Low Computational Cost Research Articles

Mixture-of-experts and semantic-guided network for brain tumor segmentation with missing MRI modalities.

Accurate brain tumor segmentation with multi-modal MRI images is crucial, but missing modalities in clinical practice often reduce accuracy. The aim of this study is to propose a mixture-of-experts and semantic-guided network to tackle the issue of missing modalities in brain tumor segmentation. We introduce a transformer-based encoder with novel mixture-of-experts blocks. In each block, four modality experts aim for modality-specific feature learning. Learnable modality embeddings are employed to alleviate the negative effect of missing modalities. We also introduce a decoder guided by semantic information, designed to pay higher attention to various tumor regions. Finally, we conduct extensive comparison experiments with other models as well as ablation experiments to validate the performance of the proposed model on the BraTS2018 dataset. The proposed model can accurately segment brain tumor sub-regions even with missing modalities. It achieves an average Dice score of 0.81 for the whole tumor, 0.66 for the tumor core, and 0.52 for the enhanced tumor across the 15 modality combinations, achieving top or near-top results in most cases, while also exhibiting a lower computational cost. Our mixture-of-experts and sematic-guided network achieves accurate and reliable brain tumor segmentation results with missing modalities, indicating its significant potential for clinical applications. Our source code is already available at https://github.com/MaggieLSY/MESG-Net .

On-tree fruit image segmentation comparing Mask R-CNN and Vision Transformer models. Application in a novel algorithm for pixel-based fruit size estimation

In situ automatic fruit monitoring is of great interest for more accurate and cost-efficient decision making in agriculture. For this purpose, the development of computer vision-based tools is essential. Deep Learning techniques have shown good performance in fruit detection and segmentation. Recently, new models based on Transformer architecture have emerged with promising potential and zero-shot inference capability. In this paper, a Deep Learning model, Mask R-CNN, was trained for on-tree pomegranate fruit segmentation and compared with foundational models based on Vision Transformer, Grounding DINO and Segment Anything Model. Results with Mask R-CNN proved a better performance, according to F1 score and AP metrics, and a lower computational cost, according to prediction time. One of the most interesting derived applications from fruit segmentation is fruit size estimation. However, segmented fruit masks are frequently incomplete due to occlusions. Therefore, image fruit size estimation is not a straightforward process. In this work, we also propose a novel algorithm to estimate and monitor the fruit size in pixel units from the automated masks. A median relative error of 1.39% was obtained, demonstrating the potential and feasibility of future fully-automatic fruit size estimators.

Quantum Quality with Classical Cost: Ab Initio Nonadiabatic Dynamics Simulations Using the Mapping Approach to Surface Hopping.

Nonadiabatic dynamics methods are an essential tool for investigating photochemical processes. In the context of employing first-principles electronic structure techniques, such simulations can be carried out in a practical manner using semiclassical trajectory-based methods or wave packet approaches. While all approaches applicable to first-principles simulations are necessarily approximate, it is commonly thought that wave packet approaches offer inherent advantages over their semiclassical counterparts in terms of accuracy and that this trait simply comes at a higher computational cost. Here we demonstrate that the mapping approach to surface hopping (MASH), a recently introduced trajectory-based nonadiabatic dynamics method, can be efficiently applied in tandem with ab initio electronic structure. Our results even suggest that MASH may provide more accurate results than on-the-fly wave packet techniques, all at a much lower computational cost.

Coordinated Control of the Onboard and Wayside Energy Storage System of an Urban Rail Train Based on Rule Mining

