Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Linear Optimization”

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 1,117 records · Page 62

Adiabatic quantum linear regression

Abstract A major challenge in machine learning is the computational expense of training these models. Model training can be viewed as a form of optimization used to fit a machine learning model to a set of data, which can take up significant amount of time on classical computers. Adiabatic quantum computers have been shown to excel at solving optimization problems, and therefore, we believe, present a promising alternative to improve machine learning training times. In this paper, we present an adiabatic quantum computing approach for training a linear regression model. In order to do this, we formulate the regression problem as a quadratic unconstrained binary optimization (QUBO) problem. We analyze our quantum approach theoretically, test it on the D-Wave adiabatic quantum computer and compare its performance to a classical approach that uses the Scikit-learn library in Python. Our analysis shows that the quantum approach attains up to $${2.8 \times }$$ 2.8 × speedup over the classical approach on larger datasets, and performs at par with the classical approach on the regression error metric. The quantum approach used the D-Wave 2000Q adiabatic quantum computer, whereas the classical approach used a desktop workstation with an 8-core Intel i9 processor. As such, the results obtained in this work must be interpreted within the context of the specific hardware and software implementations of these machines.

97 MATHEMATICS AND COMPUTING↗

SU-F-J-173: Online Replanning for Dose Painting Based On Changing ADC Map of Pancreas Cancer

Purpose: The introduction of MR-guided radiation therapy (RT), e.g., MR-Linac, would allow dose painting to adapt spatial RT response revealed from MRI data during the RT delivery. The purpose of this study is to investigate the use of an online replanning method to adapt dose painting from the MRI Apparent Diffusion Coefficient (ADC) map acquired during the delivery of RT for pancreatic cancers. Methods: Original dose painting plans were created based on multi-parametric simulation MRI including T1, T2 and ADC, using a treatment planning system (MONACO, Elekta) equipped with an online replanning algorithm (WSO, warm start optimization). Multiple GTVs, identified based on various ADC levels were prescribed to different doses ranging from 50–70 Gy with simultaneous integrated boost in 28 fractions. The MRI acquired after RT were used to mimic weekly MRI, on which the changing GTVs, pancreatic head and other organs-at-risk (OAR) (duodenum, stomach, small bowel) were delineated. The adaptive plan was generated by applying WSO algorithm starting from the deformed original plan based on the weekly MRI using a deformable image registration (DIR) software (ADMIRE, Elekta). The online replanning method takes <10 min. including DIR, target delineation, WSO execution and final dose calculation. Standard IGRT repositioning and full-blown reoptimization plans were also generated to compare with the adaptive plans. Results: The online replanning method significantly improved the multiple target coverages and OAR sparing for pancreatic cancers. For example, for a case with two GTVs with prescriptions of 60 and 70 Gy in pancreatic head, V100-GTV70 (the volume covered by 100% of prescription dose for GTV with 70 Gy)/V100-GTV60/V100-CTV50/V45-duodenum were (95.1/22.2/69.5/85.7), (95.0/97.0/98.6/34.3), and (95.0/98.1/100.0/38.7) for the IGRT, adaptive and reoptimization plans, respectively. Conclusion: The introduced online adaptive replanning method can effectively account for interfractional changes including tumor spatial response during MR-guided RT delivery, allowing precise delivery of dose painting. This study was partially supported by Elekta Inc.

60 APPLIED LIFE SCIENCES↗

SU-E-T-53: Benchmarking a Monte Carlo Model for Patient Plane Leakage Calculations of Low Energy 6MV Unique Linacs

