Physical priors accelerate neural network training for imaging through multimode fiber

Rapid prediction of cellulose, hemicellulose, and lignin content in biomass using FTIR based on transfer learning

Optimizing Time-Series forecasting using stacked deep learning framework with enhanced adaptive moment estimation and error correction

Single‐fiber imaging technology offers unique insights for research and inspection in difficult‐to‐reach and narrow spaces. In particular, ultra‐compact multimode fiber (MMF) imaging, has received increasing interest over the past decade. However, MMF imaging will be seriously distorted when subjected to dynamic perturbations due to time‐varying mode coupling, and the imaging of space objects via Gaussian beam will be relatively degraded at the edge due to insufficient contrast. Here, a robust super‐resolution image transmission method based on a ring core fiber (RCF) with orbital angular momentum (OAM) is proposed and experimentally demonstrated. The OAM modes propagating in the RCF form a series of weakly coupled mode groups, making the imaging system robust to external perturbations. In addition, a spiral phase plate is used as a vortex filter to produce OAM for edge enhancement, thus improving the image resolution. Furthermore, a few‐shot U‐Transformer neural network is proposed to enhance the resilience of the developed RCF‐OAM imaging system against environmental perturbations. Finally, the developed RCF‐OAM imaging system achieves biological image transmission, demonstrating the practicality of the scheme. This pioneering RCF OAM image transmission system may have broad applications, potentially revolutionizing fields such as biological imaging and industrial non‐destructive testing.

This paper pioneers a novel approach in electromagnetic (EM) system analysis by synergistically combining Bayesian Neural Networks (BNNs) informed by Latin Hypercube Sampling (LHS) with advanced thermal-mechanical surrogate modeling within COMSOL simulations for high-frequency low-pass filter modeling. Our methodology transcends traditional EM characterization by integrating physical dimension variability, thermal effects, mechanical deformation, and real-world operational conditions, thereby achieving a significant leap in predictive modeling fidelity. Through rigorous evaluation using Mean Squared Error (MSE), Maximum Learning Error (MLE), and Maximum Test Error (MTE) metrics, as well as comprehensive validation on unseen data, the model's robustness and generalization capability is demonstrated. This research challenges conventional methods, offering a nuanced understanding of multiphysical phenomena to enhance reliability and resilience in electronic component design and optimization. The integration of thermal variables alongside dimensional parameters marks a novel paradigm in filter performance analysis, significantly improving simulation accuracy. Our findings not only contribute to the body of knowledge in EM diagnostics and complex-environment analysis but also pave the way for future investigations into the fusion of machine learning with computational physics, promising transformative impacts across various applications, from telecommunications to medical devices.

PreAIS: Prediction of A-to-I editing sites based on DNN-CNN deep learning models.

Aiming at the problem that the boiler thermal efficiency can't be on-line calculated in the field, data collected are divided into training data and testing data by researching on the 1000MW unit of Sanbaimen Power Plant in Datang Chaozhou, and then prediction model for boiler thermal efficiency is established by applying support vector regression and particle swarm optimization. This method uses optimization function of particle swarm algorithm to optimize the parameters C and g of the prediction model, and applys test data and the randon data to test the model's accuracy and generalization capability. The simulation results show that prediction model for boiler thermal efficiency has higher pre- diction accuracy that relative error is controlled within 1% and has better generalization ability, which provides feasible scheme for online measurement of boiler thermal efficiency.

Apparent diffusion coefficient (ADC) maps derived from diffusion weighted magnetic resonance imaging (DWI MRI) provides functional measurements about the water molecules in tissues. However, DWI is time consuming and very susceptible to image artifacts, leading to inaccurate ADC measurements. This study aims to develop a deep learning framework to synthesize ADC maps from multi-parametric MR images. We proposed the multiparametric residual vision transformer model (MPR-ViT) that leverages the long-range context of vision transformer (ViT) layers along with the precision of convolutional operators. Residual blocks throughout the network significantly increasing the representational power of the model. The MPR-ViT model was applied to T1w and T2-fluid attenuated inversion recovery images of 501 glioma cases from a publicly available dataset including preprocessed ADC maps. Selected patients were divided into training (N=400), validation (N=50), and test (N=51) sets, respectively. Using the preprocessed ADC maps as ground truth, model performance was evaluated and compared against the Vision Convolutional Transformer (VCT) and residual vision transformer (ResViT) models with the peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and mean squared error (MSE). The results are as follows using T1w + T2-FLAIR MRI as inputs: MPR-ViT-PSNR: 31.0±2.1, MSE: 0.009±0.0005, SSIM: 0.950±0.015. In addition, ablation studies showed the relative impact on performance of each input sequence. Both qualitative and quantitative results indicate that the proposed MR-ViT model performs favorably against the ground truth data. We show that high-quality ADC maps can be synthesized from structural MRI using a MPR-ViT model. Our predicted images show better conformality to the ground truth volume than ResViT and VCT predictions. These high-quality synthetic ADC maps would be particularly useful for disease diagnosis and intervention, especially when ADC maps have artifacts or are unavailable.

