Search

Acoustic holography (AH) is a technique with significant potential in realms, such as biomedicine, industry, and augmented reality. The implementation of acoustic holograms can be realized by a passive approach or active ones. Although the passive approach (by a 3D printer) can achieve high-quality acoustic field generation, it is constrained by high manufacturing costs and static field control. On the contrary, the active approach with a phased transducer array (PTA) as the latest technique stands out since it supports dynamic, flexible, and reconfigurable acoustic field generation. However, current PTA-based AH techniques face the drawback of inferior acoustic field fineness due to the Spatial Bandwidth Product (SBP) limit of PTA, which hinders the application of PTA in precise tasks, such as neural electrodes and microfluidics control. To address this issue, we propose a super-resolution acoustic holography (SRAH) method inspired by the concept of super-resolution in ultrasonic imaging and computer vision, by which we can generate acoustic fields reaching the physical diffraction limit of acoustic waves regardless SBP of PTA. In other words, this method enables high-SBP acoustic field generation with low-SBP PTA. The method is based on self-supervised learning, integrating a generative adversarial network and a physical model of acoustic wave propagation, specifically the linear accumulation method. Both simulation and experimental results demonstrate that the proposed method can generate high-fidelity acoustic fields suitable for intricate tasks with low-SBP PTA. Moreover, the performance of the algorithm improves as the target SBP increases. Therefore, the proposed SRAH method shows great potential for applications requiring elaborate manipulation. [ABSTRACT FROM AUTHOR]

An accurate estimation of three-dimensional (3D) temperature fields in channel flows is challenging but critical for many important applications such as heat exchangers, radiation energy collectors, and enhanced geothermal systems. In this paper, we demonstrate the possibility of inferring temperature fields from concentration fields for laminar convection flows in a 3D channel using a machine learning (ML) approach. The study involves generation of data using 3D numerical simulations, application of deep learning methodology using conditional generative adversarial networks (cGANs), and analysis of how dataset selection affects model performance. The model is also tested for applicability in different convection scenarios. Results show that cGANs can successfully infer temperature fields from concentration fields, and the reconstruction accuracy is sensitive to the training dataset selected. In this study, we demonstrate how ML can be used to overcome the limitations of traditional heat and mass analogy functions widely used in heat transfer research. [ABSTRACT FROM AUTHOR]

Laser-based powder bed fusion (L-PBF) has become the de facto choice for metal additive manufacturing (AM) processes. Even after considerable research investments, components manufactured using L-PBF lack consistency in their quality. Realizing the crucial role of the melt pool in controlling the final build quality, we predict the morphology of the melt pool directly from the build commands in an L-PBF process. We leverage machine learning techniques to predict quantitative attributes like the size as well as qualitative attributes like the shape of the melt pool. The area of the melt pool is predicted using an LSTM network. The outlined LSTM-based approach estimates the area with $ 90.7\% $ 90.7 % accuracy. The shape is inferred by synthesising the images of the melt pool by using a Melt Pool Generative Adversarial Network (MP-GAN). The synthetic images attain a structural similarity score of 0.91. The precision and accuracy of the results showcase the efficacy of the outlined approach and pave the way for real-time monitoring and control of the melt pool to build products with consistently better quality. [ABSTRACT FROM AUTHOR]

4D MRI with high spatiotemporal resolution is desired for image-guided liver radiotherapy. Acquiring densely sampled k-space data is time-consuming. Accelerated acquisition with sparse sampling is desirable but often causes degraded image quality or long reconstruction time. We propose the Reconstruct Paired Conditional Generative Adversarial Network (Re-Con-GAN) for shortening the 4D MRI reconstruction time while maintaining the reconstruction quality. Methods: Patients underwent free-breathing liver 4D MRI were included in the study. Fully- and retrospectively under-sampled data at 3, 6 and 10 times (3x, 6x and 10x) were first reconstructed using the nuFFT algorithm. Re-Con-GAN then trained input and output in pairs. Three types of networks, ResNet9, UNet and reconstruction swin transformer, were explored as generators. PatchGAN was selected as the discriminator. Re-Con-GAN processed the data (3D+t) as temporal slices (2D+t). A total of 48 patients with 12332 temporal slices were split into training (37 patients with 10721 slices) and test (11 patients with 1611 slices). Compressed sensing (CS) reconstruction with spatiotemporal sparsity constraint was used as a benchmark. Reconstructed image quality was further evaluated with a liver gross tumor volume (GTV) localization task using Mask-RCNN trained from a separate 3D static liver MRI dataset (70 patients; 103 GTV contours).Results: Re-Con-GAN consistently achieved comparable/better PSNR, SSIM, and RMSE scores compared to CS/UNet models. The inference time of Re-Con-GAN, UNet and CS are 0.15s, 0.16s, and 120s. The GTV detection task showed that Re-Con-GAN and CS, compared to UNet, better improved the dice score (3x Re-Con-GAN 80.98%; 3x CS 80.74%; 3x UNet 79.88%) of unprocessed under-sampled images (3x 69.61%). Conclusion: A generative network with adversarial training is proposed with promising and efficient reconstruction results demonstrated on an in-house dataset. The rapid and qualitative reconstruction of 4D liver MR has the potential to facilitate online adaptive MR-guided radiotherapy for liver cancer.

The demand for specialized hardware to train AI models has increased in tandem with the increase in the model complexity over the recent years. Graphics processing unit (GPU) is one such hardware that is capable of parallelizing operations performed on a large chunk of data. Companies like Nvidia, AMD, and Google have been constantly scaling-up the hardware performance as fast as they can. Nevertheless, there is still a gap between the required processing power and processing capacity of the hardware. To increase the hardware utilization, the software has to be optimized too. In this paper, we present some general GPU optimization techniques we used to efficiently train the optiGAN model, a Generative Adversarial Network that is capable of generating multidimensional probability distributions of optical photons at the photodetector face in radiation detectors, on an 8GB Nvidia Quadro RTX 4000 GPU. We analyze and compare the performances of all the optimizations based on the execution time and the memory consumed using the Nvidia Nsight Systems profiler tool. The optimizations gave approximately a 4.5x increase in the runtime performance when compared to a naive training on the GPU, without compromising the model performance. Finally we discuss optiGANs future work and how we are planning to scale the model on GPUs.

