Search Results (12,630)

Background: This study investigates the impact of anatomically constrained preprocessing and deep learning architecture selection on benign versus malignant breast lesion classification in contrast-enhanced mammography (CEM), with the goal of improving robustness and clinical reliability across heterogeneous data sources. Methods: In this retrospective multicenter study, CEM images from 300 patients (314 lesions) were combined with 1003 publicly available CEM images, yielding a total of 1120 breast cases. Automatic breast segmentation was performed using the LIBRA framework to generate breast-mask images. Eleven deep learning models, including classical convolutional neural networks, attention-based networks, hybrid convolutional neural networks (CNNs), Transformer architectures, and mammography-specific models, were trained and evaluated using both original DICOM images and breast-mask inputs. Performance was assessed using accuracy, balanced accuracy, sensitivity, specificity, AUROC, and AUPRC on cross-validation and independent test sets. Hyperparameter optimization was conducted for the best-performing architecture. Results: Models trained on breast-mask images consistently outperformed those trained on original DICOM images across all architectures and metrics, with AUROC improvements ranging from +0.06 to +0.21. Among all models, ResNet50 trained on breast-mask images achieved the best performance (AUROC = 0.931; AUPRC = 0.933; balanced accuracy = 0.834), further improved after optimization (balanced accuracy = 0.886; sensitivity = 0.842; specificity = 0.930). Classical CNN architectures demonstrated performance comparable to or exceeding that of more complex hybrid CNN–Transformer models when anatomically focused preprocessing and rigorous optimization were applied. Conclusions: Anatomically constrained preprocessing through breast-mask segmentation substantially enhances deep learning performance and stability in CEM-based breast lesion classification. These findings indicate that input representation quality and training optimization are critical determinants of clinically relevant performance, often outweighing architectural complexity, and may support more reliable AI-assisted decision support in CEM workflows. Full article

Industrial maintenance and predictive diagnostics constitute fundamental pillars of modern manufacturing that prevent equipment failures, minimize operational downtime, and optimize maintenance costs across diverse industrial environments. Vibration-based fault classification plays an important role in industrial operations, necessitating highly sophisticated diagnostic methodologies. This research addresses these industrial imperatives through a comprehensive investigation of novel hybrid deep learning architectures for vibration-based fault classification. This study introduces a strategic integration of Quadratic Neural Networks (QNNs), which demonstrate superior non-linear feature extraction capabilities on a vibration signal compared to traditional convolutional approaches. A systematic evaluation of seven sophisticated architectures establishes a clear performance hierarchy, with QuCNN-LSTM-Transformer emerging as the optimal model achieving 99.26% average accuracy. All proposed models demonstrate excellence, with test accuracies consistently surpassing 95% across all evaluated scenarios. The data analyzed is emprical utilizing sensor data collected from an experimental rig and shows exceptional performance consistency on CWRU and HUST datasets. This investigation establishes a new paradigm in intelligent diagnostics, offering functional guidance and definitive analysis of hybrid architectures that advance industrial fault classification applications. Full article

Accurate and robust battery health prognostics are critical for reliable battery management in electronic devices and electric vehicles. Previous studies have demonstrated that combining electrochemical impedance spectroscopy (EIS) with machine learning enables accurate health-state forecasting in LiCoO₂ coin cells. However, the applicability of this EIS-AI paradigm across diverse chemistries and industrial-grade battery formats remains unvalidated, limiting its practical deployment in energy storage systems. Here, we develop an EIS–AI battery prognostic framework and validate its performance on LiNi_1/3Mn_1/3Co_1/3O₂ (NMC111) cylindrical cells and LiNi_0.8Mn_0.1Co_0.1O₂ (NMC811) pouch cells. A hybrid Convolutional Neural Network–Bidirectional Long Short-Term Memory (CNN–BiLSTM) architecture is developed to estimate state of health (SoH) and predict remaining useful life (RUL) from EIS spectra. Trained on an in-house dataset comprising over 13,000 impedance spectra from 22 cells (8 NMC111 and 14 NMC811), the model achieves robust performance, with average coefficients of determination (R²) exceeding 0.92 for SoH estimation and 0.90 for RUL prediction across various batteries and cycling protocols. Salient feature analysis further reveals chemistry- and protocol-dependent frequency regimes associated with degradation. These results demonstrate that impedance spectra constitute physically informative descriptors for data-driven battery prognostics and provide a scalable and interpretable pathway for deploying EIS-AI frameworks in real-world battery management systems (BMSs). Full article

