Reduction In Computation Time Research Articles

Fast model calibration for predicting the response of breast cancer to chemotherapy using proper orthogonal decomposition

Constructing digital twins for predictive tumor treatment response models can have a high computational demand that presents a practical barrier for their clinical adoption. In this work, we demonstrate that proper orthogonal decomposition, by which a low-dimensional representation of the full model is constructed, can be used to dramatically reduce the computational time required to calibrate a partial differential equation model to magnetic resonance imaging (MRI) data for rapid predictions of tumor growth and response to chemotherapy. In the proposed formulation, the reduction basis is based on each patient’s own MRI data and controls the overall size of the “reduced order model”. Using the full model as the reference, we validate that the reduced order mathematical model can accurately predict response in 50 triple negative breast cancer patients receiving standard of care neoadjuvant chemotherapy. The concordance correlation coefficient between the full and reduced order models was 0.986 ± 0.012 (mean ± standard deviation) for predicting changes in both tumor volume and cellularity across the entire model family, with a corresponding median local error (inter-quartile range) of 4.36 % (1.22 %, 15.04 %). The total time to estimate parameters and to predict response dramatically improves with the reduced framework. Specifically, the reduced order model accelerates our calibration by a factor of (mean ± standard deviation) 378.4 ± 279.8 when compared to the full order model for a non-mechanically coupled model. This enormous reduction in computational time can directly help realize the practical construction of digital twins when the access to computational resources is limited.

Coupled axial and transverse currents method for finite element modelling of periodic superconductors

In this paper, we propose the Coupled Axial and Transverse currents (I) (CATI) method, as an efficient and accurate finite element approach for modelling the electric and magnetic behavior of periodic composite superconducting conductors. The method consists of a pair of two-dimensional models coupled via circuit equations to account for the conductor geometrical periodicity. This allows to capture three-dimensional effects with two-dimensional models and leads to a significant reduction in computational time compared to conventional three-dimensional models. After presenting the method in detail, we verify it by comparison with reference finite element models, focussing on its application to twisted multifilamentary superconducting strands. In particular, we show that the CATI method captures the transition from uncoupled to coupled filaments, with accurate calculation of the interfilament coupling time constant. We then illustrate the capabilities of the method by generating detailed loss maps and magnetization curves of given strand types for a range of external transverse magnetic field excitations, with and without transport current.

A combined EEG motor and speech imagery paradigm with automated successive halving for customizable command selection

ABSTRACT The classification performance of endogenous electroencephalogram (EEG) brain-computer interfaces (BCIs) can be improved by hybridizing the paradigm through the use of commands from multiple paradigms. Hybrid paradigms using motor imagery (MI) and speech imagery (SI) have shown promise, but there is a lack of research into: i) their effectiveness when compared to pure MI and SI for multiclass problems, and ii) automated command selection. This study investigates multiclass MI and SI hybrid paradigms and compares the results to those obtained using pure MI and SI. Performance was assessed using F1 score and accuracy. The performances of all possible hybrid paradigm designs were assessed. The analysis indicated that hybridization does not always guarantee improved performance when compared to the pure paradigms, and there is inter-subject variation in the best paradigm. This confirmed the need for automated subject-specific hybrid paradigm designs. An automated hybrid paradigm selection technique using successive halving (SH) for expedited computational times was developed and results were compared to those obtained using a standard grid search. The SH approach resulted in an improvement in F1 score of 21.09% and 36.86% compared to MI and SI and led to a reduction in computational times of 82.80% compared to grid search.