There are three major challenges to the broad implementation of energy storage systems (ESSs) in urban rail transit: maximizing the absorption of regenerative braking power, enabling online global optimal control, and ensuring algorithm portability. To address these problems, a coordinated control framework between onboard and wayside ESSs is proposed in this study, and the related control strategy is obtained by transforming the global optimization problem into a combination of rule-based and local optimization problems. The design process of the strategy is divided into three parts: offline optimization, rule extraction, and local optimization. First, a genetic algorithm is used to optimize the charge/discharge threshold curves under typical operational conditions. Then, the rules of offline optimization results are obtained from an expert system, and the nonlinear rules are mined with local optimization methods. Finally, a coordinated strategy is presented based on the outcomes of the three parts. A power hardware-in-the-loop (PHIL) experimental platform based on RTLAB is built, and the above coordination strategy is verified based on data for the Beijing Metro Batong Line. The experimental results show that the strategy can effectively improve the energy-saving rate and reduce the regeneration failure rate substantially, with the effect being close to the offline global optimal solution. The algorithm proposed in this paper achieves near global optimal energy-saving optimization results with lower computational costs, and has strong portability, providing a good solution for the large-scale application of onboard and wayside energy storage coordination control.

Improved motion correction in brain MRI using 3D radial trajectory and projection moment analysis.

To develop a generalized rigid body motion correction method in 3D radial brain MRI to deal with continuous motion pattern through projection moment analysis. An assumption was made that the multichannel coil moves with the head, which was achieved by using a flexible head coil. A two-step motion correction scheme was proposed to directly extract the motion parameters from the acquired k-space data using the analysis of center-of-mass with high noise robustness, which were used for retrospective motion correction. A recursive least-squares model was introduced to recursively estimate the motion parameters for every single spoke, which used the smoothness of motion and resulted in high temporal resolution and low computational cost. Five volunteers were scanned at 3 T using a 3D radial multidimensional golden-means trajectory with instructed motion patterns. The performance was tested through both simulation and in vivo experiments. Quantitative image quality metrics were calculated for comparison. The proposed method showed good accuracy and precision in both translation and rotation estimation. A better result was achieved using the proposed two-step correction compared to traditional one-step correction without significantly increasing computation time. Retrospective correction showed substantial improvements in image quality among all scans, even for stationary scans. The proposed method provides an easy, robust, and time-efficient tool for motion correction in brain MRI, which may benefit clinical diagnosis of uncooperative patients as well as scientific MRI researches.

Enhancing security and privacy in smart agriculture: A novel homomorphic signcryption system

Technology driven Smart Agriculture is transforming modern day farming practices through digital tools and real-time data analysis. It monitors plant health in both soil and soilless cultivation, promoting environmentally responsible practices by utilizing integrated sensors and precision methods. Despite the significant potential benefits, challenges persist in ensuring data security in the smart agriculture frameworks. The existing research has explored various mechanisms to cater these issues, yet these solutions susceptible to known attacks i.e., Confidentiality, Authentication, Privacy preservation. Notably, the available schemes in open literature face performance deficiencies due to an overreliance on Rivest-Shamir-Adleman (RSA), bilinear pairing, and elliptic curve (EC). To comprehensively address these challenges, this study presents a novel approach via employing homomorphic Signcryption based on Hyper-Elliptic Curves (HEC). Our HEC based Signcryption technique enhances security with lower computational cost and communication overhead as compared to the existing techniques. Furthermore, our solution provides the best performance in terms of efficiency, addressing the privacy preservation issues with conformance to the (Confidentiality, Integrity, and Availability) CIA triad. For proof of concept and correctness, we have validated the results in Scyther tool, with the assurance that proposed scheme is correct, verified, and resistant against known set of attacks. The proposed scheme achieves a remarkable reduction in both computational and communication costs around 95.08 % and 89.28 % respectively, enhancing practicality and suitability for real-world implementation. Thus, making it a suitable candidate for employing in smart agriculture environments, encompassing both soil and soilless cultivation scenarios. This research contributes a robust, secure, cost effective, and efficient solution to the challenges posed by the evolving landscape of agriculture, paving the way for enhanced data protection and system resilience in smart agriculture systems.