Predicting distribution coefficient and effective diffusion coefficient of radionuclides in bentonite: Multi-output neural network simulation and diffusion experimental study.

Machine learning has demonstrated significant application potential in rolling bearing fault detection. However, traditional deep learning approaches typically depend on large-scale labeled datasets and are limited to identifying a single fault type, which poses challenges in meeting the complex fault diagnosis in real-world industrial scenarios, particularly under small sample conditions. To address these issues, this study proposes a multi-task attention fault diagnosis framework, convolutional attention neural network by neural discriminant dimension reduction (NDDR-CANN). The model employs multiple task-specific neural networks to extract distinct features such as fault type and severity, respectively, and achieves adaptive feature sharing across tasks by NDDR, thereby facilitating the transfer and complementarity of task-related knowledge and enhancing the model's generalization capability and predictive accuracy. Furthermore, by integrating a convolutional block attention module, the model can focus on salient features across channel and spatial dimensions, thereby improving feature discrimination. Experimental validation on the CWRU bearing dataset demonstrates that the proposed model exhibits robust effectiveness and adaptability under various fault conditions and small sample scenarios. Comparative experimental results further confirm that NDDR-CANN outperforms current mainstream few-shot fault diagnosis methodologies.

This study investigates the effectiveness of an Artificial Neural Network (ANN) model in enhancing the classification of dry beans varieties. The dry beans dataset, comprising 13,611 samples and 16 numerical features, was preprocessed using Z-score normalization for outlier removal and Principal Component Analysis (PCA) to remove 702 outliers and reduce dimensionality to 10 components while preserving 99.99% of the original variance. A feedforward ANN architecture was designed, constructed as a feedforward neural network with three fully connected layers namely, an input layer that receives 10 features (as reduced by PCA), a first hidden layer with 16 neurons and ReLU activation, a second hidden layer with 32 neurons and ReLU activation, and an output layer with neurons equal to the number of classes, using softmax activation for multi-class probability prediction. The model was trained over 60 epochs using TensorFlow/Keras. Evaluation on a held-out test set demonstrated that the ANN model achieved a classification accuracy of 94.50%. Performance metrics, including accuracy, precision, recall, F1-score, and confusion matrix analysis, confirmed the model's robustness and generalization capability across all dry beans’ classes. The findings affirm that ANN models, when supported by appropriate preprocessing, can serve as scalable and accurate alternatives to conventional machine learning approaches in agricultural classification tasks.

In the medical research domain, limited data and high annotation costs have made efficient classification under few-shot conditions a popular research area. This paper proposes a meta-learning framework, termed MedOptNet, for few-shot medical image classification. The framework enables the use of various high-performance convex optimization models as classifiers, such as multi-class kernel support vector machines, ridge regression, and other models. End-to-end training is then implemented using dual problems and differentiation in the paper. Additionally, various regularization techniques are employed to enhance the model's generalization capabilities. Experiments on the BreakHis, ISIC2018, and Pap smear medical few-shot datasets demonstrate that the MedOptNet framework outperforms benchmark models. Moreover, the model training time is also compared to prove its effectiveness in the paper, and an ablation study is conducted to validate the effectiveness of each module.

