamic susceptibility contrast- (DSC-) MRI in a group of patients with internal carotid artery stenosis. Methods: Six patients with unilateral ICA stenosis enrolled in the study on a 3T clinical MRI scanner. The ASL-cerebral blood flow (-CBF) maps were calculated by averaging different number of dynamic points (N=1-45) acquired by using a Q2TIPS sequence. For DSC perfusion analysis, arterial input function was selected to derive the relative cerebral blood flow (rCBF) map and the delay (Tmax) map. Patient-specific CF was calculated from the mean ASL- and DSC-CBF obtained from three different masks: (1)Tmax< 3s, (2)combined gray matter mask with mask 1, (3)mask 2 with large vessels removed. One CF value was created for each number of averages by using each of the three masks for calibrating the DSC-CBF map. The CF value of the largest number of averages (NL=45) was used to determine the acceptable range(< 10%, <15%, and <20%) of CF values corresponding to the minimally acceptable number of average (NS) for each patient. Results: Comparing DSC CBF maps corrected by CF values of NL (CBFL) in ACA, MCA and PCA territories, all masks resulted in smaller CBF on the ipsilateral side than the contralateral side of the MCA territory(p<.05). The values obtained from mask 1 were significantly different than the mask 3(p<.05). Using mask 3, the medium values of Ns were 4(<10%), 2(<15%) and 2(<20%), with the worst case scenario (maximum Ns) of 25, 4, and 4, respectively. Conclusion: This study found that reliable calibration of DSC-CBF can be achieved from a short pulsed ASL scan. We suggested use a mask based on the Tmax threshold, the inclusion of gray matter only and the exclusion of large vessels for performing the calibration.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2014

SU-E-J-67: Evaluation of Breathing Patterns for Respiratory-Gated Radiation Therapy Using Respiration Regularity Index

Purpose: Despite the importance of accurately estimating the respiration regularity of a patient in motion compensation treatment, an effective and simply applicable method has rarely been reported. The authors propose a simple respiration regularity index based on parameters derived from a correspondingly simplified respiration model. Methods: In order to simplify a patient's breathing pattern while preserving the data's intrinsic properties, we defined a respiration model as a power of cosine form with a baseline drift. According to this respiration formula, breathing-pattern fluctuation could be explained using four factors: sample standard deviation of respiration period, sample standard deviation of amplitude and the results of simple regression of the baseline drift (slope and standard deviation of residuals of a respiration signal. Overall irregularity (δ) was defined as a Euclidean norm of newly derived variable using principal component analysis (PCA) for the four fluctuation parameters. Finally, the proposed respiration regularity index was defined as ρ=ln(1+(1/ δ))/2, a higher ρ indicating a more regular breathing pattern. Subsequently, we applied it to simulated and clinical respiration signals from real-time position management (RPM; Varian Medical Systems, Palo Alto, CA) and investigated respiration regularity. Moreover, correlations between the regularity of the first session and the remaining fractions were investigated using Pearson's correlation coefficient. Results: The respiration regularity was determined based on ρ; patients with ρ<0.3 showed worse regularity than the others, whereas ρ>0.7 was suitable for respiratory-gated radiation therapy (RGRT). Fluctuations in breathing cycle and amplitude were especially determinative of ρ. If the respiration regularity of a patient's first session was known, it could be estimated through subsequent sessions. Conclusions: Respiration regularity could be objectively determined using a respiration regularity index, ρ. Such single-index testing of respiration regularity can facilitate determination of RGRT availability in clinical settings, especially for free-breathing cases. This work was supported by a Korea Science and Engineering Foundation (KOSEF) grant funded by the Korean Ministry of Science, ICT and Future Planning (No. 2013043498)

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2014

SU-E-T-333: Towards Customizable Radiotherapy Enhancement (CuRE) for Prostate Cancer Using Cisplatin Nanoparticles