Acoustic full waveform inversion with elastic wave equation forward modeling

Full-waveform inversion (FWI) is a high-resolution velocity inversion method. Currently, acoustic full waveform inversion (AFWI) has achieved promising results in marine data inversion, but it still faces significant challenges in land data inversion. One of the reasons for the failure of AFWI in land data inversion is that the acoustic wave equation does not account for the elastic effects of seismic waves, such as surface waves and converted waves. This results in errors between the synthesized acoustic waveform and the observed elastic waveform, particularly in the amplitude and phase of the P-wave. To reduce the error caused by the elastic effect, we have improved the inversion process of AFWI. During the forward modeling stage, we replace the original acoustic wave equation with the elastic wave equation to simulate the wavefield. This allows us to obtain synthetic waveforms that incorporate elasticity, thereby reducing the corresponding error. When computing model gradients, we still base on the acoustic wave equation and only update the P-wave velocity model. This approach helps maintain a lower computational cost. We tested the method on the Marmousi model and field land data, and the results demonstrate that the method successfully reduces the error caused by the elastic effect, thereby improving the clarity of the inversion results.

An electrical analogy model for performance evaluation of natural air-cooled TEG modules considering nonlinear effect of heat convection

With the outstanding merits of the simple principle and low computational cost, the electrical analogy model of thermoelectric generator (TEG) modules plays an important role in evaluating the performance of thermoelectric energy harvesting systems. However, owing to the simplification of the heat dissipation structure and difficulty in determining the heat dissipation parameters, it is difficult to accurately describe the nonlinear influence of thermal convection on the heat dissipation performance of the TEG module under natural air-cooled conditions, thereby reducing the estimation accuracy of the module temperature and output. A new nonlinear electrical analogy model to evaluate the performance of natural air-cooled TEG modules was proposed in this study. The proposed model more accurately describes the nonlinear influence of thermal convection on the thermal dynamic characteristics of the TEG module by introducing a heat transfer branch between the cold and hot side surfaces of the TEG module and the environment. A parameter extraction method based on multi-objective optimization and the non-dominated sorting genetic algorithm II (NSGA-II) was proposed to overcome the problem of the heat dissipation parameters of natural air-cooled TEG modules. The experimental results show that under natural convection conditions, the maximum estimation errors of the temperature and output voltage evaluated by the proposed model and parameter identification method are less than 6.18 % and 9.14 %, respectively. Under forced convection, the maximum estimation errors of the temperature and output voltage were less than 7.36 % and 8.71 %, respectively. The above estimation errors are significantly lower than the traditional model.

Efficient blind super-resolution imaging via adaptive degradation-aware estimation

Single image super-resolution (SISR) has achieved prominent success based on deep learning. However, most SISR methods based on the specific degradation pattern, e.g., bicubic interpolation with a bulky network structure, are unsuitable for real-world images reconstructed on edge devices. To mitigate the above limitations, unlike the majority of blind super-resolution methods, which consist of a separate complex degradation kernel projection network and a SR reconstruction network, we propose an end-to-end lightweight multi-degradation oriented network that not only can estimate degradation kernel but also reconstructs HR image fast and accurately. The core components of the designed network adopt a lightweight SR model with region non-local feature similarity learning, i.e., Region Non-Local Feature Block (RNLFB), to establish region-wise global feature correlation. Concretely, more and fewer RNLFBs are used to learn more sophisticated feature representation and relatively simple degradation representation. Then, a degradation kernel projector is equipped to fulfill adaptive degradation-aware estimation from the degradation representation. With the help of the encoded degradation kernel, the reconstruction model can learn to restore a high-resolution image from the feature representation. By doing so, the degradation kernel estimation task and the HR image reconstruction task can be accomplished without designing two complex networks separately, which guides more stable and effective network training under low computational resource costs. Extensive experiments on synthetic and real-world datasets demonstrate that our method can fulfill precise degradation kernel prediction and HR image reconstruction compared with other state-of-the-art SR methods with lower model complexity.