Control over the two-dimensional electron gas (2DEG) in AlGaN/GaN heterostructures is crucial for their practical applications in current semiconducting devices. However, the oxide surface structures inducing 2DEG are still ambiguous because oxide-stoichiometry (OS) matching structures possess occupied surface donor states at 1.0–1.8 eV below the conduction band minimum of bulk but are usually not available in energy than electron counting (EC) rule structures. In this work, a global optimization algorithm was introduced to explore the possible oxidation structures on GaN (0001) and AlN (0001) surfaces; the method was demonstrated to be available due to the fact that the reported oxidized structures were reproduced at each stoichiometry. Interestingly, the two similar oxide structures with close energy were found in each oxide-bilayer, which can be used to clarify the experimental observations of disordered surface oxide layers below 550 °C. Additionally, new stable oxidation structures with low surface energy were proposed. Interestingly, the new OS matching structures are proposed with remarkably lower energy than EC rule structures under cation-rich and oxygen-poor conditions, which is caused by the large formation enthalpy of Al2O3 and Ga2O3. Further electronic structure calculations demonstrate that the new OS structures possess highest occupied states above the half of the gap and are the origin of 2DEG in AlGaN/GaN heterostructures. [ABSTRACT FROM AUTHOR]

Generative Adversarial Networks (GANs) have shown immense potential in fields such as text and image generation. Only very recently attempts to exploit GANs to statistical-mechanics models have been reported. Here we quantitatively test this approach by applying it to a prototypical stochastic process on a lattice. By suitably adding noise to the original data we succeed in bringing both the Generator and the Discriminator loss functions close to their ideal value. Importantly, the discreteness of the model is retained despite the noise. As typical for adversarial approaches, oscillations around the convergence limit persist also at large epochs. This undermines model selection and the quality of the generated trajectories. We demonstrate that a simple multi-model procedure where stochastic trajectories are advanced at each step upon randomly selecting a Generator leads to a remarkable increase in accuracy. This is illustrated by quantitative analysis of both the predicted equilibrium probability distribution and of the escape-time distribution. Based on the reported findings, we believe that GANs are a promising tool to tackle complex statistical dynamics by machine learning techniques. [ABSTRACT FROM AUTHOR]

Summary: Anomaly event recognition and identification has a crucial part in several areas, particularly in night vision environments. Conventional techniques of event recognition are hugely based upon data extracted from certain images for classification purposes. This needs users to select suitable features to establish the feature depictions for actual images per definite situations. Manual feature selection is laborious as well as heuristic tasks and the features obtained in this manner generally have worse robustness. Here, a Faster Region‐based Convolutional fused Social Generative Adversarial Network (FRC‐SGAN) is designed for anomaly event recognition in a night vision environment. At the cloud, key frame extraction, pre‐processing, feature extraction, human detection (HD) and anomalous event recognition are carried out. Initially, input video from the database is subjected to perform pre‐processing. The visibility enhancement is utilized for pre‐processing. Thereafter, features like ResNet features, texture features and statistical features are extracted. Then, HD is accomplished by DeepJoint segmentation with chord distance. Finally, anomalous detection is done by FRC‐SGAN that is the incorporation of Fast Regional Convolutional Neural Network (FR‐CNN) and Social Generative Adversarial Network (SGAN). In addition, FRC‐SGAN acquired 90.8% of accuracy, 89.7% of precision, and 89.2% of recall. [ABSTRACT FROM AUTHOR]

Summary: Wireless body sensor network (WBSN) is essential for monitoring patients' health problems and offers a low‐cost option for various healthcare applications. In this manuscript, a Novel Health Monitoring Approach for WBSNs (DIWGAN‐WBSN) is proposed, which uses Dual Interactive Wasserstein Generative Adversarial Network (DIWGAN) optimized with War Strategy Optimization Algorithm (WSOA). After sensing the aforementioned attribute information, it is the responsibility of WBSN nodes to transfer the sensed data to the sink node. The Volcano Eruption Algorithm (VEA) is applied to select the optimum cluster heads in WBSN. The results from VEA are fed to the target node; it consists of DIWGAN to classify the health records and to portray the patient's health status. Generally, DIWGAN does not adopt any optimization methods for measuring the ideal parameters and guaranteeing accurate health monitoring and risk assessment. So the proposed WSOA is considered to enhance the DIWGAN. The proposed method is activated in MATLAB; its efficacy is estimated under performance metrics, like precision, specificity, accuracy, and energy utilization. The proposed approach attains 23.9%, 21.34%, and 51.09% higher accuracy; 21.45%, 13.94%, and 20.6% higher precision; 31.32%, 29.61%, and 11.03% higher specificity; and 20.9%, 19.87%, and 24.6% lower energy utilization for HD classification using the Cleveland database than the existing methods like back propagation neural network‐based risk detection in WBSN for health monitoring, random forest algorithm–based health monitoring in WBSN, and ensemble deep learning and feature fusion for health monitoring using WBSN methods, respectively. [ABSTRACT FROM AUTHOR]

Cancer involves abnormal cell growth, with types like intestinal and oesophageal cancer often diagnosed in advanced stages, making them hard to cure. Symptoms are like burning sensations in the stomach and swallowing difficulties are specified as colorectal cancer. Deep learning significantly impacts the medical image processing and diagnosis, offering potential improvements in accuracy and efficiency. The Kruskal Szekeres Generative Adversarial Network Augmented Deep Autoencoder (KSGANA-DA) is introduced for early colorectal cancer detection and it comprises two stages; Initial stage, data augmentation uses Affine Transform via Random Horizontal Rotation and Geometric Transform via Kruskal–Szekeres that coordinates to improve the training dataset diversity, boosting detection performance. The second stage, a Deep Autoencoder Anatomical Landmark-based Image Segmentation preserves edge pixel spatial locations, improving precision and recall for early boundary detection. Experiments validate KSGANA-DA performance and different existing methods are implemented into Python. The results of KSGANA-DA are to provide higher precision by 41%, recall by 7%, and lesser training time by 46% than compared to conventional methods. [ABSTRACT FROM AUTHOR]