Purpose: Replacing routinely used brachytherapy spacers with multifunctional ones loaded with cisplatin nanoparticles (CNP), which can be released into the tumor after implantation, could enable customizable radiation boosting to the prostate tumor in addition to chemotherapy effect. This study investigates the feasibility of customizing the intra-tumor biodistribution and corresponding dose enhancement (DEF) over time for the released CNP as a function of nanoparticle size. Methods: Dose enhancement factors (DEF) due to photon-induced emission of photo-/Auger electrons from CNPs were calculated as a function of concentration using previously published analytical calculation method. An experimentally determined diffusion coefficient (D) for 10 nm nanoparticles in mouse tumor model was employed to estimate D for other sizes using the Stoke- Einstein equation. The error function diffusion model in the experimental study was applied to generate the intra-tumor concentration profile for a burst release of CNPs from the spacer over time. The corresponding DEF profiles were then determined for brachytherapy using Pd-103 and I-125 sources. Results: As expected, the generated profiles showed greater DEF over time for smaller CNP sizes at sample distances from the spacer. For example, for a centrally located spacer, clinically significant DEF (> 20%) could be achieved near the tumor periphery (ca. 0.85 cm distance from the spacer for average PCa tumor size) after 20, and 100 days, respectively for CNPs sizes of 2 nm, and 10 nm, using I-125. Meanwhile for Pd-103, clinically significant DEF could be achieved at the same position after 22 and 108 days, respectively, for same size particles. Conclusion: Our preliminary results demonstrate the feasibility of customizing dose enhancement to prostate tumors as a function of spacer location, brachytherapy source type or size of CNPs released from multifunctional spacers. Such an approach could enable customizable radiation boosting to tumor sub-volumes, while minimizing dose to healthy tissues.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2014

TU-C-17A-10: Patient Features Based Dosimetric Pareto Front Prediction In Esophagus Cancer Radiotherapy

Purpose: The purpose of this study is to study the feasibility of the dosimetric pareto front (PF) prediction based on patient anatomic and dosimetric parameters for esophagus cancer patients. Methods: Sixty esophagus patients in our institution were enrolled in this study. A total 2920 IMRT plans were created to generated PF for each patient. On average, each patient had 48 plans. The anatomic and dosimetric features were extracted from those plans. The mean lung dose (MLD), mean heart dose (MHD), spinal cord max dose and PTV homogeneous index (PTVHI) were recorded for each plan. The principal component analysis (PCA) was used to extract overlap volume histogram (OVH) features between PTV and other critical organs. The full dataset was separated into two parts include the training dataset and the validation dataset. The prediction outcomes were the MHD and MLD for the current study. The spearman rank correlation coefficient was used to evaluate the correlation between the anatomical features and dosimetric features. The PF was fit by the the stepwise multiple regression method. The cross-validation method was used to evaluation the model. Results: The mean prediction error of the MHD was 465 cGy with 100 repetitions. The most correlated factors were the first principal components of the OVH between heart and PTV, and the overlap between heart and PTV in Z-axis. The mean prediction error of the MLD was 195 cGy. The most correlated factors were the first principal components of the OVH between lung and PTV, and the overlap between lung and PTV in Z-axis. Conclusion: It is feasible to use patients anatomic and dosimetric features to generate a predicted PF. Additional samples and further studies were required to get a better prediction model.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2014

TH-E-BRF-07: Raman Spectroscopy for Radiation Treatment Response Assessment in a Lung Metastases Mouse Model