Purpose: To validate the physics parameters of a Monte Carlo model for patient plane leakage calculations on the 6MV Unique linac by comparing the simulations against IEC patient plane leakage measurements. The benchmarked model can further be used for shielding design optimization, to predict leakage in the proximity of intended treatment fields, reduce the system weight and cost, and improve components reliability. Methods: The treatment head geometry of the Unique linac was simulated in Geant4 (v9.4.p02 with “Opt3” standard electromagnetic physics list) based on CAD drawings of all collimation and shielding components projected from the target to the area within 2m from isocenter. A 4×4m2 scorer was inserted 1m from the target in the patient plane and multiple phase space files were recorded by performing a 40-node computing cluster simulation on the EC2 cloud. The photon energy fluence was calculated relative to the value at isocenter for a 10×10cm2 field using 10×10mm2 bins. Tungsten blocks were parked accordingly to represent MLC120. The secondary particle contamination to patient plane was eliminated by “killing” those particles prior to the primary collimator entrance using a “kill-plane”, which represented the upper head shielding components not being modeled. Both IEC patient-plane leakage and X/Y-jaws transmission were simulated. Results: The contribution of photons to energy fluence was 0.064% on average, in excellent agreement with the experimental data available at 0.5, 1.0, and 1.5m from isocenter, characterized by an average leakage of 0.045% and a maximum leakage of 0.085%. X- and Y-jaws transmissions of 0.43% and 0.44% were found in good agreement with measurements of 0.48% and 0.43%, respectively. Conclusion: A Geant4 model based on energy fluence calculations for the 6MV Unique linac was created and validated using IEC patient plane leakage measurements. The “kill-plane” has effectively eliminated electron contamination to patient plane in these simulations.

60 APPLIED LIFE SCIENCES↗

Determination of output factors for stereotactic radiosurgery beams

Accurate dosimetry of the narrow beam tends to be difficult to perform due to the absence of lateral electronic equilibrium and the steep dose gradient, as well as the finite size of detectors. Thus, although the high dose rate 6 MV beam on the VARIAN Trilogy accelerator is increasingly utilized for stereotactic radiosurgery (SRS) treatment, there is no general agreement in the SRS beam output factor values among the Trilogy user community. Trilogy SRS beams are confined by cone collimators and the available collimator sizes range from 5 and 10 to 30 mm, in every 2 mm increment. A range of the relative output factors are in clinic use. This variation may impair observations of dose response and optimizations of the prescribed dose. It is necessary to investigate an accurate, easily performable, and detector independent method for the narrow beam output factor measurement. In this study, a scanning beam/scanning chamber method was proposed to overcome the limitation/difficulty of using a relatively large detector in narrow beam output factor measurement. Specifically, for the scanning beam method, multiple narrow beams are used for the dose measurement using a finite size chamber. These multiple scanning beams form an equivalent large uniform field which provides lateral electron equilibrium condition. After the measurement, the contributions from neighboring beams are deconvolved and the value is used for output factor determinations. For a Linac that cannot move a beam laterally, the scanning chamber method can be used to achieve the same result. The output factors determined in such a method were compared to chambers (a 0.015 cc PTW PinPoint ion chamber and a 0.125 cc PTW ion chamber) and film measurement, as well as with Monte Carlo simulation. Film and Monte Carlo results are found to be in excellent agreement with the measurement using the scan beam method. However, the VARIAN recommended output factors measured directly by Wellhoefer CC01 chamber and Scanditronix photon field diode are consistently higher for all the cones. Especially for the 5 mm cone, the difference is more than 10%. Overall, the results suggested that the new method can help overcoming the detector volume averaging effect and the positioning uncertainties, which constitute the major challenge in small radiosurgical beam output factor measurement, and provide reliable output factors.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Efficiency in nonequilibrium molecular dynamics Monte Carlo simulations

Hybrid algorithms combining nonequilibrium molecular dynamics and Monte Carlo (neMD/MC) offer a powerful avenue for improving the sampling efficiency of computer simulations of complex systems. These neMD/MC algorithms are also increasingly finding use in applications where conventional approaches are impractical, such as constant-pH simulations with explicit solvent. However, selecting an optimal nonequilibrium protocol for maximum efficiency often represents a non-trivial challenge. This work evaluates the efficiency of a broad class of neMD/MC algorithms and protocols within the theoretical framework of linear response theory. The approximations are validated against constant pH-MD simulations and shown to provide accurate predictions of neMD/MC performance. An assessment of a large set of protocols confirms (both theoretically and empirically) that a linear work protocol gives the best neMD/MC performance. Lastly, a well-defined criterion for optimizing the time parameters of the protocol is proposed and demonstrated with an adaptive algorithm that improves the performance on-the-fly with minimal cost.