P-glycoprotein (P-gp) is a transmembrane protein widely involved in the absorption, distribution, metabolism, excretion, and toxicity (ADMET) of drugs within the human body. Accurate prediction of P-gp inhibitors and substrates is crucial for drug discovery and toxicological assessment. However, existing models rely on limited molecular information, leading to suboptimal model performance for predicting P-gp inhibitors and substrates. To overcome this challenge, we compiled an extensive dataset from public databases and literature, consisting of 5,943 P-gp inhibitors and 4,018 substrates, notable for their high quantity, quality, and structural uniqueness. In addition, we curated two external test sets to validate the model's generalization capability. Subsequently, we developed a multimodal graph contrastive learning (GCL) model for the prediction of P-gp inhibitors and substrates (MC-PGP). This framework integrates three types of features from Simplified Molecular Input Line Entry System (SMILES) sequences, molecular fingerprints, and molecular graphs using an attention-based fusion strategy to generate a unified molecular representation. Furthermore, we employed a GCL approach to enhance structural representations by aligning local and global structures. Extensive experimental results highlight the superior performance of MC-PGP, which achieves improvements in the area under the curve of receiver operating characteristic (AUC-ROC) of 9.82% and 10.62% on the external P-gp inhibitor and external P-gp substrate datasets, respectively, compared with 12 state-of-the-art methods. Furthermore, the interpretability analysis of all three molecular feature types offers comprehensive and complementary insights, demonstrating that MC-PGP effectively identifies key functional groups involved in P-gp interactions. These chemically intuitive insights provide valuable guidance for the design and optimization of drug candidates.

Multimode fibers provide a unique opportunity for exploring the spatial degrees of freedom for high throughput light transmission. However, the modal dispersion prevents from the straightforward application of multimode fibers for space division multiplexing, such as image transmission. Herein, we propose and experimentally demonstrate a deep neural network termed multimode fiber inverse-scattering net for overcoming the modal dispersion induced scrambling in multimode fibers. Such a network is capable of transmitting grayscale image through the multimode fiber with high fidelity. 256-level grayscale images with 128 × 128 spatial channels encoded in the input wavefront can be retrieved from the output optical speckle patterns, where the average Pearson correlation coefficient and structural similarity index are as large as 0.97 and 0.95, respectively. Our results demonstrate that the proposed deep neural network has an excellent ability for learning the relationship between the input and output optical fields of a multimode fiber, which might facilitate the realization of high throughput space division multiplexing through multimode fibers.

Reconstructing as‐built building information modeling from point clouds in underground garages holds great potential for enhancing facility management. However, this task is particularly challenging due to the complex and diverse architectural and mechanical, electrical, and plumbing (MEP) components present in such environments. Moreover, there is no publicly available dataset and no dedicated semantic segmentation algorithm for underground garages environments. To address this gap, this paper first builds an expert‐labeled underground garage dataset, which includes 10 categories of components. Subsequently, this paper introduces GarageNet, a deep learning (DL) network designed for semantic segmentation of point clouds for scenes including both building and MEP components. GarageNet integrates a residual squeeze‐and‐excitation module to enhance layer and channel connection and a multi‐head attention feature aggregation module to capture local geometric structures. The results indicate that the GarageNet achieved a high overall accuracy of 95.09% and a mean intersection over union of 87.15%, surpassing other DL algorithms such as PointNext. Ablation studies and cross‐validation further demonstrate the model's reliability and generalization capability.

This study introduces significant improvements in the construction of deep convolutional neural network models for classifying agricultural products, specifically oranges, based on their shape, size, and color. Utilizing the MobileNetV2 architecture, this research leverages its efficiency and lightweight nature, making it suitable for mobile and embedded applications. Key techniques such as depthwise separable convolutions, linear bottlenecks, and inverted residuals help reduce the number of parameters and computational load while maintaining high performance in feature extraction. Additionally, the study employs comprehensive data augmentation methods, including horizontal and vertical flips, grayscale transformations, hue adjustments, brightness adjustments, and noise addition to enhance the model's robustness and generalization capabilities. The proposed model demonstrates superior performance, achieving an overall accuracy of 99.53%∼100% with nearly perfect precision, recall of 95.7%, and F1-score of 94.6% for both “orange_good” and “orange_bad” classes, significantly outperforming previous models which typically achieved accuracies between 70% and 90%. While the classification performance was near-perfect in some aspects, there were minor errors in specific detection tasks. The confusion matrix shows that the model has high sensitivity and specificity, with very few misclassifications. Finally, this study highlights the practical applicability of the proposed model, particularly its easy deployment on resource-constrained devices and its effectiveness in agricultural product quality control processes. These findings affirm the model in this research as a reliable and highly efficient tool for agricultural product classification, surpassing the capabilities of traditional models in this field.