Purpose: Raman spectroscopy is an optical spectroscopic method used to probe chemical information about a target tissue. Our goal was to investigate whether Raman spectroscopy is able to distinguish lung tumors from normal lung tissue and whether this technique can identify the molecular changes induced by radiation. Methods: 4T1 mouse breast cancer cells were implanted subcutaneously into the flanks of 6 Balb/C female mice. Four additional mice were used as “normal lung” controls. After 14 days, 3 mice bearing tumors received 6Gy to the left lung with 6MV photons and the other three were treated as “unirradiated tumor” controls. At a 24-hour time point, lungs were excised and the specimens were sectioned using a cryostat; alternating sections were either stained with hematoxylin and eosin (H and E) for evaluation by a pathologist or unstained for Raman measurements. 240 total Raman spectra were collected; 84 from normal lung controls; 63 from unirradiated tumors and 64 from tumors irradiated with 6Gy in a single fraction. Raman spectra were also collected from normal lung tissues of mice with unirradiated tumors. Principal component analysis (PCA) and discriminant function analysis (DFA) were performed to analyze the data. Results: Raman bands assignable to DNA/RNA showed prominent contributions in tumor tissues while Raman bands associated with hemoglobin showed strong contributions in normal lung tissue. PCA/DFA analysis identified normal lung tissue and tumor with 100% and 98.4% accuracy, respectively, relative to pathologic scoring. Additionally, normal lung tissues from unirradiated mice bearing tumors were classified as normal with 100% accuracy. In a model consisting of unirradiated and irradiated tumors identification accuracy was 79.4% and 93.8% respectively, relative to pathologic assessment. Conclusion: Initial results demonstrate the promise for Raman spectroscopy in the diagnosis normal vs. lung metastases as well as the assessment of radiation therapy response. Further investigations are needed to validate these results. This study was supported in part by a research grant from Varian Medical Systems, Palo Alto, CA.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

SU-E-J-256: Predicting Metastasis-Free Survival of Rectal Cancer Patients Treated with Neoadjuvant Chemo-Radiotherapy by Data-Mining of CT Texture Features of Primary Lesions

Purpose: The purpose of this study is to investigate the relationship between computed tomographic (CT) texture features of primary lesions and metastasis-free survival for rectal cancer patients; and to develop a datamining prediction model using texture features. Methods: A total of 220 rectal cancer patients treated with neoadjuvant chemo-radiotherapy (CRT) were enrolled in this study. All patients underwent CT scans before CRT. The primary lesions on the CT images were delineated by two experienced oncologists. The CT images were filtered by Laplacian of Gaussian (LoG) filters with different filter values (1.0–2.5: from fine to coarse). Both filtered and unfiltered images were analyzed using Gray-level Co-occurrence Matrix (GLCM) texture analysis with different directions (transversal, sagittal, and coronal). Totally, 270 texture features with different species, directions and filter values were extracted. Texture features were examined with Student’s t-test for selecting predictive features. Principal Component Analysis (PCA) was performed upon the selected features to reduce the feature collinearity. Artificial neural network (ANN) and logistic regression were applied to establish metastasis prediction models. Results: Forty-six of 220 patients developed metastasis with a follow-up time of more than 2 years. Sixtyseven texture features were significantly different in t-test (p<0.05) between patients with and without metastasis, and 12 of them were extremely significant (p<0.001). The Area-under-the-curve (AUC) of ANN was 0.72, and the concordance index (CI) of logistic regression was 0.71. The predictability of ANN was slightly better than logistic regression. Conclusion: CT texture features of primary lesions are related to metastasisfree survival of rectal cancer patients. Both ANN and logistic regression based models can be developed for prediction.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

MO-AB-BRA-10: Cancer Therapy Outcome Prediction Based On Dempster-Shafer Theory and PET Imaging