97 MATHEMATICS AND COMPUTING↗

Network-Level Optimization for Unbalanced Power Distribution System: Approximation and Relaxation

The nonlinear programming (NLP) problem to solve distribution-level optimal power flow (D-OPF) poses convergence issues and does not scale well for unbalanced distribution systems. The existing scalable D-OPF algorithms either use approximations that are not valid for an unbalanced power distribution system, or apply relaxation techniques to the nonlinear power flow equations that do not guarantee a feasible power flow solution. In this paper, we propose scalable D-OPF algorithms that simultaneously achieve optimal and feasible solutions by solving multiple iterations of approximate, or relaxed, D-OPF subproblems of low complexity. The first algorithm is based on a successive linear approximation of the nonlinear power flow equations around the current operating point, where the D-OPF solution is obtained by solving multiple iterations of a linear programming (LP) problem. The second algorithm is based on the relaxation of the nonlinear power flow equations as conic constraints together with directional constraints, which achieves optimal and feasible solutions over multiple iterations of a second-order cone programming (SOCP) problem. Finally, it is demonstrated that the proposed algorithms are able to reach an optimal and feasible solution while significantly reducing the computation time as compared to an equivalent NLPD-OPF model for the same distribution system.

24 POWER TRANSMISSION AND DISTRIBUTION↗

A trust region algorithm for PDE-constrained optimization with bound constraints using reduced order modeling

This paper presents a novel trust region algorithm that relies on proper orthogonal de-composition techniques to construct accurate reduced order models during optimization.The algorithm samples high fidelity snapshots to compute the POD functions that are used to generate reduced order models. The reduced order models are used to replace the computationally intensive high fidelity finite element evaluations during optimization.The proposed algorithm employs a trust region framework to detect loss in predictive ac-curacy in the reduced order model and automatically update the POD functions during optimization. The trust region framework allows the algorithm to use sound mathematical metrics to effectively improve the accuracy and robustness of the reduced order models during optimization. The algorithm also employs a projected gradient algorithm to model bound constraints and compute optimal and feasible controls.This paper also presents an accurate Hessian formulation for topology optimization problems. The proposed trust region framework relies on a quadratic model to update the control. This quadratic model needs reliable second order derivative information to predict the behavior of the objective function within a suitable trust region. If a nonlinear Hessian formulation is used, the computational effort increases due to additional finite element evaluations. The proposed linear Hessian formulation reduces the computational effort and enables the calculation of the second order derivative information without additional finite element model evaluations. Examples in topology optimization are presented to demonstrate the applicability of the proposed algorithm and linear Hessian formulation for large-scale PDE-constrained optimization.

bound constraints↗

Optimizing colormaps with consideration for color vision deficiency to enable accurate interpretation of scientific data

Color vision deficiency (CVD) affects more than 4% of the population and leads to a different visual perception of colors. Though this has been known for decades, colormaps with many colors across the visual spectra are often used to represent data, leading to the potential for misinterpretation or difficulty with interpretation by someone with this deficiency. Until the creation of the module presented here, there were no colormaps mathematically optimized for CVD using modern color appearance models. While there have been some attempts to make aesthetically pleasing or subjectively tolerable colormaps for those with CVD, our goal was to make optimized colormaps for the most accurate perception of scientific data by as many viewers as possible. We developed a Python module, cmaputil , to create CVD-optimized colormaps, which imports colormaps and modifies them to be perceptually uniform in CVD-safe colorspace while linearizing and maximizing the brightness range. The module is made available to the science community to enable others to easily create their own CVD-optimized colormaps. In this paper, we present an example CVD-optimized colormap created with this module that is optimized for viewing by those without a CVD as well as those with red-green colorblindness. This colormap, cividis, enables nearly-identical visual-data interpretation to both groups, is perceptually uniform in hue and brightness, and increases in brightness linearly.

59 BASIC BIOLOGICAL SCIENCES↗

Learning Optimal Multigrid Smoothers via Neural Networks

Multigrid methods are one of the most efficient techniques for solving large sparse linear systems arising from partial differential equations (PDEs) and graph Laplacians from machine learning applications. One of the key components of multigrid is smoothing, which aims at reducing high-frequency errors on each grid level. However, finding optimal smoothing algorithms is problem-dependent and can impose challenges for many problems. In this paper, we propose an efficient adaptive framework for learning optimized smoothers from operator stencils in the form of convolutional neural networks (CNNs). Here, the CNNs are trained on small-scale problems from a given type of PDEs based on a supervised loss function derived from multigrid convergence theories and can be applied to large-scale problems of the same class of PDEs. Numerical results on anisotropic rotated Laplacian problems and variable coefficient diffusion problems demonstrate improved convergence rates and solution time compared with classical hand-crafted relaxation methods.