Manual survey data entry remains a bottleneck in large-scale research, marketing, and public policy, where survey sheets are still widely used due to accessibility and high response rates. Despite the progress in Optical Character Recognition (OCR) and Optical Mark Recognition (OMR), existing systems treat these tasks separately and are typically tailored to clean, standardized forms, making them unreliable for real-world survey sheets with diverse markings and handwritten inputs. These limitations hinder automation and introduce significant error rates in data transcription. To address this, we propose SurveyNet, a unified deep learning framework that combines OCR and OMR capabilities to automatically digitize complex survey responses within a single model. SurveyNet processes both handwritten digits and a wide variety of mark types including ticks, circles, and crosses across multiple question formats. We also introduce SurveySet, a novel dataset comprising 135 real-world survey forms annotated across four key response types. Experimental results demonstrate that SurveyNet achieves between 50% and 97% classification accuracy across tasks, with strong performance even on small and imbalanced datasets. This framework offers a scalable solution for streamlining survey digitization workflows, reducing manual errors, and enabling timely analysis in domains ranging from consumer research to public health and education. Full article

Accurately predicting the temperature field of turbine blades is of great significance for evaluating the thermal reliability and service life of high-temperature components in aero-engines. However, due to the high computational cost of numerical simulations and the limitations imposed by complex geometric structures and harsh operating environments, experimental measurements can usually only obtain sparse sensor data, making the acquisition of complete temperature distributions still challenging. Therefore, reconstructing the complete temperature field under sparse measurement conditions has become a key research issue in turbine thermal analysis. To address this problem, this paper proposes an attention-enhanced CNN–LSTM framework for reconstructing transient turbine blade temperature fields from sparse data. The model combines the spatial feature extraction capability of Convolutional Neural Networks (CNNs) with the time-series modeling capability of Long Short-Term Memory networks (LSTM). An SE channel attention module is introduced in the CNN feature extraction stage to achieve adaptive recalibration of channel features, and a temporal attention mechanism is incorporated after the LSTM layer to highlight key transient thermal features. A multi-condition temperature field dataset was constructed by conducting Computational Fluid Dynamics (CFD) simulations on low-pressure turbine guide vanes, and the model was experimentally validated through thermal shock tests. The results show that the proposed model can accurately reconstruct the spatial distribution and transient evolution of the turbine blade temperature field under sparse measurement conditions. Under different operating conditions, the predicted temperature fields are highly consistent with the CFD results, with the maximum Reconstruction error remaining below 19 °C. Error distribution analysis indicates that the model has stable Reconstruction performance and good generalization ability. Full article

Electroencephalogram (EEG) signals serve as the core information carrier for brain–computer interfaces (BCIs); however, their highly non-stationary nature, extremely low signal-to-noise ratio, and significant inter-individual variability pose considerable challenges for signal decoding. Existing deep learning methods struggle to strike a balance between multi-scale, fine-grained feature extraction and efficient long-range temporal modeling. To overcome this limitation, this study proposes a novel deep cascaded architecture, MSResBiMamba, which deeply integrates multi-scale spatiotemporal feature learning with cutting-edge long-sequence modeling techniques. The model first utilizes an enhanced multi-scale spatiotemporal convolutional network (MS-CNN) combined with a SE-channel attention mechanism to adaptively extract local multi-band features and dynamically suppress redundant artefacts. Subsequently, it innovatively introduces an enhanced bidirectional Mamba (Bi-Mamba) module to efficiently capture non-causal long-range temporal dependencies with linear computational complexity, whilst cascading multi-head self-attention mechanisms to establish global higher-order feature interactions. Extensive experiments on the BCI Competition IV-2a dataset demonstrate that MSResBiMamba achieves outstanding classification performance in multi-class motor imagery tasks, significantly outperforming traditional methods and existing state-of-the-art neural networks. Ablation studies and t-SNE visualisations further confirm the model’s robustness in feature decoupling and cross-subject applications, providing a high-precision, high-efficiency decoding solution for BCI systems. Full article