Purpose: In cancer therapy, utilizing FDG-18 PET image-based features for accurate outcome prediction is challenging because of 1) limited discriminative information within a small number of PET image sets, and 2) fluctuant feature characteristics caused by the inferior spatial resolution and system noise of PET imaging. In this study, we proposed a new Dempster-Shafer theory (DST) based approach, evidential low-dimensional transformation with feature selection (ELT-FS), to accurately predict cancer therapy outcome with both PET imaging features and clinical characteristics. Methods: First, a specific loss function with sparse penalty was developed to learn an adaptive low-rank distance metric for representing the dissimilarity between different patients’ feature vectors. By minimizing this loss function, a linear low-dimensional transformation of input features was achieved. Also, imprecise features were excluded simultaneously by applying a l2,1-norm regularization of the learnt dissimilarity metric in the loss function. Finally, the learnt dissimilarity metric was applied in an evidential K-nearest-neighbor (EK- NN) classifier to predict treatment outcome. Results: Twenty-five patients with stage II–III non-small-cell lung cancer and thirty-six patients with esophageal squamous cell carcinomas treated with chemo-radiotherapy were collected. For the two groups of patients, 52 and 29 features, respectively, were utilized. The leave-one-out cross-validation (LOOCV) protocol was used for evaluation. Compared to three existing linear transformation methods (PCA, LDA, NCA), the proposed ELT-FS leads to higher prediction accuracy for the training and testing sets both for lung-cancer patients (100+/−0.0, 88.0+/−33.17) and for esophageal-cancer patients (97.46+/−1.64, 83.33+/−37.8). The ELT-FS also provides superior class separation in both test data sets. Conclusion: A novel DST- based approach has been proposed to predict cancer treatment outcome using PET image features and clinical characteristics. A specific loss function has been designed for robust accommodation of feature set incertitude and imprecision, facilitating adaptive learning of the dissimilarity metric for the EK-NN classifier.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

MO-F-CAMPUS-J-02: Automatic Recognition of Patient Treatment Site in Portal Images Using Machine Learning

Purpose: To investigate the method to automatically recognize the treatment site in the X-Ray portal images. It could be useful to detect potential treatment errors, and to provide guidance to sequential tasks, e.g. automatically verify the patient daily setup. Methods: The portal images were exported from MOSAIQ as DICOM files, and were 1) processed with a threshold based intensity transformation algorithm to enhance contrast, and 2) where then down-sampled (from 1024×768 to 128×96) by using bi-cubic interpolation algorithm. An appearance-based vector space model (VSM) was used to rearrange the images into vectors. A principal component analysis (PCA) method was used to reduce the vector dimensions. A multi-class support vector machine (SVM), with radial basis function kernel, was used to build the treatment site recognition models. These models were then used to recognize the treatment sites in the portal image. Portal images of 120 patients were included in the study. The images were selected to cover six treatment sites: brain, head and neck, breast, lung, abdomen and pelvis. Each site had images of the twenty patients. Cross-validation experiments were performed to evaluate the performance. Results: MATLAB image processing Toolbox and scikit-learn (a machine learning library in python) were used to implement the proposed method. The average accuracies using the AP and RT images separately were 95% and 94% respectively. The average accuracy using AP and RT images together was 98%. Computation time was ∼0.16 seconds per patient with AP or RT image, ∼0.33 seconds per patient with both of AP and RT images. Conclusion: The proposed method of treatment site recognition is efficient and accurate. It is not sensitive to the differences of image intensity, size and positions of patients in the portal images. It could be useful for the patient safety assurance. The work was partially supported by a research grant from Varian Medical System.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

WE-G-BRD-06: Volumetric Cine MRI (VC-MRI) Estimated Based On Prior Knowledge for On-Board Target Localization

Purpose: To develop a technique to generate on-board VC-MRI using patient prior 4D-MRI, motion modeling and on-board 2D-cine MRI for real-time 3D target verification of liver and lung radiotherapy. Methods: The end-expiration phase images of a 4D-MRI acquired during patient simulation are used as patient prior images. Principal component analysis (PCA) is used to extract 3 major respiratory deformation patterns from the Deformation Field Maps (DFMs) generated between end-expiration phase and all other phases. On-board 2D-cine MRI images are acquired in the axial view. The on-board VC-MRI at any instant is considered as a deformation of the prior MRI at the end-expiration phase. The DFM is represented as a linear combination of the 3 major deformation patterns. The coefficients of the deformation patterns are solved by matching the corresponding 2D slice of the estimated VC-MRI with the acquired single 2D-cine MRI. The method was evaluated using both XCAT (a computerized patient model) simulation of lung cancer patients and MRI data from a real liver cancer patient. The 3D-MRI at every phase except end-expiration phase was used to simulate the ground-truth on-board VC-MRI at different instances, and the center-tumor slice was selected to simulate the on-board 2D-cine images. Results: Image subtraction of ground truth with estimated on-board VC-MRI shows fewer differences than image subtraction of ground truth with prior image. Excellent agreement between profiles was achieved. The normalized cross correlation coefficients between the estimated and ground-truth in the axial, coronal and sagittal views for each time step were >= 0.982, 0.905, 0.961 for XCAT data and >= 0.998, 0.911, 0.9541 for patient data. For XCAT data, the maximum-Volume-Percent-Difference between ground-truth and estimated tumor volumes was 1.6% and the maximum-Center-of-Mass-Shift was 0.9 mm. Conclusion: Preliminary studies demonstrated the feasibility to estimate real-time VC-MRI for on-board target localization before or during radiotherapy treatments. National Institutes of Health Grant No. R01-CA184173; Varian Medical System.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