97 MATHEMATICS AND COMPUTING↗

Development and optimization of a beam shaper device for a mobile dedicated IOERT accelerator

Purpose: The aim of this study was to design and build a prototype beam shaper to be used on a dedicated mobile accelerator that protects organs at risk within the radiation field and conforms the beam to the target geometry during intraoperative electron radiotherapy (IOERT). A dosimetric characterization of the beam shaper device was performed based on Monte Carlo (MC) simulations, as well as experimental data, at different energies, field sizes, and source to skin distances. Methods: A mobile light intraoperative accelerator (LIAC{sup Registered-Sign }, Sordina, Italy) was used. The design of the beam shaper prototype was based on MC simulations (BEAMnrc/OMEGA and DOSXYZnrc code) for a selection of materials and thicknesses, as well as for dosimetric characterization. Percentage depth dose (PDD) and profile measurements were performed using a p-type silicon diode and a commercial water phantom, while output factors were measured using a PinPoint ion chamber in a PMMA phantom. Planar doses in planes of interest were carried out using radiochromic films (Gafchromic{sup TM} EBT and EBT2) in PMMA and in a Solid Water{sup Registered-Sign} phantom. Several experimental set-ups were investigated with the beam shaper device fixed on the top of the phantom, varying both the short side of the rectangular field and the air gap between the device and the phantom surface, simulating the clinical situation. The output factors (OFs) were determined using different geometrical set-ups and energies. Results: The beam shaper prototype consists of four blades sliding alongside each other and mounted on a special support at the end of the 10 cm diameter PMMA circular applicator. Each blade is made of an upper layer of 2.6 cm of Teflon{sup Registered-Sign} and a lower layer of 8 mm of stainless steel. All rectangles inscribed in a 5 cm diameter can be achieved in addition to any 'squircle-shaped' field. When one side of the rectangular field is held constant and the second side is reduced, both R{sub 50} and R{sub max} move towards the phantom surface. Comparing the PDDs obtained with the 5 cm circular applicator and with a 4.4 Multiplication-Sign 4.4 cm{sup 2} square field (that is the equivalent square of the 5 cm circular field) obtained with the beam shaper, a different behavior was observed in the region extending from the surface to a depth of 50% of the maximum dose. Isodoses measured for rectangular fields used for clinical cases (i.e., 4 Multiplication-Sign 9 cm{sup 2} 8 MeV) are shown, with different air gaps. For each energy investigated, the normalized OFs slowly increase, when the length of the side decreases down to about 4 cm, and then rapidly decreases for smaller field widths. MC simulation showed an excellent agreement with experimental data (<2%). Conclusions: The beam shaper device is able to provide square/rectangular/squircle fields with adequate dose homogeneity for mobile dedicated accelerators, thus allowing conformal treatment with IOERT. Monte Carlo simulation can be a very useful tool to simulate any clinical set up and can be used to create a data set to calculate MUs, thereby increasing the accuracy of the delivered dose during IOERT procedures.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Instrumental noise estimates stabilize and quantify endothelial cell micro-impedance barrier function parameter estimates

A novel transcellular micro-impedance biosensor, referred to as the electric cell-substrate impedance sensor or ECIS, has become increasingly applied to the study and quantification of endothelial cell physiology. In principle, frequency dependent impedance measurements obtained from this sensor can be used to estimate the cell cell and cell matrix impedance components of endothelial cell barrier function based on simple geometric models. Few studies, however, have examined the numerical optimization of these barrier function parameters and established their error bounds. This study, therefore, illustrates the implementation of a multi-response Levenberg Marquardt algorithm that includes instrumental noise estimates and applies it to frequency dependent porcine pulmonary artery endothelial cell impedance measurements. The stability of cell cell, cell matrix and membrane impedance parameter estimates based on this approach is carefully examined, and several forms of parameter instability and refinement illustrated. Including frequency dependent noise variance estimates in the numerical optimization reduced the parameter value dependence on the frequency range of measured impedances. The increased stability provided by a multi-response non-linear fit over one-dimensional algorithms indicated that both real and imaginary data should be used in the parameter optimization. Error estimates based on single fits and Monte Carlo simulations showed that the model barrier parameters were often highly correlated with each other. Independently resolving the different parameters can, therefore, present a challenge to the experimentalist and demand the use of non-linear multivariate statistical methods when comparing different sets of parameters.