Deep learning-based voxel sampling for particle therapy treatment planning

Objective. Scanned particle therapy often requires complex treatment plans, robust optimization, as well as treatment adaptation. Plan optimization is especially complicated for heavy ions due to the variable relative biological effectiveness. We present a novel deep-learning model to select a subset of voxels in the planning process thus reducing the planning problem size for improved computational efficiency. Approach. Using only a subset of the voxels in target and organs at risk (OARs) we produced high-quality treatment plans, but heuristic selection strategies require manual input. We designed a deep-learning model based on P-Net to obtain an optimal voxel sampling without relying on patient-specific user input. A cohort of 70 head and neck patients that received carbon ion therapy was used for model training (50), validation (10) and testing (10). For training, a total of 12 500 carbon ion plans were optimized, using a highly efficient artificial intelligence (AI) infrastructure implemented into a research treatment planning platform. A custom loss function increased sampling density in underdosed regions, while aiming to reduce the total number of voxels. Main results. On the test dataset, the number of voxels in the optimization could be reduced by 84.8% (median) at <1% median loss in plan quality. When the model was trained to reduce sampling in the target only while keeping all voxels in OARs, a median reduction up to 71.6% was achieved, with 0.5% loss in the plan quality. The optimization time was reduced by a factor of 7.5 for the total AI selection model and a factor of 3.7 for the model with only target selection. Significance. The novel deep-learning voxel sampling technique achieves a significant reduction in computational time with a negligible loss in the plan quality. The reduction in optimization time can be especially useful for future real-time adaptation strategies.

Branch-Price-and-Cut algorithms for the team orienteering problem with interval-varying profits

This paper studies the team orienteering problem (TOP), wherein each vertex requires two visits, and the service profit of each vertex depends on the time interval between these two visits. Motivated by scenarios like perishable product delivery, home health care scheduling, and earth observation satellite scheduling, the paper addresses two variants: the periodic orienteering problem with a focus on the number of days (NoDs) and the team orienteering problem with attention to a combination of time windows (CoTWs). It develops mixed-integer linear programming (MILP) models for both variants and devises exact Branch-Price-and-Cut (BPC) algorithms tailored to their block diagonal structures. Furthermore, this paper proposes two strategies to improve algorithm performance. A simplification strategy streamlines the directed graph network by removing redundant vertices and arcs without compromising optimality, thereby accelerating the solution of pricing problems. Additionally, a matheuristic algorithm is proposed to obtain integer solutions quickly. Computational results demonstrate the effectiveness of these algorithms, showcasing their superiority over CPLEX. The BPC algorithm, coupled with the simplification strategy, exhibits an average computational time reduction of 70 % compared to the basic strategy. The effectiveness of the matheuristic algorithm is also confirmed. Finally, the findings presented herein offer valuable insights for refining and applying BPC algorithms.

Collision-free trajectory planning for UAVs based on sequential convex programming

This paper proposes an optimization-based approach to solve the collision-free trajectory planning problem for unmanned aerial vehicles (UAVs) with nonlinear dynamical systems. The problem is modeled as a non-convex optimization problem, and then iteratively solved by sequential convex programming algorithm which approximates non-convex constraints by convexification methods. We linearize the dynamical equations of the UAV and penalize collision-free constraints with a hinge loss. Furthermore, a variable trust region method is employed to ensure the convergence and stability of the algorithm. We apply the algorithm to three-degree-of-freedom UAV case studies. Numerical results validate that our method can provide collision-free trajectories with a 27.19% reduction in total flight time and a 93.61% reduction in computation time on average compared with the genetic algorithm method.

On Alpha-Expansion-Based Graph-Cut Optimization for Decoder-Side Depth Estimation

In order to achieve high realism an acceptable level of user experience in immersive videos, it is crucial to provide both the best possible quality of depth maps and minimize computational time. In this paper, we propose a new approach to the decoder-side depth estimation (DSDE) problem, which uses the hierarchical alpha-expansion algorithm with additional improvements for the estimation designed to be more robust to compressed video artifacts and limited computational resources. As shown by the experimental results, the proposal simultaneously results in reduction of computation time of the estimation process (by almost 40%) and an improvement of quality of estimated depth maps. The increased quality is demonstrated by more than 6% Bjøntegaard delta gain compared to the Moving Picture Experts Group (MPEG) immersive video state-of-the-art DSDE method.

Fast identification of transients: applying expectation maximization to neutrino data

We present a novel method for identifying transients suitable for both strong signal-dominated and background-dominated objects. By employing the unsupervised machine learning algorithm known as expectation maximization, we achieve computing time reductions of over 104 on a single CPU compared to conventional brute-force methods. Furthermore, this approach can be readily extended to analyze multiple flares. We illustrate the algorithm's application by fitting the IceCube neutrino flare of TXS 0506+056.