Structural optimization of serpentine channel water-cooled plate for lithium-ion battery modules based on multi-objective Bayesian optimization algorithm

Maintaining the battery within its optimal operating temperature range while preventing thermal runaway is crucial. Serpentine channel water-cooled plate (SCWCP) has been widely employed in battery pack cooling. The challenge lies in enhancing the cooling efficiency of SCWCP while minimizing energy consumption. Due to the high efficiency and robustness of the multi-objective Bayesian optimization (MOBO), it is employed to systematically optimize the SCWCP for lithium batteries. The width, depth, and turning radius of the serpentine flow channels are optimized to minimize both the maximum battery module temperature (Tmax) and the pumping power (PP) of the SCWCP. The MOBO process integrates structural parameter adjustments into Computational Fluid Dynamics (CFD) simulations through an automated iterative approach. Subsequently, a Pareto front is generated based on Tmax and PP, and the K-means clustering algorithm identifies four design solutions with different performance orientations. Compared with the initial design, Tmax of the optimal design decreases slightly, but with a reduction in PP of 71 %. Compared to other evolutionary algorithms, the MOBO algorithm exhibits superior computational efficiency in providing an optimal design solution set at a lower computational cost.

A kinetic partitioning method for simulating the condensation mass flux of organic vapors in a wide volatility range

Organic aerosols are ubiquitous, playing important roles in various atmospheric physicochemical processes such as the formation of cloud droplets and haze. Condensation of organic vapors, as a net effect of association with particles and dissociation from the condensed phase, is a fundamental process that drives the formation of organic aerosols. Kinetic models are often used to simulate the condensation fluxes of low-volatility organic vapors and aerosol growth. However, the widely used kinetic growth models usually calculate the evaporation of a certain species based on previous particulate compositions, without including the co-condensation of other species. Here we present a new kinetic partitioning method for calculating the condensation fluxes of organic vapors in a wide volatility range with low computational cost. In this method, the organic vapors are assumed to be in a quasi-steady state, but never reach real association-dissociation equilibrium during the simultaneous condensation of multiple species. We show a good consistency between the kinetic partitioning method and kinetic models in simulating particle mass fractions and condensation fluxes. Under relevant atmospheric conditions, we reveal that the kinetic partitioning method also reproduce the trend that low-volatility species are almost non-volatile while volatile organic compounds almost reach association-dissociation equilibrium, while there is a transition regime between them. This transition regime varies with atmospheric conditions, such as temperature and vapor concentrations. Compared with previous studies combining kinetic growth methods with equilibrium partitioning theories to simplify the condensation flux calculation, this method helps to improve accuracy without a significant expense of computation cost, and it can be applied in a wider range of atmospheric conditions such as in extremely cold atmospheres and polluted exhaust plumes.

Detailed Investigation of the Eddy Current and Core Losses in Coaxial Magnetic Gears through a Two-Dimensional Analytical Model

This work introduces a 2D model that calculates power losses in coaxial magnetic gears (CMGs). The eddy current losses of the magnets are computed analytically, whereas the core losses of the ferromagnetic segments are computed using an analytical–finite element hybrid model. The results were within 1.51% and 3.18% of those obtained from an FEA for the eddy current and core losses in the CMG for an indicative inner rotor speed of 2500 rpm. In addition, the significance of the circumferential magnet segmentation is demonstrated in the CMGs. Furthermore, a parametric investigation of the efficiency of the system for different applied external loads is carried out. Finally, a mesh sensitivity analysis is performed, along with the computation of the average power losses throughout one full period, resulting in an at least 80% reduction in computational costs with a negligible effect on accuracy. The developed model could be a valuable tool for the minimization of power losses in CMGs since it combines high accuracy with a low computational cost.

A new identification of Lagrange multipliers to study solutions of nonlinear Caputo–Fabrizio fractional problems

This research aims to provide a new identification of Lagrange multipliers using a novel hybrid method based on the Khalouta transform method and the variational iteration method and to study the solutions of nonlinear fractional problems. This method is called Khalouta variational iteration method (KHVIM). The fractional derivative is considered in Caputo–Fabrizio sense. The uniqueness and convergence results are investigated using Banach fixed point theorem. The results are obtained in the form of successive approximations corresponding to the proposed problem. To demonstrate the accuracy and effectiveness of the proposed method, three different nonlinear fractional partial differential equations are provided. Approximate solutions of the fractional equations were obtained. These solutions quickly converged to exact solutions with lower computational cost. Furthermore, the method used in this study is more generalized and allows our results to be more extensive and cover several new and existing fractional problems in the literature.