SU-C-BRA-07: Variability of Patient-Specific Motion Models Derived Using Different Deformable Image Registration Algorithms for Lung Cancer Stereotactic Body Radiotherapy (SBRT) Patients

Purpose: To study the variability of patient-specific motion models derived from 4-dimensional CT (4DCT) images using different deformable image registration (DIR) algorithms for lung cancer stereotactic body radiotherapy (SBRT) patients. Methods: Motion models are derived by 1) applying DIR between each 4DCT image and a reference image, resulting in a set of displacement vector fields (DVFs), and 2) performing principal component analysis (PCA) on the DVFs, resulting in a motion model (a set of eigenvectors capturing the variations in the DVFs). Three DIR algorithms were used: 1) Demons, 2) Horn-Schunck, and 3) iterative optical flow. The motion models derived were compared using patient 4DCT scans. Results: Motion models were derived and the variations were evaluated according to three criteria: 1) the average root mean square (RMS) difference which measures the absolute difference between the components of the eigenvectors, 2) the dot product between the eigenvectors which measures the angular difference between the eigenvectors in space, and 3) the Euclidean Model Norm (EMN), which is calculated by summing the dot products of an eigenvector with the first three eigenvectors from the reference motion model in quadrature. EMN measures how well an eigenvector can be reconstructed using another motion model derived using a different DIR algorithm. Results showed that comparing to a reference motion model (derived using the Demons algorithm), the eigenvectors of the motion model derived using the iterative optical flow algorithm has smaller RMS, larger dot product, and larger EMN values than those of the motion model derived using Horn-Schunck algorithm. Conclusion: The study showed that motion models vary depending on which DIR algorithms were used to derive them. The choice of a DIR algorithm may affect the accuracy of the resulting model, and it is important to assess the suitability of the algorithm chosen for a particular application. This project was supported, in part, through a Master Research Agreement with Varian Medical Systems, Inc, Palo Alto, CA.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

SU-D-204-01: A Methodology Based On Machine Learning and Quantum Clustering to Predict Lung SBRT Dosimetric Endpoints From Patient Specific Anatomic Features

Purpose: To develop a data-mining methodology based on quantum clustering and machine learning to predict expected dosimetric endpoints for lung SBRT applications based on patient-specific anatomic features. Methods: Ninety-three patients who received lung SBRT at our clinic from 2011–2013 were retrospectively identified. Planning information was acquired for each patient, from which various features were extracted using in-house semi-automatic software. Anatomic features included tumor-to-OAR distances, tumor location, total-lung-volume, GTV and ITV. Dosimetric endpoints were adopted from RTOG-0195 recommendations, and consisted of various OAR-specific partial-volume doses and maximum point-doses. First, PCA analysis and unsupervised quantum-clustering was used to explore the feature-space to identify potentially strong classifiers. Secondly, a multi-class logistic regression algorithm was developed and trained to predict dose-volume endpoints based on patient-specific anatomic features. Classes were defined by discretizing the dose-volume data, and the feature-space was zero-mean normalized. Fitting parameters were determined by minimizing a regularized cost function, and optimization was performed via gradient descent. As a pilot study, the model was tested on two esophageal dosimetric planning endpoints (maximum point-dose, dose-to-5cc), and its generalizability was evaluated with leave-one-out cross-validation. Results: Quantum-Clustering demonstrated a strong separation of feature-space at 15Gy across the first-and-second Principle Components of the data when the dosimetric endpoints were retrospectively identified. Maximum point dose prediction to the esophagus demonstrated a cross-validation accuracy of 87%, and the maximum dose to 5cc demonstrated a respective value of 79%. The largest optimized weighting factor was placed on GTV-to-esophagus distance (a factor of 10 greater than the second largest weighting factor), indicating an intuitively strong correlation between this feature and both endpoints. Conclusion: This pilot study shows that it is feasible to predict dose-volume endpoints based on patient-specific anatomic features. The developed methodology can potentially help to identify patients at risk for higher OAR doses, thus improving the efficiency of treatment planning. R01-184173.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