Flexible piezoresistive sensors are often vulnerable to modal ambiguity and bending-induced drift, both of which can obscure true pressure and strain signals under practical operation. Here, we address these limitations by suppressing bending sensitivity at the device level and disambiguating tactile modes at the algorithmic level. We propose and fabricate a Ti₃C₂T_x/graphene oxide (GO) sandwich sensor in which the conductive network is positioned near the neutral axis, thereby ensuring that bending induces negligible axial strain in the active layer. In contrast, out-of-plane pressing enlarges microcontacts, while in-plane stretching disrupts percolation pathways. We develop a composite-beam model to quantify neutral-axis alignment and the resultant bending immunity, realize the device via a straightforward casting process, and systematically characterize its electromechanical response under bending, pressing, nail pressing, and stretching. To further reduce modal ambiguity and improve tactile recognition, a lightweight one-dimensional convolutional neural network (1D-CNN) was introduced to classify temporal resistance signals from the sensor. Experimental results showed that the 1D-CNN achieved a high classification accuracy of 98.52% under flat-state training and testing conditions, and maintained 96.67% accuracy when evaluated on bending-state samples, demonstrating strong robustness against bending-induced interference. Together, the neutral-axis device architecture and the learning-based inference pipeline deliver high sensitivity to pressing and stretching while markedly suppressing the response to bending, thereby enabling wrist-worn pulse monitoring, soft-robotic joint sensing, and plantar pressure insoles. Full article

Deep learning models represent one of the most advanced and effective approaches in predictive modeling. Their hierarchical architectures enable the extraction of complex, non-linear feature relationships and the identification of latent patterns within data, making them highly suitable for tasks involving high-dimensional or unstructured inputs. However, these models are computationally demanding, requiring significant processing resources and time. Furthermore, their predictive performance is largely contingent upon the availability of large-scale datasets. In this study, a Deep Green Framework is employed for the prediction of two computer vision tasks. CIFAR-10 and CIFAR-00 have been taken for image classification. Fifteen convolutional neural network (CNN) variants categorized into light-weight and heavy-weight are trained for the prediction of these two datasets. Based on energy footprint, time, memory usage, Top-1 accuracy, Top-3 accuracy, model size, and model parameters. The study highlights that MobileNetV3-Small produces the best outcomes when compared to other trained models having low task latency and higher efficiency, making it highly suitable for edge environments where resources are scarce. Full article

Clothing classification by occasion is an important area in computer vision and artificial intelligence (AI). This task is particularly challenging because of the subtle visual similarities among clothing categories such as formal, party, and casual attire. Variations in color, fabric, patterns, and lighting further increase the complexity of this task. To address this challenge, we used the Fashionpedia dataset to create a balanced subset of 15,000 images. Specifically, we adopted two different methods for labeling these images: automated classification, which relies on category identifications (IDs) and components, and manual labeling performed by human annotators. We then implemented our preprocessing pipeline, which includes several steps: resizing, image normalization, background removal using segmentation masks, and class balancing. We benchmarked traditional models, including artificial neural networks (ANNs), support vector machines (SVMs), and k-nearest neighbors (KNNs), which use a histogram of oriented gradient (HOG) features, as well as deep learning models such as convolutional neural networks (CNNs), the Visual Geometry Group 16 (VGG16) model utilizing transfer learning, and the vision transformer (ViT) model, all evaluated using identical data splits and preprocessing procedures. The traditional models achieved moderate accuracy, ranging from 54% to 66%. In contrast, the ViT model achieved an accuracy of 81.78% with automated classification and 98.09% with manual labeling. This indicates that a higher label accuracy, along with the preprocessing steps used, significantly enhances the performance. Together, these factors improve the effectiveness of ViT in context-aware apparel classification and establish a reliable baseline for future research. Full article