Automatic Route Design by Stepwise Subdivision of Virtual Walls —Reduces Route Length and Speeds Up Execution Time—

With the development of autonomous driving technology utilizing machine learning, AI, and sensors, research on autonomous driving control has become more active, and a large number of innovative studies are underway. In the near future, all autonomous vehicle fleets will be able to communicate with each other for sharing information and overall optimal traffic control will be achieved. One of the vehicle control systems that are based on the premise of such a fully automated society is the “signal-less intersection.” There is an intersection traffic control method that achieves safe and rational route selection by using virtual walls (VWs), which are virtual obstacles, but there are issues in terms of total route length and reduction of computation time. To address the issues, we propose a method that (1) prunes unneeded paths and (2) arranges VWs in a stepwise manner. The effectiveness of the proposed method was evaluated by simulation, and the results showed that the total route length and execution time were reduced.

High-Performance Krawtchouk Polynomials of High Order Based on Multithreading

Orthogonal polynomials and their moments serve as pivotal elements across various fields. Discrete Krawtchouk polynomials (DKraPs) are considered a versatile family of orthogonal polynomials and are widely used in different fields such as probability theory, signal processing, digital communications, and image processing. Various recurrence algorithms have been proposed so far to address the challenge of numerical instability for large values of orders and signal sizes. The computation of DKraP coefficients was typically computed using sequential algorithms, which are computationally extensive for large order values and polynomial sizes. To this end, this paper introduces a computationally efficient solution that utilizes the parallel processing capabilities of modern central processing units (CPUs), namely the availability of multiple cores and multithreading. The proposed multi-threaded implementations for computing DKraP coefficients divide the computations into multiple independent tasks, which are executed concurrently by different threads distributed among the independent cores. This multi-threaded approach has been evaluated across a range of DKraP sizes and various values of polynomial parameters. The results show that the proposed method achieves a significant reduction in computation time. In addition, the proposed method has the added benefit of applying to larger polynomial sizes and a wider range of Krawtchouk polynomial parameters. Furthermore, an accurate and appropriate selection scheme of the recurrence algorithm is introduced. The proposed approach introduced in this paper makes the DKraP coefficient computation an attractive solution for a variety of applications.

3D analytical magnetic model calculation for Halbach PM induction heater planar topology

In this work, a 3D analytical magnetic model based on hybridation of the sub-domain’s method and the image’s theory to compute magnetic field and translational motion eddy current in the conducting plate of the Halbach permanent magnet induction heater planar topology is developed. The main objective is to remedy the problem of the transverse edge effect, and hence improving the efficiency of analytical model and the accuracy of results. The developed model also allows fast and precise simulations of 3D magnetic phenomena, and presents an important reduction in computation time compared to 3D finite element simulations.

AN EFFICIENT ITERATIVE METHOD TO SIMULATE THE VARYING STATIC RECEPTANCE OF A THIN-WALLED WORKPIECE

This paper introduces an accelerated iterative Conjugate Gradient method (CG-method) for simulating the varying static receptance of a thin-walled workpiece. The proposed approach enhances computational efficiency by computing an optimal initial displacement guess for a given cutter location. This guess is based on the previously calculated displacement vector from the preceding cutter location, serving as the start iteration vector for the CG-method. The validation of efficiency gains through this method is demonstrated by a significant reduction in computation time for a numerical example. Moreover, to ensure the accuracy of the accelerated iterative CG-method, comparisons are made with the Cholesky method. The results confirm the precision of the proposed approach.