SU-D-201-05: On the Automatic Recognition of Patient Safety Hazards in a Radiotherapy Setup Using a Novel 3D Camera System and a Deep Learning Framework

Purpose: Patient safety hazards such as a wrong patient/site getting treated can lead to catastrophic results. The purpose of this project is to automatically detect potential patient safety hazards during the radiotherapy setup and alert the therapist before the treatment is initiated. Methods: We employed a set of co-located and co-registered 3D cameras placed inside the treatment room. Each camera provided a point-cloud of fraxels (fragment pixels with 3D depth information). Each of the cameras were calibrated using a custom-built calibration target to provide 3D information with less than 2 mm error in the 500 mm neighborhood around the isocenter. To identify potential patient safety hazards, the treatment room components and the patient’s body needed to be identified and tracked in real-time. For feature recognition purposes, we used a graph-cut based feature recognition with principal component analysis (PCA) based feature-to-object correlation to segment the objects in real-time. Changes in the object’s position were tracked using the CamShift algorithm. The 3D object information was then stored for each classified object (e.g. gantry, couch). A deep learning framework was then used to analyze all the classified objects in both 2D and 3D and was then used to fine-tune a convolutional network for object recognition. The number of network layers were optimized to identify the tracked objects with >95% accuracy. Results: Our systematic analyses showed that, the system was effectively able to recognize wrong patient setups and wrong patient accessories. The combined usage of 2D camera information (color + depth) enabled a topology-preserving approach to verify patient safety hazards in an automatic manner and even in scenarios where the depth information is partially available. Conclusion: By utilizing the 3D cameras inside the treatment room and a deep learning based image classification, potential patient safety hazards can be effectively avoided.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

SU-F-J-199: Predictive Models for Cone Beam CT-Based Online Verification of Pencil Beam Scanning Proton Therapy

Purpose: To utilize online CBCT scans to develop models for predicting DVH metrics in proton therapy of head and neck tumors. Methods: Nine patients with locally advanced oropharyngeal cancer were retrospectively selected in this study. Deformable image registration was applied to the simulation CT, target volumes, and organs at risk (OARs) contours onto each weekly CBCT scan. Intensity modulated proton therapy (IMPT) treatment plans were created on the simulation CT and forward calculated onto each corrected CBCT scan. Thirty six potentially predictive metrics were extracted from each corrected CBCT. These features include minimum/maximum/mean over and under-ranges at the proximal and distal surface of PTV volumes, and geometrical and water equivalent distance between PTV and each OARs. Principal component analysis (PCA) was used to reduce the dimension of the extracted features. Three principal components were found to account for over 90% of variances in those features. Datasets from eight patients were used to train a machine learning model to fit these principal components with DVH metrics (dose to 95% and 5% of PTV, mean dose or max dose to OARs) from the forward calculated dose on each corrected CBCT. The accuracy of this model was verified on the datasets from the 9th patient. Results: The predicted changes of DVH metrics from the model were in good agreement with actual values calculated on corrected CBCT images. Median differences were within 1 Gy for most DVH metrics except for larynx and constrictor mean dose. However, a large spread of the differences was observed, indicating additional training datasets and predictive features are needed to improve the model. Conclusion: Intensity corrected CBCT scans hold the potential to be used for online verification of proton therapy and prediction of delivered dose distributions.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