In recent years, the growth of social media has significantly increased in people's daily lives. Social media platforms offer various benefits to users. Still, they also come with vulnerabilities, including spam, posing threats to media owners and users, leading to user fatigue, financial and security damages, sensitive information disclosure, and media credibility decline. Due to the high volume of spam, manual detection is challenging. However, employing machine and deep learning methods makes it possible to identify and block them effectively. Also, class imbalances in real-world datasets are prevalent, with more non-spam instances. This poses challenges for models in effectively identifying samples with limited data. In this regard, we propose a neural network model to detect spam based on its textual content. Furthermore, using a code-based Generative Adversarial Networks, the model balances the number of spam and non-spam samples. We used the Wasserstein distance with penalty regularization to measure the convergence between the actual and generated data distributions to address the mode collapse and vanishing gradient issues. Our evaluation of experimental results from two benchmark datasets demonstrates a significant improvement in spam classification and addresses the issue of imbalanced datasets. We compared our model with three other spam detection models, one of which employed data augmentation. In our evaluation, the proposed model achieved an accuracy of 98.2% and an F1-score of 93.0%. In this study, we also assessed the quality of text generated by our proposed model using the perplexity metric to evaluate the fluency and accuracy of the spam content. [ABSTRACT FROM AUTHOR]

Background and Aim Methods Results Conclusion Computed tomography of the abdomen exhibits subtle and complex features of liver lesions, subjectively interpreted by physicians. We developed a deep learning‐based localization and classification (DLLC) system for focal liver lesions (FLLs) in computed tomography imaging that could assist physicians in more robust clinical decision‐making.We conducted a retrospective study (approval no. EMRP‐109‐058) on 1589 patients with 17 335 slices with 3195 FLLs using data from January 2004 to December 2020. The training set included 1272 patients (male: 776, mean age 62 ± 10.9), and the test set included 317 patients (male: 228, mean age 57 ± 11.8). The slices were annotated by annotators with different experience levels, and the DLLC system was developed using generative adversarial networks for data augmentation. A comparative analysis was performed for the DLLC system versus physicians using external data.Our DLLC system demonstrated mean average precision at 0.81 for localization. The system's overall accuracy for multiclass classifications was 0.97 (95% confidence interval [CI]: 0.95–0.99). Considering FLLs ≤ 3 cm, the system achieved an accuracy of 0.83 (95% CI: 0.68–0.98), and for size > 3 cm, the accuracy was 0.87 (95% CI: 0.77–0.97) for localization. Furthermore, during classification, the accuracy was 0.95 (95% CI: 0.92–0.98) for FLLs ≤ 3 cm and 0.97 (95% CI: 0.94–1.00) for FLLs > 3 cm.This system can provide an accurate and non‐invasive method for diagnosing liver conditions, making it a valuable tool for hepatologists and radiologists. [ABSTRACT FROM AUTHOR]

The current quantitative retrieval of Aerosol Optical Depth (AOD) typically uses Top-of-atmosphere (TOA) reflectance data obtained by the scale and offset terms of radiometric calibration. Errors can be introduced during the conversion of Digital Number (DN) values to TOA reflectance, restraining the accurate retrieval of AOD. Particularly when surface reflectance is high, conversion errors can significantly distort the retrieval of Aerosol Optical Depth (AOD), given the minimal impact of aerosols on the radiation detected by satellite sensors. To counteract this, we introduce an innovative retrieval algorithm that harnesses a Generative Adversarial Network (GANN), a deep learning (DL) model, to accurately derive AOD from the scaled digital number (DN) values of multi-spectral satellite imagery. The DL technology enables the use of scaled DN values for AOD retrieval, bypassing the conversion errors associated with TOA reflectance since it relies on a sample learning approach instead of radiative transfer calculations. To ensure the number and representativeness of training samples, a total of 37,103 samples from different underlying surfaces from 2014 to 2018 were obtained using a space-time matching strategy. Our K-cross validation demonstrates that employing DN values for AOD retrieval promises to yield higher accuracy compared to using TOA reflectance. Independent 2019 data were used to estimate GANN, MCD19A2 and MOD04_3K aerosol products. Our model showed superior performance compared to the other products in terms of a higher correlation (R = 0.8184) with AERONET AOD and obtaining more reliable retrievals (7601). The sensitivity experiment suggests that GANN exhibits limitations in areas characterized by high aerosol loads and heterogeneity. Our study can give a novel insight into AOD retrieval based on multi-spectral satellite and DL models. [ABSTRACT FROM AUTHOR]

Advancements in infrastructure management have significantly benefited from automatic pavement crack detection systems, relying on image processing enhanced by high‐resolution imaging and machine learning. However, image and motion blur substantially challenge the accuracy of crack detection and analysis. Nevertheless, research on mitigating motion blur remains sparse. This study introduces an effective image processing system adept at deblurring and segmentation, employing a generative adversarial network (GAN) with UNet as the generator and Wasserstein GAN with Gradient Penalty (WGAN‐gp) as the loss function. This approach performs exceptionally in deblurring pavement crack images and improves segmentation accuracy. Models were trained with sharp and artificially blurred images, with WGAN‐gp surpassing other loss functions in effectiveness. This research innovatively suggests assessing deblurring quality through segmentation accuracy in addition to peak signal‐to‐noise ratio (PSNR) and structural similarity (SSIM), revealing that PSNR and SSIM may not fully capture deblurring effectiveness for pavement crack images. An extensive evaluation of various generators, including UNet, lightweight UNet, TransUNet, DeblurGAN, DeblurGAN‐v2, and MIMO‐UNet, identifies the superior performance of UNet on simulated motion blur. Validation with actual motion‐blurred images confirms the effectiveness of the proposed model. These findings demonstrate that GAN‐based models have great potential in overcoming motion blur challenges in pavement crack detection systems, marking a notable advancement in the field. [ABSTRACT FROM AUTHOR]

This study proposes a synthetic data generation model to create a classification framework for cerebellar ataxia patients using trajectory data from the visuomotor adaptation task. The classification objectives include patients with cerebellar ataxia, age-matched normal individuals, and young healthy subjects. Synthetic data for the three classes is generated based on class conditions and random noise by leveraging a combination of conditional adversarial generative neural networks and reconstruction networks. This synthetic data, alongside real data, is utilized as training data for the patient classification model to enhance classification accuracy. The fidelity of the synthetic data is assessed visually to measure the validity and diversity of the generated data qualitatively while quantitatively evaluating distribution similarity to real data. Furthermore, the clinical efficacy of the patient classification model employing synthetic data is demonstrated by showcasing improved classification accuracy through a comparative analysis between results obtained using solely real data and those obtained when both real and synthetic data are utilized. This methodological approach holds promise in addressing data insufficiency in the digital healthcare domain, employing deep learning methodologies, and developing early disease diagnosis tools. [ABSTRACT FROM AUTHOR]