A simplified GNSS/LEO joint orbit determination method

Joint orbit determination (JOD) of Global Navigation Satellite Systems (GNSS) satellites and low Earth orbit (LEO) satellites has been substantiated as an efficacious approach to compensate for the ground station geometry. In the traditional JOD, the precise orbit determination (POD) of LEO satellite is mainly processed by the reduced-dynamic approach, however, this approach involves complex calculations and the credibility of the determined positions diminishes when LEO satellites undergo orbital maneuvers. Therefore, a simplified JOD method is designed that employs kinematic approach to determine the LEO satellites orbit. To verify the effectiveness of the proposed method, the orbit of GPS satellites and LEO satellites are jointly estimated utilizing the regional and global networks. 8 LEO satellites, including GRACE-C/D, SWARM-A/B/C, SENTINEL-3A/B, and JASON-3, are chosen for JOD. The comparative analysis between the proposed method and traditional method are achieved in terms of GPS orbit accuracy, LEO orbit accuracy, computation time and the JOD performance during LEO maneuvers. Under regional station scenario, the GPS orbit accuracy determined using the proposed method and the traditional method is 3.64 cm and 2.52 cm, respectively. In the case of global station scenario, the accuracies are 1.71 cm and 1.64 cm. Additionally, the traditional method yields superior enhancement and higher accuracy of the LEO orbits. However, it exhibits a noticeable increase in computation time compared to the proposed method and the performance of JOD declines significantly when LEO satellites undergo orbital maneuvers. Alternatively, although the accuracy of the LEO orbits using the proposed method is comparatively lower, it offers a substantial reduction in the overall network computation time compared to traditional method. Moreover, the proposed method based on LEO kinematic precise orbit determination (KPOD) is nearly unaffected by orbital maneuvers of LEO satellites, presenting unique advantages in practical data processing.

Extended finite element method with cell-based smoothing for modeling frictional contact crack-induced acoustic nonlinearity involving distorted mesh

This study proposes a 2D cell-based smoothed extended finite element method (CS-XFEM) for accurate and efficient simulation of nonlinear ultrasonic wave propagation in solid structures, specifically addressing the effects of frictional contact in cracks. Traditional mesh discretization methods for cracks often suffer from mesh distortion and computational instability owing to their high aspect ratios. To overcome this, CS-XFEM integrates a cell-based smoothing technique into XFEM to model the frictional contact of a crack. A comprehensive numerical example demonstrates the advantages of CS-XFEM. The results show that CS-XFEM exhibits a higher convergence rate and enables a larger critical time increment than XFEM. Specifically, the critical time increment of CS-XFEM was found to be twice that of XFEM, leading to a 50% reduction in the total computational time. These findings confirm that CS-XFEM is an efficient, accurate, and robust numerical method for studying the acoustic nonlinearity induced by crack-induced frictional contact.

Sample-Adaptive Classification Inference Network

Existing pre-trained models have yielded promising results in terms of computational time reduction. However, these models only focus on pruning simple sentences or less salient words, while neglecting the treatment of relatively complex sentences. It is frequently these sentences that cause the loss of model accuracy. This shows that the adaptation of the existing models is one-sided. To address this issue, in this paper, we propose a sample-adaptive training and inference model. Specifically, complex samples are extracted from the training datasets and a dedicated data augmentation module is trained to extract global and local semantic information of complex samples. During inference, simple samples can exit the model via the Sample Adaptive Exit Mechanism, Normal samples pass through the whole backbone model before inference, while complex samples are processed by the Characteristic Enhancement Module after passing through the backbone model. In this way, all samples are processed adaptively. Our extensive experiments on classification tasks datasets in the field of Natural Language Processing demonstrate that our method enhances model accuracy and reduces model inference time for multiple datasets. Moreover, our method is transferable and can be applied to multiple pre-trained models.

Efficient matrix profile computation with Euclidean distance using Eigen transformation: Performance evaluation based on beat-to-beat interval (BBI) data.

The matrix profile serves as a fundamental tool to provide insights into similar patterns within time series. Existing matrix profile algorithms have been primarily developed for the normalized Euclidean distance, which may not be a proper distance measure in many settings. The methodology work of this paper was motivated by statistical analysis of beat-to-beat interval (BBI) data collected from smartwatches to monitor e-cigarette users' heart rate change patterns for which the original Euclidean distance ( -norm) would be a more suitable choice. Yet, incorporating the Euclidean distance into existing matrix profile algorithms turned out to be computationally challenging, especially when the time series is long with extended query sequences. We propose a novel methodology to efficiently compute matrix profile for long time series data based on the Euclidean distance. This methodology involves four key steps including (1) projection of the time series onto eigenspace; (2) enhancing singular value decomposition (SVD) computation; (3) early abandon strategy; and (4) determining lower bounds based on the first left singular vector. Simulation studies based on BBI data from the motivating example have demonstrated remarkable reductions in computational time, ranging from one-fourth to one-twentieth of the time required by the conventional method. Unlike the conventional method of which the performance deteriorates sharply as the time series length or the query sequence length increases, the proposed method consistently performs well across a wide range of the time series length or the query sequence length.