Convolutional neural networks (CNNs) have achieved remarkable performance in industrial image-based surface defect inspection in recent years. However, many state-of-the-art (SOTA) networks have become increasingly complex and computationally expensive, which limits their deployment in resource-constrained environments or high-throughput, real-time, practical industrial applications. To address this challenge, this paper proposes a novel approach, Light SDI-NAS, to automatically design lightweight CNN architectures for real-time industrial surface defect inspection through neural architecture search (NAS). First, a task-oriented search space for industrial image inspection is constructed by integrating prior knowledge of neural network architecture design with empirical observations. Second, a novel loss function is introduced to balance model accuracy and computational efficiency during the architecture search process. Finally, the lightweight networks generated by Light SDI-NAS demonstrate strong performance on three industrial image datasets. Experimental results show that the proposed models achieve comparable or superior accuracy to manually designed SOTA networks while significantly reducing the number of parameters and improving inference speed by 1.8 times, making them highly suitable for real-time industrial inspection applications. Full article

With increasing port automation and operational intensity, the gearboxes of gantry cranes widely used in bulk cargo terminals are prone to bearing and gear failures under prolonged heavy loads, intense vibrations, and complex operating conditions. Since fault samples often exhibit imbalanced distributions, this imposes two higher requirements on diagnostic methods—first, the ability to effectively address sample imbalance and, second, the capability to simultaneously identify multiple fault categories. To address these challenges, this paper proposes a joint diagnostic method integrating an improved Conditional Wasserstein Generative Adversarial Network with Gradient Penalty (CWGAN-GP) and Multi-Task Learning (MTL). First, the modified CWGAN-GP performs conditional augmentation for minority fault classes, evaluating synthetic sample authenticity and diversity through multiple metrics. Subsequently, a multi-channel diagnostic network is constructed, in which vibration signals are fed into two parallel sub-networks: time–frequency features are extracted from the Short-Time Fourier Transform (STFT)-based time–frequency representations via a residual-block Convolutional Neural Network (CNN), while temporal features are captured from the raw time-domain signal using a Bidirectional Long Short-Term Memory (Bi-LSTM) with an attention mechanism. An attention fusion layer then integrates these two feature types, enabling joint classification of bearings and gears within a multi-task learning framework. Experimental validation on public gearbox datasets and port gantry crane gearbox datasets demonstrates that this method achieves an average diagnostic accuracy exceeding 97%. The proposed method reduces the impact of class imbalance, thereby improving the accuracy and stability of multi-task fault identification. Full article

Monitoring oil pipelines is crucial for effective infrastructure management and maintenance, as it helps prevent threats such as vandalism and leaks that can lead to catastrophic events. Pipeline leaks pose significant environmental and economic risks; however, existing detection methods are often expensive, slow, or unreliable, limiting their effectiveness for real-time applications. This study proposes a lightweight thermal-imaging-based intelligent leak detection system that integrates Convolutional Neural Networks (CNN), Autoencoder (AE), and Knowledge Distillation (KD), suitable for deployment on edge devices. The proposed system addresses challenges associated with existing pipeline detection techniques, including large model sizes, high transmission latency, and excessive energy consumption. Thermal images of pipelines are captured and compressed using an autoencoder before being processed by a CNN model optimized through knowledge distillation. The model was trained and tested on a locally collected thermal image dataset and designed for deployment on edge devices such as Raspberry Pi to simulate edge computing scenarios. Experimental results demonstrate that the proposed CNN + KD + AE model achieved 98% accuracy, 98% precision, 98% recall, and an F1-score of 98%, outperforming baseline models such as MobileNetV2 (91%), InceptionV3 (84%), EfficientNet-Lite (81%), and ResNet (74%). Furthermore, the number of trainable parameters was significantly reduced to 1.18 million, with a compact model size of 4.51 MB. These findings confirm the system’s suitability for real-time leak detection in remote and resource-constrained environments, contributing to the development of cost-effective, scalable, and energy-efficient solutions for intelligent pipeline monitoring. Full article