Cyber-physical system is one of the essential components of wireless sensor network (WSN). WSN conjointly senses, collects, analyzes and communicates data on recognized items at service part before transmitting this data to the owner of the network. Blackhole, Grayhole, Flooding, Scheduling are the common WSN attacks that rapidly damage the system. Intrusion detection systems, such as WSN have drawbacks, like less detection rates, high computation costs and high percentage of false alarms, because sensor nodes have limited resources. Therefore, a Self Attention Generative Adversarial Capsule Network as an intrusion detection scheme for WSN (SAGACN-ID-WSN) is proposed in this paper to address the aforementioned issues. Initially, the data are amassed from WSN-DS dataset. Then, the data are sent for pre-processing. The input data are preprocessed using the Guided Box Filtering (GBF) method. Then, the preprocessed data are provided to the Atomic Orbital Search (AOS), Tasmanian Devil Optimization (TDO) and Ebola Optimization Search (EOS) algorithms for feature selection. The attacks are divided into four categories: normal, Grayhole attack, Black hole attack, flooding attack, and scheduling attack. The proposed ID-WSN is implemented in Network simulator 2 (NS-2) utilizing the dataset of WSN-DS. The metrics, like accuracy, precision, recall, F-measure, ROC and computational time are analyzed. The proposed SAGACN-ID-WSN technique attains 25.5%, 20.12% and 20.7% high accuracy, 51.136%, 59.04% and 32.81%; higher precision, 2.292%, 1.51% and 3.915% lower error rate compared to the existing models, like Dataset for Intrusion Detection Systems in WSN (SLGBM-ID-WSN), multi-layer machine learning-based ID-WSN (NB-RF-ID-WSN), deep learning and entity embedding-base intrusion detection systems in WSN, respectively. [ABSTRACT FROM AUTHOR]

The surface quality of aluminium alloy castings is crucial to quality control. Aiming to address the challenges of limited samples and extensive computation in deep learning‐based surface defect detection for aluminium alloy castings, this paper proposes a surface defect detection method based on data enhancement and the Casting Real‐Time DEtection TRansformer. First, to tackle the issue of small sample sizes and uneven distribution in surface defect data sets of aluminium alloy castings, ECA‐MetaAconC Deep Convolution Generative Adversarial Networks is proposed for generating defects with fewer samples and employ the image augmentation (IMGAUG) library for sample enhancement. Second, building upon the Real‐Time DEtection TRansformer (RT‐DETR), a lightweight partial‐rep convolution is designed to decrease the network's parameter count. Simultaneously, the Deformable attention module and the DRBC3 module are introduced to enhance the neck network, thereby improving the model's capability to capture information and enhancing its detection performance. Compared to RT‐DETR, this method reduces the number of model parameters by 38.7%, increases mAP by 1.5%, and achieves a frame rate that is 1.58 times higher than the original model. The experimental results demonstrate that this method can effectively and accurately detect surface defects in aluminium alloy castings, satisfying industrial requirements. [ABSTRACT FROM AUTHOR]

Insulators, as typical equipment for distribution networks, provide good electrical insulation between live conductors and earth. Timely and accurate detection is essential for insulator detection issues. However, as the complexity of neural networks increases, the detection efficiency is often lower. Therefore, this paper proposes a fast insulator positioning and defect detection method. Firstly, for insulator target localization, the SqueezeNet network is improved using ECA attention mechanism. In addition, to address the issue of low defect detection accuracy, a joint algorithm has been proposed. The integration of convolutional variational autoencoder (CVAE) and generative adversarial network (GAN) solve their own shortcomings due to different image focus angles. The target localization accuracy reaches 94.30%, and the defect detection accuracy reaches 89.60%. It solves the problems of difficulty in locating small targets in a large field of view and inaccurate detection due to a small number of abnormal samples. This method has been tried and tested in practical distribution network systems. [ABSTRACT FROM AUTHOR]

Background: Bioactive peptides are important bioactive molecules composed of short-chain amino acids that play various crucial roles in the body, such as regulating physiological processes and promoting immune responses and antibacterial effects. Due to their significance, bioactive peptides have broad application potential in drug development, food science, and biotechnology. Among them, understanding their biological mechanisms will contribute to new ideas for drug discovery and disease treatment. Results: This study employs generative adversarial capsule networks (CapsuleGAN), gated recurrent units (GRU), and convolutional neural networks (CNN) as base classifiers to achieve ensemble learning through voting methods, which not only obtains high-precision prediction results on the angiotensin-converting enzyme (ACE) inhibitory peptides dataset and the anticancer peptides (ACP) dataset but also demonstrates effective model performance. For this method, we first utilized the protein language model—evolutionary scale modeling (ESM-2)—to extract relevant features for the ACE inhibitory peptides and ACP datasets. Following feature extraction, we trained three deep learning models—CapsuleGAN, GRU, and CNN—while continuously adjusting the model parameters throughout the training process. Finally, during the voting stage, different weights were assigned to the models based on their prediction accuracy, allowing full utilization of the model's performance. Experimental results show that on the ACE inhibitory peptide dataset, the balanced accuracy is 0.926, the Matthews correlation coefficient (MCC) is 0.831, and the area under the curve is 0.966; on the ACP dataset, the accuracy (ACC) is 0.779, and the MCC is 0.558. The experimental results on both datasets are superior to existing methods, demonstrating the effectiveness of the experimental approach. Conclusion: In this study, CapsuleGAN, GRU, and CNN were successfully employed as base classifiers to implement ensemble learning, which not only achieved good results in the prediction of two datasets but also surpassed existing methods. The ability to predict peptides with strong ACE inhibitory activity and ACPs more accurately and quickly is significant, and this work provides valuable insights for predicting other functional peptides. The source code and dataset for this experiment are publicly available at https://github.com/Zhou-Jianren/bioactive-peptides. [ABSTRACT FROM AUTHOR]