59 BASIC BIOLOGICAL SCIENCES↗

Reduced order modeling for flow and transport problems with Barlow Twins self-supervised learning

Abstract We propose a unified data-driven reduced order model (ROM) that bridges the performance gap between linear and nonlinear manifold approaches. Deep learning ROM (DL-ROM) using deep-convolutional autoencoders (DC–AE) has been shown to capture nonlinear solution manifolds but fails to perform adequately when linear subspace approaches such as proper orthogonal decomposition (POD) would be optimal. Besides, most DL-ROM models rely on convolutional layers, which might limit its application to only a structured mesh. The proposed framework in this study relies on the combination of an autoencoder (AE) and Barlow Twins (BT) self-supervised learning, where BT maximizes the information content of the embedding with the latent space through a joint embedding architecture. Through a series of benchmark problems of natural convection in porous media, BT–AE performs better than the previous DL-ROM framework by providing comparable results to POD-based approaches for problems where the solution lies within a linear subspace as well as DL-ROM autoencoder-based techniques where the solution lies on a nonlinear manifold; consequently, bridges the gap between linear and nonlinear reduced manifolds. We illustrate that a proficient construction of the latent space is key to achieving these results, enabling us to map these latent spaces using regression models. The proposed framework achieves a relative error of 2% on average and 12% in the worst-case scenario (i.e., the training data is small, but the parameter space is large.). We also show that our framework provides a speed-up of $$7 \times 10^{6}$$ 7 × 10 6 times, in the best case, and $$7 \times 10^{3}$$ 7 × 10 3 times on average compared to a finite element solver. Furthermore, this BT–AE framework can operate on unstructured meshes, which provides flexibility in its application to standard numerical solvers, on-site measurements, experimental data, or a combination of these sources.

97 MATHEMATICS AND COMPUTING↗

ELSI — An open infrastructure for electronic structure solvers

Routine applications of electronic structure theory to molecules and periodic systems need to compute the electron density from given Hamiltonian and, in case of non-orthogonal basis sets, overlap matrices. System sizes can range from few to thousands or, in some examples, millions of atoms. Different discretization schemes (basis sets) and different system geometries (finite non-periodic vs. infinite periodic boundary conditions) yield matrices with different structures. The ELectronic Structure Infrastructure (ELSI) project provides an open-source software interface to facilitate the implementation and optimal use of high-performance solver libraries covering cubic scaling eigensolvers, linear scaling density-matrix-based algorithms, and other reduced scaling methods in between. In this paper, we present recent improvements and developments inside ELSI, mainly covering (1) new solvers connected to the interface, (2) matrix layout and communication adapted for parallel calculations of periodic and/or spin-polarized systems, (3) routines for density matrix extrapolation in geometry optimization and molecular dynamics calculations, and (4) general utilities such as parallel matrix I/O and JSON output. The ELSI interface has been integrated into four electronic structure code projects (DFTB+, DGDFT, FHI-aims, SIESTA), allowing us to rigorously benchmark the performance of the solvers on an equal footing. Based on results of a systematic set of large-scale benchmarks performed with Kohn–Sham density-functional theory and density-functional tight-binding theory, we identify factors that strongly affect the efficiency of the solvers, and propose a decision layer that assists with the solver selection process. As a result, we describe a reverse communication interface encoding matrix-free iterative solver strategies that are amenable, e.g., for use with planewave basis sets.

97 MATHEMATICS AND COMPUTING↗

Use of a segmented cell for the combinatorial development of platinum group metal-free electrodes for polymer electrolyte fuel cells

The development of novel platinum group metal (PGM)-free catalysts is a difficult task, particularly when coupled with the integration into a cathode catalyst layer (CCL). The optimization of such PGM-free electrode structures is often non-linear and iterative, making it a demanding task due to the high number of parameters that affect performance. To accelerate both materials discovery and electrode development, this work demonstrates the application of a high-resolution segmented cell (SC) with 121 segments of 0.413 cm 2 size for combinatorial high-throughput PGM-free catalyst screening and CCL optimization. The approach utilizes three flow-field strips with active areas of 4.45 cm 2 each, distributed over 11 segments. Electrodes with identical catalyst material and fabrication method result in reproducible data with typically less than 10% variation between each segment and match in performance with differential single cell data. High throughput testing of combinatorial sample sets, with varied PGM-free electrocatalyst materials or electrode composition, demonstrated the ability to rapidly discern high performing outliers, and establish or confirm trends in electrode optimization. The results indicated significant performance benefits of electrodes with ionomer content of 45 wt% over lower values, and of water rich inks with 82% water content over those with 50%.