A reduced kinetic method for investigating non-local ion heat transport in ideal multi-species plasmas

A reduced kinetic method (RKM) with a first-principles collision operator is introduced in a 1D2V planar geometry and implemented in a computationally inexpensive code to investigate non-local ion heat transport in multi-species plasmas. The RKM successfully reproduces local results for multi-species ion systems and the important features expected to arise due to non-local effects on the heat flux are captured. In addition to this, novel features associated with multi-species, as opposed to single species, cases are found. Effects of non-locality on the heat flux are investigated in mass and charge symmetric and asymmetric ion mixtures with temperature, pressure, and concentration gradients. In particular, the enthalpy flux associated with diffusion is found to be insensitive to sharp pressure and concentration gradients, increasing its significance in comparison to the conductive heat flux driven by temperature gradients in non-local scenarios. The RKM code can be used for investigating other kinetic and non-local effects in a broader plasma physics context. Due to its relatively low computational cost it can also serve as a practical non-local ion heat flux closure in hydrodynamic simulations or as a training tool for machine learning surrogates.

Mutual annotation-based prediction of protein domain functions with Domain2GO.

Identifying unknown functional properties of proteins is essential for understanding their roles in both health and disease states. The domain composition of a protein can reveal critical information in this context, as domains are structural and functional units that dictate how the protein should act at the molecular level. The expensive and time-consuming nature of wet-lab experimental approaches prompted researchers to develop computational strategies for predicting the functions of proteins. In this study, we proposed a new method called Domain2GO that infers associations between protein domains and function-defining gene ontology (GO) terms, thus redefining the problem as domain function prediction. Domain2GO uses documented protein-level GO annotations together with proteins' domain annotations. Co-annotation patterns of domains and GO terms in the same proteins are examined using statistical resampling to obtain reliable associations. As a use-case study, we evaluated the biological relevance of examples selected from the Domain2GO-generated domain-GO term mappings via literature review. Then, we applied Domain2GO to predict unknown protein functions by propagating domain-associated GO terms to proteins annotated with these domains. For function prediction performance evaluation and comparison against other methods, we employed Critical Assessment of Function Annotation 3 (CAFA3) challenge datasets. The results demonstrated the high potential of Domain2GO, particularly for predicting molecular function and biological process terms, along with advantages such as producing interpretable results and having an exceptionally low computational cost. The approach presented here can be extended to other ontologies and biological entities to investigate unknown relationships in complex and large-scale biological data. The source code, datasets, results, and user instructions for Domain2GO are available at https://github.com/HUBioDataLab/Domain2GO. Additionally, we offer a user-friendly online tool at https://huggingface.co/spaces/HUBioDataLab/Domain2GO, which simplifies the prediction of functions of previously unannotated proteins solely using amino acid sequences.

Mixed-integer quadratic optimization and iterative clustering techniques for semi-supervised support vector machines

Among the most famous algorithms for solving classification problems are support vector machines (SVMs), which find a separating hyperplane for a set of labeled data points. In some applications, however, labels are only available for a subset of points. Furthermore, this subset can be non-representative, e.g., due to self-selection in a survey. Semi-supervised SVMs tackle the setting of labeled and unlabeled data and can often improve the reliability of the results. Moreover, additional information about the size of the classes can be available from undisclosed sources. We propose a mixed-integer quadratic optimization (MIQP) model that covers the setting of labeled and unlabeled data points as well as the overall number of points in each class. Since the MIQP’s solution time rapidly grows as the number of variables increases, we introduce an iterative clustering approach to reduce the model’s size. Moreover, we present an update rule for the required big-M values, prove the correctness of the iterative clustering method as well as derive tailored dimension-reduction and warm-starting techniques. Our numerical results show that our approach leads to a similar accuracy and precision than the MIQP formulation but at much lower computational cost. Thus, we can solve larger problems. With respect to the original SVM formulation, we observe that our approach has even better accuracy and precision for biased samples.