Background Purpose Methods Results Conclusion Automatic primary gross tumor volume (GTVp) segmentation for nasopharyngeal carcinoma (NPC) is a quite challenging task because of the existence of similar visual characteristics between tumors and their surroundings, especially on computed tomography (CT) images with severe low contrast resolution. Therefore, most recently proposed methods based on radiomics or deep learning (DL) is difficult to achieve good results on CT datasets.A peritumoral radiomics‐guided generative adversarial network (PRG‐GAN) was proposed to address this challenge.A total of 157 NPC patients with CT images was collected and divided into training, validation, and testing cohorts of 108, 9, and 30 patients, respectively. The proposed model was based on a standard GAN consisting of a generator network and a discriminator network. Morphological dilation on the initial segmentation results from GAN was first conducted to delineate annular peritumoral region, in which radiomics features were extracted as priori guide knowledge. Then, radiomics features were fused with semantic features by the discriminator's fully connected layer to achieve the voxel‐level classification and segmentation. The dice similarity coefficient (DSC), 95% Hausdorff distance (HD95), and average symmetric surface distance (ASSD) were used to evaluate the segmentation performance using a paired samples t‐test with Bonferroni correction and Cohen's d (d) effect sizes. A two‐sided p‐value of less than 0.05 was considered statistically significant.The model‐generated predictions had a high overlap ratio with the ground truth. The average DSC, HD95, and ASSD were significantly improved from 0.80 ± 0.12, 4.65 ± 4.71  mm, and 1.35 ± 1.15  mm of GAN to 0.85 ± 0.18 (p = 0.001, d = 0.71), 4.15 ± 7.56 mm (p = 0.002, d = 0.67), and 1.11 ± 1.65 mm (p < 0.001, d = 0.46) of PRG‐GAN, respectively.Integrating radiomics features into GAN is promising to solve unclear border limitations and increase the delineation accuracy of GTVp for patients with NPC. [ABSTRACT FROM AUTHOR]

ABSTRACT The rapid expansion of Internet of Things (IoT) applications presents unprecedented challenges for antenna design, demanding solutions that are versatile, efficient, and capable of operating across multiple frequency bands. This paper addresses these challenges through the development of a design of a multiband antenna using Auxiliary Classifier Wasserstein Generative Adversarial Network for IoT applications (DMA‐ACWGAN‐IoT). Here, the Auxiliary Classifier Wasserstein Generative Adversarial Network (ACWGAN) is employed to generate synthetic data representing electromagnetic field distributions and antenna characteristics across various frequencies. The proposed metamaterial‐based Multiple‐Input Multiple‐Output (MIMO) antenna contains four discrete elements, each provided by a micro strip feed. The structures overall width and length are 60 and 52 mm. The metamaterial is printed on a patch and dispersed from the fields with the most effective coupling. The proposed structures strong impedance bandwidth works at 8.3 GHz, 10.8 GHz, 12.3 GHz, 13.7 GHz, 16.1 GHz, and 18.1 GHz, fabricating the proposed antenna prototype using a substrate made of FR‐4 material. The proposed DMA‐ACWGAN‐IoT design provides 6.5 dB maximum gain and 25.40%, 21.60%, and 20.05% higher efficiency compared to existing dual band antenna design with resonance frequency prediction under machine learning models (DBA‐PRF‐ML), machine learning verification depending on a distinctive SWB multiple slotted four‐port high isolated MIMO antenna loaded with metasurface for the applications of IoT (SWB‐MIMO‐IOT‐ML), and dual‐band miniaturized composite right–left‐handed transmission line ZOR antenna along machine learning method for microwave communication (DB‐ZORA‐MC‐ML). [ABSTRACT FROM AUTHOR]

A novel damage detection model, which utilizes the spatiotemporal characteristics of the acceleration data, is proposed to assess the structural integrity of railway bridges. For this, the measured acceleration data are decomposed into several intrinsic mode functions (IMFs) using the sparse random mode decomposition model. The generated IMFs are subsequently integrated into the enhanced time series conditional generative adversarial network model to identify possible damage in bridges across various frequency bands. The influence of environmental and operational variables (EOVs), particularly temperature fluctuations, was also investigated. The proposed model was verified using both numerical and experimental data from a plate girder bridge. Further validation was conducted using the Z24 bridge dataset, and damage cases under the influence of EOVs were successfully predicted. Throughout the validation process, various anomaly metrics were introduced to establish a threshold value, and a covariance‐based time domain metric was proven to be the most effective in our cases. [ABSTRACT FROM AUTHOR]

The availability of comprehensive datasets is a crucial challenge for developing artificial intelligence (AI) models in various applications and fields. The lack of large and diverse public fabric defect datasets forms a major obstacle to properly and accurately developing and training AI models for detecting and classifying fabric defects in real-life applications. Models trained on limited datasets struggle to identify underrepresented defects, reducing their practicality. To address these issues, this study suggests using a conditional generative adversarial network (cGAN) for fabric defect data augmentation. The proposed image-to-image translator GAN features a conditional U-Net generator and a 6-layered PatchGAN discriminator. The conditional U-Network (U-Net) generator can produce highly realistic synthetic defective samples and offers the ability to control various characteristics of the generated samples by taking two input images: a segmented defect mask and a clean fabric image. The segmented defect mask provides information about various aspects of the defects to be added to the clean fabric sample, including their type, shape, size, and location. By augmenting the training dataset with diverse and realistic synthetic samples, the AI models can learn to identify a broader range of defects more accurately. This technique helps overcome the limitations of small or unvaried datasets, leading to improved defect detection accuracy and generalizability. Moreover, this proposed augmentation method can find applications in other challenging fields, such as generating synthetic samples for medical imaging datasets related to brain and lung tumors. [ABSTRACT FROM AUTHOR]

ABSTRACT In practical industrial applications, equipment usually operates normally and failures are relatively rare, resulting in serious imbalances in the collected data. This imbalance leads to issues such as overfitting, instability, and poor robustness, significantly reducing the accuracy and stability of fault diagnosis system. To address these challenges, this research proposes a method for imbalanced data augmentation and industrial process fault diagnosis based on improved Generative Adversarial Network (GAN). The method adopts Wasserstein distance with gradient penalty and integrates residual connections into the architecture of the generator. This innovation not only helps improve gradient transfer in the generator, but also significantly enhances the data generation capabilities of the generative model through improving the stability of training. Limited industrial process data is used by a generative model to produce synthetic samples with high similarity and diversity. These high‐quality samples improve fault diagnosis by enriching the imbalanced dataset. Experimental results on two industrial datasets confirm the method's effectiveness in enhancing fault diagnosis performance with limited data. [ABSTRACT FROM AUTHOR]