The integration of micro-nano optical structures has become an essential strategy for overcoming the performance bottlenecks of quantum well infrared photodetectors (QWIPs), specifically by addressing the inherent inability of planar devices to couple with normally incident light due to intersubband transition selection rules. A critical factor in this integration is the precise spectral overlap between an optical mode and the material’s excitation mode. Therefore, achieving precise spectral engineering is indispensable. However, conventional electromagnetic simulations act as forward solvers, calculating optical responses based on given geometric parameters. They cannot directly perform inverse design, which involves deriving optimal geometric parameters directly from a desired optical response. Consequently, structural optimization is severely constrained by time-consuming trial-and-error iterations, which often struggle to find the global optimum in a complex design space. To overcome these limitations, this paper presents a comprehensive theoretical and numerical study proposing a deep learning framework for QWIPs coupled with all-dielectric micropillar structures. By establishing a structure-absorption spectrum dataset via finite difference time domain (FDTD) simulations, we developed a dual-network setup. For the forward prediction, a multilayer perceptron (MLP) maps geometric parameters (side length a and period p) to the absorption spectrum, achieving a computational speedup of seven orders of magnitude over traditional numerical simulations. Concurrently, a convolutional neural network (CNN) is employed for the inverse design, realizing on-demand design of geometric parameters based on target spectra with high reconstruction accuracy. Furthermore, the selected all-dielectric micropillar structures are highly compatible with mainstream semiconductor fabrication processes. This research provides an efficient, automated toolkit for the development of high-performance infrared photodetectors. Full article

This study investigated the impact of image capture distance on the performance of convolutional neural networks (CNNs) in classifying fabrics. Unlike previous works that rely solely on digital zoom and data augmentation to simulate multi-scale variations, this research explores the use of physically captured images at far, mid-range, and near focal lengths using a camera with an attached varifocal lens. Fabric samples from three categories of Cotton, Linen, and Silk were imaged under consistent lighting to create an image dataset with a total of 1350 images used to train CNN models via transfer learning, with MobileNetV2 and ResNet50 as the baseline architectures. Classification performance was evaluated separately on each focal subset and on their combined dataset to test the trained model generalization capability. Results showed an absolute accuracy gain of 20.57% with MobileNetV2 and 9.78% for ResNet50 while performing with an improved accuracy at 98.42% for MobileNetV2 and ResNet50 at 96.30% Full article

Background: Laryngeal cancer (LC), a common subtype of head and neck cancers (HNC), is most frequently represented by laryngeal squamous cell carcinoma (LSCC). Prognosis largely depends on early detection; however, traditional prognostic tools, including tumor-node-metastasis (TNM) staging, often show limited predictive accuracy. Artificial intelligence (AI), including machine learning (ML), natural language processing, and deep learning (DL), has emerged as a promising approach to improving cancer diagnosis, prognosis, and treatment planning by analyzing clinical data and medical imaging. Objective: This systematic review assesses the role of AI in prognosis, recurrence prediction, and treatment outcomes in LC. Methods: PubMed, MEDLINE, Scopus, Web of Science, IEEE Xplore, and ScienceDirect were searched up to January 2025. A total of 1062 records were identified; after title/abstract screening and full-text assessment, 29 studies were included. Eligible studies involved adult patients with LC and applied AI to diagnose, prognose, predict recurrence, or assess treatment outcomes using human datasets. Study quality and risk of bias were evaluated using the QUADAS-2 and QUIPS. Results: The 29 included studies were mostly retrospective, with sample sizes ranging from 10 to 63,000 patients. Most focused on LSCC, with a higher prevalence in males. The studies utilized various AI techniques, including deep learning models such as convolutional neural networks (CNNs) and DeepSurv, as well as ML algorithms like random survival forest, gradient boosting machines, random forest, k-nearest neighbors, naïve Bayes, and decision trees. AI models demonstrated strong prognostic performance, surpassing Cox regression and TNM staging in predicting survival and recurrence. Several studies reported outcomes related to treatment, such as chemotherapy response, occult lymph node metastasis, and the need for salvage surgery. Methodological quality varied, with biases related to patient selection and confounding factors. Conclusions: AI has the potential to improve prognosis estimation, recurrence prediction, and treatment outcome assessment in LC. However, although AI can be a helpful addition to clinical decision-making, more prospective studies, external validation, and standardized evaluation are necessary before these technologies can be confidently adopted in everyday clinical practice. Full article