SU-G-TeP1-09: Modality-Specific Dose Gradient Modeling for Prostate IMRT Using Spherical Distance Maps of PTV and Isodose Contours

Purpose: Overlapping volume histogram (OVH) and distance-to-target histogram (DTH) calculations rely on the assumption that dose gradients are symmetric with respect to primary target volume (PTV) expansion and minimum distance to PTV surface, respectively. It is desirable to lift this assumption and instead account for achievable modality-specific dose gradients (MSDG) for a given PTV shape. Methods: From a library of 96 prostate 7-beam IMRT plans, we computed spherical distance maps (SDMs) for PTVs and 3 iso-dose contours. Each SDM contains the minimum distances between plan isocenter and the object’s surface for a fixed set of azimuthal and polar angles. We performed principal component analysis (PCA)-based missing data recovery with PTV SDM as input and a single iso-dose contour SDM as output. Repeating this process for the set of iso-dose contours sparsely reconstructed the MSDG for a given PTV. DVH points were computed from the MSDG for bladder and rectum (OARs) as a natural way of casting patient-specific geometric information into modality-specific dose space (vs. OVH and DTH where data is purely geometric). For comparison, we implemented a cumulative (c)DTH-based prediction algorithm, and produced DVH for both OARs separately. We then computed linear regressions between each method and the DVH of the original patient plan for three different OAR DVH points. Results: R-squared for MSD-Gcomputed DVH vs. original plan DVH at V90%, V80% and V60% of max dose were 0.86, 0.83, and 0.78, for bladder and 0.36, 0.52, and 0.55 for rectum, respectively; R-squared for cDTH-based predicted DVH vs. original plan DVH were 0.67, 0.81, and 0.83, for bladder and 0.44, 0.50, and 0.51, for rectum, respectively. Conclusion: By simply modeling the MSDG about a given PTV, we are able to reproduce DVHs consistent with a validated cDTH-based DVH prediction. Regression analysis suggests MSDG could further improve predictions.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

MO-DE-207B-03: Improved Cancer Classification Using Patient-Specific Biological Pathway Information Via Gene Expression Data

Purpose: To develop an efficient, pathway-based classification system using network biology statistics to assist in patient-specific response predictions to radiation and drug therapies across multiple cancer types. Methods: We developed PICS (Pathway Informed Classification System), a novel two-step cancer classification algorithm. In PICS, a matrix m of mRNA expression values for a patient cohort is collapsed into a matrix p of biological pathways. The entries of p, which we term pathway scores, are obtained from either principal component analysis (PCA), normal tissue centroid (NTC), or gene expression deviation (GED). The pathway score matrix is clustered using both k-means and hierarchical clustering, and a clustering is judged by how well it groups patients into distinct survival classes. The most effective pathway scoring/clustering combination, per clustering p-value, thus generates various ‘signatures’ for conventional and functional cancer classification. Results: PICS successfully regularized large dimension gene data, separated normal and cancerous tissues, and clustered a large patient cohort spanning six cancer types. Furthermore, PICS clustered patient cohorts into distinct, statistically-significant survival groups. For a suboptimally-debulked ovarian cancer set, the pathway-classified Kaplan-Meier survival curve (p = .00127) showed significant improvement over that of a prior gene expression-classified study (p = .0179). For a pancreatic cancer set, the pathway-classified Kaplan-Meier survival curve (p = .00141) showed significant improvement over that of a prior gene expression-classified study (p = .04). Pathway-based classification confirmed biomarkers for the pyrimidine, WNT-signaling, glycerophosphoglycerol, beta-alanine, and panthothenic acid pathways for ovarian cancer. Despite its robust nature, PICS requires significantly less run time than current pathway scoring methods. Conclusion: This work validates the PICS method to improve cancer classification using biological pathways. Patients are classified with greater specificity and physiological relevance as compared to current gene-specific approaches. Focus now moves to utilizing PICS for pan-cancer patient-specific treatment response prediction.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