Viral oncoproteins play crucial roles in transforming normal cells into cancer cells, representing a significant factor in the etiology of various cancers. Traditionally, identifying these oncoproteins is both time-consuming and costly. With advancements in computational biology, bioinformatics tools based on machine learning have emerged as effective methods for predicting biological activities. Here, for the first time, we propose an innovative approach that combines Generative Adversarial Networks (GANs) with supervised learning methods to enhance the accuracy and generalizability of viral oncoprotein prediction. Our methodology evaluated multiple machine learning models, including Random Forest, Multilayer Perceptron, Light Gradient Boosting Machine, eXtreme Gradient Boosting, and Support Vector Machine. In ten-fold cross-validation on our training dataset, the GAN-enhanced Random Forest model demonstrated superior performance metrics: 0.976 accuracy, 0.976 F1 score, 0.977 precision, 0.976 sensitivity, and 1.0 AUC. During independent testing, this model achieved 0.982 accuracy, 0.982 F1 score, 0.982 precision, 0.982 sensitivity, and 1.0 AUC. These results establish our new tool, VirOncoTarget, accessible via a web application. We anticipate that VirOncoTarget will be a valuable resource for researchers, enabling rapid and reliable viral oncoprotein prediction and advancing our understanding of their role in cancer biology. [ABSTRACT FROM AUTHOR]

Our study develops a generative adversarial network (GAN)-based method that generates faithful synthetic image data of human cardiomyocytes at varying stages in their maturation process, as a tool to significantly enhance the classification accuracy of cells and ultimately assist the throughput of computational analysis of cellular structure and functions. Human induced pluripotent stem cell derived cardiomyocytes (hiPSC-CMs) were cultured on micropatterned collagen coated hydrogels of physiological stiffnesses to facilitate maturation and optical measurements were performed for their structural and functional analyses. Control groups were cultured on collagen coated glass well plates. These image recordings were used as the real data to train the GAN model. The results show the GAN approach is able to replicate true features from the real data, and inclusion of such synthetic data significantly improves the classification accuracy compared to usage of only real experimental data that is often limited in scale and diversity. The proposed model outperformed four conventional machine learning algorithms with respect to improved data generalization ability and data classification by incorporating synthetic data. This work demonstrates the importance of integrating synthetic data in situations where there are limited sample sizes and thus, effectively addresses the challenges imposed by data availability. [ABSTRACT FROM AUTHOR]

The floor plan layout of museum exhibition spaces is the skeleton network of the museum, which determines the internal circulation and spatial form of the museum. This paper studies the method and practice of using artificial intelligence technology to assist in the space design of exhibition halls in urban cultural museums. First, it introduces the limitations of traditional space design methods for exhibition halls in urban cultural museums and the superiority and application prospects of the CGAN (conditional generative adversarial network) model in space design. Second, the principle and training process of the CGAN model are explained in detail, and the experimental results and analysis are given. By learning 100 floor plans of exhibition halls of urban culture museums, the CGAN model can generate a new floor plan design for an exhibition hall, which provides a new idea and innovative method for this design task. Finally, the limitations and future research directions of the CGAN model in the space design of urban cultural museum exhibition halls are discussed. The study shows that using the CGAN model to learn the floor plans of exhibition halls of urban cultural museums can effectively improve the innovation and practicability of space design and has the following advantages: (1) It can quickly generate a large number of exhibition hall floor plans, shorten the design cycle, and improve design efficiency. (2) The generated floor plan designs of the exhibition hall are diverse and personalized, meeting the design requirements of different scenarios and needs. (3) The method promotes the deep integration of space design and artificial intelligence technology and provides new possibilities and ideas for space design. These conclusions provide new ideas and methods for the space design of exhibition halls of urban cultural museums and provide a reference and inspiration for space design and intelligent applications in other fields, such as office space design, home decoration space design, landscape space design, and historical arcade and building renovation design. [ABSTRACT FROM AUTHOR]

In image dehazing, non-uniform haze leads to issues such as image blurring, distortion, low contrast, and surface detail loss. Existing dehazing techniques struggle with the uneven distribution of haze, especially in scenes with varying depth, making it difficult to restore high-quality images. This research introduces the Central Difference Embedding Convoluted Generative Adversarial Network (CDECGAN), which enhances local contrast variations to address haze-induced blur and low-contrast areas. To further improve adaptability, the Updated Human Memory Algorithm (UHMA) is integrated to optimise hyperparameters and model performance. The approach also combines image denoising and dehazing, using the Generic Edge Attention (GAE) architecture to denoise while preserving edge sharpness. The Weighted Median Channel Prior (WMCP) model helps focus on high-frequency details and estimate scene depth, solving issues related to fixed-size local patches. Performance metrics confirm the effectiveness of CDECGAN, achieving an average peak signal-to-noise ratio of 25.58 dB, a structural similarity index of 0.884, a natural image quality evaluator score of 11.18, and other notable results. The CDECGAN method outperforms existing techniques by successfully handling non-uniform haze and depth-related challenges in image dehazing. [ABSTRACT FROM AUTHOR]

Background: Recent advancements in generative artificial intelligence (AI) for image generation have presented significant opportunities for medical imaging, offering a promising way to generate realistic virtual medical images while ensuring patient privacy. The generation of a large number of virtual medical images through AI has the potential to augment training datasets for discriminative AI models, particularly in fields with limited data availability, such as neuroimaging. Current studies on generative AI in neuroimaging have mainly focused on disease discrimination; however, its potential for simulating complex phenomena in psychiatric disorders remains unknown. In this study, as examples of a simulation, we aimed to present a novel generative AI model that transforms magnetic resonance imaging (MRI) images of healthy individuals into images that resemble those of patients with schizophrenia (SZ) and explore its application. Methods: We used anonymized public datasets from the Center for Biomedical Research Excellence (SZ, 71 patients; healthy subjects [HSs], 71 patients) and the Autism Brain Imaging Data Exchange (autism spectrum disorder [ASD], 79 subjects; HSs, 105 subjects). We developed a model to transform MRI images of HSs into MRI images of SZ using cycle generative adversarial networks. The efficacy of the transformation was evaluated using voxel-based morphometry to assess the differences in brain region volumes and the accuracy of age prediction pre- and post-transformation. In addition, the model was examined for its applicability in simulating disease comorbidities and disease progression. Results: The model successfully transformed HS images into SZ images and identified brain volume changes consistent with existing case-control studies. We also applied this model to ASD MRI images, where simulations comparing SZ with and without ASD backgrounds highlighted the differences in brain structures due to comorbidities. Furthermore, simulating disease progression while preserving individual characteristics showcased the model's ability to reflect realistic disease trajectories. Discussion: The results suggest that our generative AI model can capture subtle changes in brain structures associated with SZ, providing a novel tool for visualizing brain changes in different diseases. The potential of this model extends beyond clinical diagnosis to advances in the simulation of disease mechanisms, which may ultimately contribute to the refinement of therapeutic strategies. [ABSTRACT FROM AUTHOR]