30 DIRECT ENERGY CONVERSION↗

Sensitivity of joint contagiousness and susceptibility-based dynamic optimal control strategies for HIV prevention

Predicting the population-level effects of an infectious disease intervention that incorporate multiple modes of intervention is complicated by the joint non-linear dynamics of both infection transmission and the intervention itself. In this paper, we consider the sensitivity of Dynamic Optimal Control Profiles (DOCPs) for the optimal joint investment in both a contagiousness and susceptibility-based control of HIV to bio-behavioral, economic, and programmatic assumptions. The DOCP is calculated using recently developed numerical algorithms that allow controls to be represented by a set of piecewise constant functions that maintain a constant yearly budget. Our transmission model assumes multiple stages of HIV infection corresponding to acute and chronic infection and both within- and between-individual behavioral heterogeneity. We parameterize a baseline scenario from a longitudinal study of sexual behavior in MSM and consider sensitivity of the DOCPs to deviations from that baseline scenario. In the baseline scenario, the primary determinant of the dominant control were programmatic factors, regardless of budget. In sensitivity analyses, the qualitative aspects of the optimal control policy were often robust to significant deviation in assumptions regarding transmission dynamics. In addition, we found several conditions in which long-term joint investment in both interventions was optimal. Our results suggest that modeling in the service of decision support for intervention design can improve population-level effects of a limited set of economic resources. We found that economic and programmatic factors were as important as the inherent transmission dynamics in determining population-level intervention effects. Given our finding that the DOCPs were robust to alternative biological and behavioral assumptions it may be possible to identify DOCPs even when the data are not sufficient to identify a transmission model.

59 BASIC BIOLOGICAL SCIENCES↗

TU-D-207B-02: Delta-Radiomics: The Prognostic Value of Therapy-Induced Changes in Radiomics Features for Stage III Non-Small Cell Lung Cancer Patients

Purpose: To determine how radiomics features change during radiation therapy and whether those changes (delta-radiomics features) can improve prognostic models built with clinical factors. Methods: 62 radiomics features, including histogram, co-occurrence, run-length, gray-tone difference, and shape features, were calculated from pretreatment and weekly intra-treatment CTs for 107 stage III NSCLC patients (5–9 images per patient). Image preprocessing for each feature was determined using the set of pretreatment images: bit-depth resample and/or a smoothing filter were tested for their impact on volume-correlation and significance of each feature in univariate cox regression models to maximize their information content. Next, the optimized features were calculated from the intratreatment images and tested in linear mixed-effects models to determine which features changed significantly with dose-fraction. The slopes in these significant features were defined as delta-radiomics features. To test their prognostic potential multivariate cox regression models were fitted, first using only clinical features and then clinical+delta-radiomics features for overall-survival, local-recurrence, and distant-metastases. Leave-one-out cross validation was used for model-fitting and patient predictions. Concordance indices(c-index) and p-values for the log-rank test with patients stratified at the median were calculated. Results: Approximately one-half of the 62 optimized features required no preprocessing, one-fourth required smoothing, and one-fourth required smoothing and resampling. From these, 54 changed significantly during treatment. For overall-survival, the c-index improved from 0.52 for clinical factors alone to 0.62 for clinical+delta-radiomics features. For distant-metastases, the c-index improved from 0.53 to 0.58, while for local-recurrence it did not improve. Patient stratification significantly improved (p-value<0.05) for overallsurvival and distant-metastases when delta-radiomics features were included. The delta-radiomics versions of autocorrelation, kurtosis, and compactness were selected most frequently in leave-one-out iterations. Conclusion: Weekly changes in radiomics features can potentially be used to evaluate treatment response and predict patient outcomes. High-risk patients could be recommended for dose escalation or consolidation chemotherapy. This project was funded in part by grants from the National Cancer Institute (NCI) and the Cancer Prevention Research Institute of Texas (CPRIT).

60 APPLIED LIFE SCIENCES↗