LB-SCAM: a learning-based method for efficient large-scale sensitivity analysis and tuning of the Single Column Atmosphere Model (SCAM)

Abstract. The single-column model, with its advantages of low computational cost and fast execution speed, can assist users in gaining a more intuitive understanding of the impact of parameters on the simulated results of climate models. It plays a crucial role in the study of parameterization schemes, allowing for a more direct exploration of the influence of parameters on climate model simulations. In this paper, we employed various methods to conduct sensitivity analyses on the 11 parameters of the Single Column Atmospheric Model (SCAM). We explored their impact on output variables such as precipitation, temperature, humidity, and cloud cover, among others, across five test cases. To further expedite experimentation, we utilized machine learning methods to train surrogate models for the aforementioned cases. Additionally, three-parameter joint perturbation experiments were conducted based on these surrogate models to validate the combined parameter effects on the results. Subsequently, targeting the sensitive parameter combinations identified from the aforementioned experiments, we further conducted parameter tuning for the corresponding test cases to minimize the discrepancy between the results of SCAM and observational data. Our proposed method not only enhances model performance but also expedites parameter tuning speed, demonstrating good generality at the same time.

Multifidelity topology design of a maritime survey operation with UUVs

Multifidelity topology design of a maritime survey operation with UUVs

Exploring the Mechanisms behind Non-aromatic Fluorescence with the Density Functional Tight Binding Method.

Recent experimental findings reveal nonconventional fluorescence emission in biological systems devoid of conjugated bonds or aromatic compounds, termed non-aromatic fluorescence (NAF). This phenomenon is exclusive to aggregated or solid states and remains absent in monomeric solutions. Previous studies focused on small model systems in vacuum show that the carbonyl stretching mode along with strong interaction of short hydrogen bonds (SHBs) remains the primary vibrational mode explaining NAF in these systems. In order to simulate larger model systems taking into account the effects of the surrounding environment, in this work we propose using the density functional tight-binding (DFTB) method in combination with non-adiabatic molecular dynamics (NAMD) and the mixed quantum/molecular mechanics (QM/MM) approach. We investigate the mechanism behind NAF in the crystal structure of l-pyroglutamine-ammonium, comparing it with the related nonfluorescent amino acid l-glutamine. Our results extend our previous findings to more realistic systems, demonstrating the efficiency and robustness of the proposed DFTB method in the context of NAMD in biological systems. Furthermore, due to its inherent low computational cost, this method allows for a better sampling of the nonradiative events at the conical intersection which is crucial for a complete understanding of this phenomenon. Beyond contributing to the ongoing exploration of NAF, this work paves the way for future application of this method in more complex biological systems such as amyloid aggregates, biomaterials, and non-aromatic proteins.

Data‐driven based optimal output feedback control with low computation cost

SummaryA partial model‐free, data‐driven adaptive optimal output feedback (OPFB) control scheme with low computational cost continuous‐time is proposed in this paper. The design objective is to obtain the optimal control law by using measurable input and output data, without some knowledge of system model information. Firstly, the system states are decoupled into measurable and unmeasurable parts, and a new state‐space equation is built to estimate the unmeasurable states by using a reduced‐order observer. Based on this, a parametrization method is utilized to reconstruct the system states. Subsequently, by using the reconstructed states, the adaptive dynamic programming (ADP) Bellman equations based on policy‐iteration (PI) and value‐iteration (VI) are presented to solve the control problems with initially stable and unstable conditions, respectively. Then, the convergence of the system is proved. Compared with the early proposed OPFB algorithms, only the unknown internal state needs to be reconstructed. Therefore, the computation cost and design complexity are reduced for the proposed scheme. The effectiveness of the proposed scheme is verified through two numerical simulations. In addition, a practical inverted pendulum experiment is carried out to demonstrate the performance of the proposed scheme.