Traditional dental prosthetics require a significant amount of work, labor, and time. To simplify the process, a method to convert teeth scan images, scanned using an intraoral scanner, into 3D images for design was developed. Furthermore, several studies have used deep learning to automate dental prosthetic processes. Tooth images are required to train deep learning models, but they are difficult to use in research because they contain personal patient information. Therefore, we propose a method for generating virtual tooth images using image-to-image translation (pix2pix) and contextual reconstruction fill (CR-Fill). Various virtual images can be generated using pix2pix, and the images are used as training images for CR-Fill to compare the real image with the virtual image to ensure that the teeth are well-shaped and meaningful. The experimental results demonstrate that the images generated by the proposed method are similar to actual images. In addition, only using virtual images as training data did not perform well; however, using both real and virtual images as training data yielded nearly identical results to using only real images as training data. [ABSTRACT FROM AUTHOR]

ABSTRACT Energy efficiency and secure data transmission are considered as the main design targets of wireless sensor network (WSN). Previous energy‐aware cluster head (CH) selection approaches could not provide secured data transmission. To address this problem, a self‐attention–based cycle‐consistent generative adversarial network (SaCyCsGAN) with honey badger algorithm (HyBA)–adopted energy‐aware routing (EgAR) in WSN is proposed. Initially, the proposed system uses a CH to carry out the routing practice. Accordingly, SaCyCsGAN classifier is exploited for CH selection under firm fitness function: delay, distance, energy, cluster density, and traffic rate. Afterward CH selection, a spiteful node may enter the cluster and acts as a CH. Hence, best path selection is needed. For that, HyBA is utilized, it selects the best path depending upon triplet parameter: trust, connectivity, and service degree. Finally, data are conveying into base station (BS) and vice versa under best path. Finally, the proposed EgAR‐WSN‐SaCyCsGAN‐HyBA method attains 24.13%, 28.57%, 5.88%, and 20.37% higher throughput and 18.50%, 21.67%, 13.33%, and 22.22% lesser energy consumption evaluated to the existing methods such as multiple‐objective CH utilizing self‐attention basis progressive generative adversarial network for safe data aggregation (EgAR‐WSN‐SAPrGAN‐ArVOA), reinforcement learning–based energy‐efficient optimized routing protocol in WSN (EgAR‐WSN‐RLGT‐BCSA), data aggregation including clustering protocol in WSN under machine learning (EgAR‐WSN‐AfNN), and optimum cluster along trusted path for routing formation with categorization of intrusion under machine learning categorization (EgAR‐WSN‐RsBPDT‐HoCSOA). [ABSTRACT FROM AUTHOR]

Vehicle re-identification (Vehicle Re-ID) aims at retrieving and tracking the specified target vehicle with multiple other cameras, which can provide help in checking violations and catching fugitives, but there are still the following problems that need to be solved urgently. First, the existing collected Vehicle Re-ID data often have low resolution and blur in local regions, so that the Vehicle Re-ID algorithm cannot accurately extract subtle feature representations. In addition, small features are easy to cause the disappearance of features under the operation of a large convolution kernel, which makes the model unable to capture and learn subtle features, resulting in inaccurate judgment of vehicles. In this study, we propose a Vehicle Re-ID method based on super resolution and pyramidal convolution residual network. Firstly, a super-resolution image generation network leveraging generative adversarial networks (GANs) is proposed. This network employs both content loss and adversarial loss as optimization criteria, ensuring an efficient transformation from a low-resolution image into a super-resolution counterpart, while meticulously preserving intricate high-frequency details. Then, multi levels of pyramidal convolution operations are designed to generate multi-scale features, which can capture information on different scales. Moreover, the concept of residual learning is applied between the multi levels of pyramidal convolution operations to expedite model optimization and enhance recognition capabilities. Ultimately, the double pyramidal convolutions are meticulously employed on both the original image and the super-resolution image, yielding low-noise feature representations and intricate semantic information respectively. By seamlessly fusing these two diverse sources of information, the resultant combined features exhibit heightened discrimination capabilities and significantly bolster the robustness of image features. In order to verify the effectiveness of the proposed method, extensive experiments are carried out on VeRi-776 and VehicleID datasets. The experimental results show that the method proposed in this paper effectively captures the detail information of vehicle images, accurately distinguishes the subtle differences between different vehicles of the same type, and is superior to state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Accurately predicting the degradation and remaining useful life (RUL) of Proton Exchange Membrane Fuel Cells (PEMFCs) is crucial for enhancing their reliability, particularly in automotive and energy sectors. Traditional models often fail to capture the complex, non-linear degradation patterns of PEMFCs, resulting in suboptimal predictions. This study addresses these technological gaps by introducing a novel approach, the Self-Attention Temporal Dual Discriminator Generative Adversarial Network (SAT-DD-GAN). This model is designed to overcome the limitations of existing methods by integrating advanced Long Short-Term Memory (LSTM) networks for time-series analysis, a self-attention mechanism to focus on critical degradation signals, and dual discriminators to improve prediction accuracy and robustness. By addressing the gap in handling long-term dependencies and complex degradation patterns, the proposed SAT-DD-GAN bridges the disconnect between traditional models and the real-world degradation behavior of PEMFCs. The SAT-DD-GAN achieved groundbreaking predictive performance, with the lowest RMSE (0.983E-3 for static load, 1.012E-3 for dynamic load), MAPE (0.036E-1 and 0.253E-1), MAE (0.776E-3 and 0.806E-3), and the highest R2-score (0.9998 and 0.9983), alongside zero relative error for both datasets. These results demonstrate the model's ability to capture the intricate degradation dynamics more effectively than existing state-of-the-art methods. By filling the technology gap of accurately modeling PEMFC degradation under varying load conditions, this study establishes a new standard for PEMFC prognostics and paves the way for future innovations in optimizing fuel cell performance and longevity. • Introduced a novel SAT-DD-GAN model for accurate prediction of PEMFC stack voltage degradation and Remaining Useful Life (RUL). • Innovatively combined LSTM, self-attention, and dual discriminator mechanisms for enhanced predictive robustness. • Achieved superior accuracy, with RMSE as low as 0.000983 (static load) and R2-score of 0.9998 (static load). • Demonstrated advanced handling of static and dynamic loads, outperforming recent models in key prediction metrics. • Set a new benchmark in predictive maintenance for PEMFCs, offering potential advancements in neural network-driven prognostics. [ABSTRACT FROM AUTHOR]