Efficient Parallel FDTD Method Based on Non-Uniform Conformal Mesh

The finite-difference time-domain (FDTD) method is a versatile electromagnetic simulation technique, widely used for solving various broadband problems. However, when dealing with complex structures and large dimensions, especially when applying perfectly matched layer (PML) absorbing boundaries, tremendous computational burdens will occur. To reduce the computational time and memory, this paper presents a Message Passing Interface (MPI) parallel scheme based on non-uniform conformal FDTD, which is suitable for convolutional perfectly matched layer (CPML) absorbing boundaries, and adopts a domain decomposition approach, dividing the entire computational domain into several subdomains. More importantly, only one magnetic field exchange is required during the iterations, and the electric field update is divided into internal and external parts, facilitating the synchronous communication of magnetic fields between adjacent subdomains and internal electric field updates. Finally, unmanned helicopters, helical antennas, 100-period folded waveguides, and 16 × 16 phased array antennas are designed to verify the accuracy and efficiency of the algorithm. Moreover, we conducted parallel tests on a supercomputing platform, showing its satisfactory reduction in computational time and excellent parallel efficiency.

An Incremental Interpolation Scheme With Discrete Cosine Series Expansion for Multimaterial Topology Optimization

Abstract Topology optimization is a powerful tool for structural design, while its computational cost is quite high due to the large number of design variables, especially for multilateral systems. Herein, an incremental interpolation approach with discrete cosine series expansion (DCSE) is established for multilateral topology optimization. A step function with shape coefficients (i.e., ensuring that no extra variables are required as the number of materials increases) and the use of the DCSE together reduces the number of variables (e.g., from 8400 to 120 for the optimization of the clamped–clamped beam with four materials). Remarkably, the proposed approach can effectively bypass the checkerboard problem without using any filter. The enhanced computational efficiency (e.g., a ∼89.2% reduction in computation time from 439.1 s to 47.4 s) of the proposed approach is validated via both 2D and 3D numerical cases.

A computational efficient approach for distributionally robust unit commitment with enhanced disjointed layered ambiguity set

AbstractTo achieve the sustainable development of the society, renewable energy dominated power systems are gradually formed. However, the uncertainty of renewable power poses challenges for power system operations, such as balancing the load and generation in day‐ahead unit commitment problems. To address this issue, an enhanced disjointed layered ambiguity set is proposed in this paper to effectively capture the uncertainty of renewable generation forecast errors. On this basis, a two‐stage distributionally robust unit commitment optimization problem is formulated, which is computational challenge. To this end, a novel reformation approach and a worst‐case support set identification method are proposed to derive the optimal solution in a computationally efficient way. Simulation results demonstrate that the proposed ambiguity set brings lower operation costs of unit commitment while maintaining similar reliability compared to existing methods. In detail, the proposed ambiguity set achieves a reduction in operating costs of approximately 10.01%. The proposed novel reformation approach and solution method are also shown to have higher computation efficiency than the existing Benders decomposition method, resulting in a reduction in computation time to less than 3.78% of the original duration.

In Search of the Best Low-Cost Methods for Efficient Screening of Conformers.

Locating the lowest energy conformer is crucial for the accurate computation of equilibrium properties of molecular systems. This paper examines the performance of efficient low-cost methods in terms of the alignment and relative energies of their energy minima against the benchmark revDSD-PBEP86-D4/def2-TZVPP//MP2/cc-pVTZ potential energy surface. The low-cost methods considered include GFN-FF, GFN2-xTB, DFTB3, HF-3c, B97-3c, PBEh-3c, and r2SCAN-3c composite methods against a diverse test set of 20 compounds including alkanes, perfluoroalkyl molecules, peptides, open-shell radicals, and Zn(II) complexes of varying sizes. The "3c" composite methods are generally more accurate, but are at least 2-3 orders of magnitude more expensive than tight-binding methods which have energy minima that align well with the benchmark potential energy surface. The findings of this paper were further exploited to introduce a simple strategy involving Grimme's CENSO energy-sorting algorithm that resulted in up to an order of magnitude reduction in computational time for locating the lowest energy conformer on the revDSD-PBEP86-D4/def2-TZVPP//MP2/cc-pVTZ surface.