TU-FG-209-03: Exploring the Maximum Count Rate Capabilities of Photon Counting Arrays Based On Polycrystalline Silicon

Purpose: Photon counting arrays (PCAs) offer several advantages over conventional, fluence-integrating x-ray imagers, such as improved contrast by means of energy windowing. For that reason, we are exploring the feasibility and performance of PCA pixel circuitry based on polycrystalline silicon. This material, unlike the crystalline silicon commonly used in photon counting detectors, lends itself toward the economic manufacture of radiation tolerant, monolithic large area (e.g., ∼43×43 cm2) devices. In this presentation, exploration of maximum count rate, a critical performance parameter for such devices, is reported. Methods: Count rate performance for a variety of pixel circuit designs was explored through detailed circuit simulations over a wide range of parameters (including pixel pitch and operating conditions) with the additional goal of preserving good energy resolution. The count rate simulations assume input events corresponding to a 72 kVp x-ray spectrum with 20 mm Al filtration interacting with a CZT detector at various input flux rates. Output count rates are determined at various photon energy threshold levels, and the percentage of counts lost (e.g., due to deadtime or pile-up) is calculated from the ratio of output to input counts. The energy resolution simulations involve thermal and flicker noise originating from each circuit element in a design. Results: Circuit designs compatible with pixel pitches ranging from 250 to 1000 µm that allow count rates over a megacount per second per pixel appear feasible. Such rates are expected to be suitable for radiographic and fluoroscopic imaging. Results for the analog front-end circuitry of the pixels show that acceptable energy resolution can also be achieved. Conclusion: PCAs created using polycrystalline silicon have the potential to offer monolithic large-area detectors with count rate performance comparable to those of crystalline silicon detectors. Further improvement through detailed circuit simulations and prototyping is expected. Partially supported by NIH grant R01-EB000558. This work was partially supported by NIH grant no. R01-EB000558.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

WE-AB-209-05: Development of an Ultra-Fast High Quality Whole Breast Radiotherapy Treatment Planning System

Purpose: To enable near-real-time (<20sec) and interactive planning without compromising quality for whole breast RT treatment planning using tangential fields. Methods: Whole breast RT plans from 20 patients treated with single energy (SE, 6MV, 10 patients) or mixed energy (ME, 6/15MV, 10 patients) were randomly selected for model training. Additional 20 cases were used as validation cohort. The planning process for a new case consists of three fully automated steps:1. Energy Selection. A classification model automatically selects energy level. To build the energy selection model, principle component analysis (PCA) was applied to the digital reconstructed radiographs (DRRs) of training cases to extract anatomy-energy relationship.2. Fluence Estimation. Once energy is selected, a random forest (RF) model generates the initial fluence. This model summarizes the relationship between patient anatomy’s shape based features and the output fluence. 3. Fluence Fine-tuning. This step balances the overall dose contribution throughout the whole breast tissue by automatically selecting reference points and applying centrality correction. Fine-tuning works at beamlet-level until the dose distribution meets clinical objectives. Prior to finalization, physicians can also make patient-specific trade-offs between target coverage and high-dose volumes.The proposed method was validated by comparing auto-plans with manually generated clinical-plans using Wilcoxon Signed-Rank test. Results: In 19/20 cases the model suggested the same energy combination as clinical-plans. The target volume coverage V100% was 78.1±4.7% for auto-plans, and 79.3±4.8% for clinical-plans (p=0.12). Volumes receiving 105% Rx were 69.2±78.0cc for auto-plans compared to 83.9±87.2cc for clinical-plans (p=0.13). The mean V10Gy, V20Gy of the ipsilateral lung was 24.4±6.7%, 18.6±6.0% for auto plans and 24.6±6.7%, 18.9±6.1% for clinical-plans (p=0.04, <0.001). Total computational time for auto-plans was < 20s. Conclusion: We developed an automated method that generates breast radiotherapy plans with accurate energy selection, similar target volume coverage, reduced hotspot volumes, and significant reduction in planning time, allowing for near-real-time planning.

60 APPLIED LIFE SCIENCES↗