The editorial in "Frontiers in Marine Science" discusses the advancements in deep learning technology for marine science, focusing on applications such as marine image enhancement, visual recognition, and predictive modeling of marine processes. The research topic covers a wide range of areas, including physical oceanography, underwater image processing, and marine biogeochemical variable prediction. Various innovative deep learning models are proposed to address challenges in marine science, such as enhancing underwater image quality, improving species recognition, and predicting marine phenomena. These advancements contribute to a better understanding of marine environments and processes, offering valuable insights for researchers in the field. [Extracted from the article]

With the development of brain–computer interfaces in recent years, deep learning models have been widely used in EEG-based personal identification. The training of complex models requires a large amount of data, but the acquisition of EEG signals is very time-consuming and energy-consuming. EEG data augmentation can reduce the overhead of EEG data acquisition and provide the data volume required by various complex models, which is conducive to applying EEG identification in practice. The generative adversarial network (GAN) has been applied to augment EEG data features. But the augmentation of temporal EEG signals is less studied. Therefore, this paper proposes a method of temporal EEG data augmentation based on an attention mechanism time series generative adversarial network (ATGAN), which can expand the original EEG dataset to facilitate the training of downstream deep learning classifiers. The experiment is conducted on the BCI Competition IV 2a dataset. The ATGAN model is trained with the temporal EEG signals after preprocessing, and the generated augmentation data by ATGAN is used to expand the original dataset. The accuracy of the EEG-based identification task on the augmented dataset is 7.78% higher than that on the original dataset, which proves the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

The degree of correlation between the sounds received by the ears significantly influences the spatial perception of a sound image. Audio signal decorrelation is, therefore, a commonly used tool in various spatial audio rendering applications. In this paper, we propose a multi-channel extension of a previously proposed decorrelation method based on generative adversarial networks. A separate generator network is employed for each output channel. All generator networks are optimized jointly to obtain a multi-channel output signal with the desired properties. The training objective includes a number of individual loss terms to control both the input-output and the inter-channel correlation as well as the quality of the individual output channels. The proposed approach is trained on music signals and evaluated both objectively and through formal listening tests. Thereby, a comparison with two classical signal processing-based multi-channel decorrelators is performed. Additionally, the influence of the number of output channels, the individual loss term weightings, and the employed training data on the proposed method's performance is investigated. [ABSTRACT FROM AUTHOR]

Underwater simultaneous localization and mapping (SLAM) is essential for effectively navigating and mapping underwater environments; however, traditional SLAM systems have limitations due to restricted vision and the constantly changing conditions of the underwater environment. This study thoroughly examined the underwater SLAM technology, particularly emphasizing the incorporation of deep learning methods to improve performance. We analyzed the advancements made in underwater SLAM algorithms. We explored the principles behind SLAM and deep learning techniques, examining how these methods tackle the specific difficulties encountered in underwater environments. The main contributions of this work are a thorough assessment of the research into the use of deep learning in underwater image processing and perception and a comparison study of standard and deep learning-based SLAM systems. This paper emphasizes specific deep learning techniques, including generative adversarial networks (GANs), convolutional neural networks (CNNs), long short-term memory (LSTM) networks, and other advanced methods to enhance feature extraction, data fusion, scene understanding, etc. This study highlights the potential of deep learning in overcoming the constraints of traditional underwater SLAM methods, providing fresh opportunities for exploration and industrial use. [ABSTRACT FROM AUTHOR]

Tropical cyclones (TC) are dynamic atmospheric phenomena featuring extreme low-pressure systems and powerful winds, known for their devastating impacts on weather and the environment. The main purpose of this paper is to consider the subtle involvement of TCs in the air quality index (AQI), focusing on aspects related to the air quality before, during and after cyclones. This research employs multimodal methods, which include meteorological data and different satellite observations. Deep learning approaches, i.e., ConvLSTM, CNN and Real-ESRGAN models, are combined with a regression model to analyze the temporal variability in the air quality associated with tropical cyclones. Deep learning models are deployed to uncover complex patterns and non-linear interdependencies between cyclones' features and the AQI to give predictive insights into the air quality fluctuations throughout the different stages of tropical cyclones. Furthermore, this study explores the aftermaths of TCs in terms of the air quality with respect to post-cyclone recovery. The findings offer an enhanced view of the role of TCs in the regional or global air quality, which will be useful for policymakers, meteorologists and environmental researchers. Utilizing a CNN for tropical cyclone (TC) classification and the extra trees regressor (ETR) for AQI prediction results in accuracy of 92.02% for the CNN and an R 2 of 83.33% for the ETR. Hence, this work adds to our knowledge and enlightens us on the complex interactions between TCs and the air quality, highlighting wider public health concerns regarding climate adaptation and urban renewal. [ABSTRACT FROM AUTHOR]

In order to solve the problem that existing visible and infrared image fusion methods rely only on the original local or global information representation, which has the problem of edge blurring and non-protrusion of salient targets, this paper proposes a layered fusion method based on channel attention mechanism and improved Generative Adversarial Network (HFCA_GAN). Firstly, the infrared image and visible image are decomposed into a base layer and fine layer, respectively, by a guiding filter. Secondly, the visible light base layer is fused with the infrared image base layer by histogram mapping enhancement to improve the contour effect. Thirdly, the improved GAN algorithm is used to fuse the infrared and visible image refinement layer, and the depth transferable module and guided fusion network are added to enrich the detailed information of the fused image. Finally, the multilayer convolutional fusion network with channel attention mechanism is used to correlate the local information of the layered fusion image, and the final fusion image containing contour gradient information and useful details is obtained. TNO and RoadSence datasets are selected for training and testing. The results show that the proposed algorithm retains the global structure features of multilayer images and has obvious advantages in fusion performance, model generalization and computational efficiency. [ABSTRACT FROM AUTHOR]