Search Results (1,762)

Appliance recognition from aggregate household measurements is challenging under real deployment conditions, where multiple devices operate concurrently and sensor data are affected by imperfections such as noise, missing samples, and nonlinear meter response. In contrast to many studies that rely on curated or idealized datasets, this work investigates appliance recognition using real multimodal utility data (electricity, water, gas) collected at the building entry point, in the presence of substantial uninstrumented background activity. We present a case study evaluating transparent, rule-based detectors designed to exploit characteristic temporal dependencies between modalities while remaining interpretable and robust to sensing imperfections. Four household appliances—washing machine, dishwasher, tumble dryer, and kettle—are analyzed over six weeks of data. The proposed approach achieves reliable detection for structured, water-related appliances (22/30 washing cycles, 19/21 dishwashing cycles, and 23/27 drying cycles), while highlighting the limitations encountered for short, high-power events such as kettle usage. The results illustrate both the potential and the limitations of conservative rule-based detection under realistic conditions and provide a well-documented baseline for future hybrid systems combining interpretable rules with data-driven adaptation. Full article

Wildfires are increasing globally and pose major challenges for assessing post-fire vegetation recovery and ecosystem resilience. We analyzed long-term Landsat time series in two contrasting fire-prone ecosystems in the United States and Australia. Vegetation area was extracted using the Enhanced Vegetation Index (EVI) with Otsu thresholding. Recovery to pre-fire baseline levels was modeled using linear, logistic, locally estimated scatterplot smoothing (LOESS), and generalized additive models (GAM), and their performance was compared using multiple metrics. The results indicated rapid recovery of Australian forests to baseline levels, whereas this was not the case for forests in the United States. Among climatic factors, temperature was the dominant parameter in Australia (Spearman ρ = 0.513, p < 10⁻⁸), while no climatic variable significantly influenced recovery in California. Methodologically, GAM consistently performed best in both regions due to its success in capturing multiphase and heterogeneous recovery patterns, yielding the lowest values of AIC (United States: 142.89; Australia: 46.70) and RMSE_cv (United States: 112.86; Australia: 2.26). Linear and logistic models failed to capture complex recovery dynamics, whereas LOESS was highly sensitive to noise and unstable for long-term prediction. These findings indicate that post-fire recovery is inherently nonlinear and ecosystem-specific and that simple models are insufficient for accurate estimation, with GAM emerging as an appropriate method for assessing vegetation recovery using remote sensing data. This study provides a transferable approach using remote sensing and GAM to monitor forest resilience under accelerating global fire regimes. Full article

In a high-dimensional classification dataset, feature selection is crucial for improving classification performance and computational efficiency by identifying an informative subset of features while reducing noise, redundancy, and overfitting. This study proposes a novel metaheuristic-based ensemble feature selection approach by combining the complementary strengths of Modified Binary Particle Swarm Optimization (MBPSO) and Binary Dynamic Grey Wolf Optimization (BDGWO). The proposed MBPSO–BDGWO ensemble method is specifically designed for high-dimensional classification problems. The performance of the proposed MBPSO–BDGWO ensemble method was rigorously evaluated through an extensive simulation study under multiple high-dimensional scenarios with varying correlation structures. The ensemble method was further validated on several real datasets. Comparative analyses were conducted against single-stage feature selection methods, including BPSO, BGWO, MBPSO, and BDGWO, using evaluation metrics such as accuracy, the F1-score, the true positive rate (TPR), the false positive rate (FPR), the AUC, precision, and the Jaccard stability index. Simulation studies conducted under various dimensionality and correlation scenarios show that the proposed ensemble method achieves a low FPR, a high TPR/Precision/F1/AUC, and strong selection stability, clearly outperforming both classical and advanced single-stage methods, even as dimensionality and collinearity increase. In contrast, single-stage methods typically experience substantial performance degradation in high-correlation and high-dimensional settings, particularly BPSO and BGWO. Moreover, on the real datasets, the ensemble method outperformed all compared single-stage methods and produced consistently low MAD values across repetitions, indicating robustness and stability even in ultra-high-dimensional genomic datasets. Overall, the findings indicate that the proposed ensemble method demonstrates consistent performance across the evaluated scenarios and achieves higher selection stability compared with the single-stage methods. Full article

Surface deformation induced by the extraction of natural resources constitutes a non-stationary spatiotemporal process. Modeling surface deformation time series obtained through Interferometric Synthetic Aperture Radar (InSAR) technology using deep learning methods is crucial for disaster prevention and mitigation. However, the complexity of model hyperparameter configuration and the lack of interpretability in the resulting predictions constrain its engineering applications. To enhance the reliability of model outputs and their decision-making value for engineering applications, this study presents a workflow that combines a Tree-structured Parzen Estimator (TPE)-based Bayesian optimization approach with ensemble inference. Using the Rhineland coalfield in Germany as a case study, we systematically evaluated six deep learning architectures in conjunction with various spatiotemporal coding strategies. Pairwise comparisons were conducted using a Welch t-test to evaluate the performance differences across each architecture under two parameter-tuning approaches. The Benjamini–Hochberg method was applied to control the false discovery rate (FDR) at 0.05 for multiple comparisons. The results indicate that TPE-optimized models demonstrate significantly improved performance compared to their manually tuned counterparts, with the ResNet+Transformer architecture yielding the most favorable outcomes. A comprehensive analysis of the spatial residuals further revealed that TPE optimization not only enhances average accuracy, but also mitigates the model’s prediction bias in fault zones and mineralize areas by improving the spatial distribution structure of errors. Based on this optimal architecture, we combined the ten highest-performing models from the optimization stage to generate a quantile-based susceptibility map, using the ensemble median as the central predictor. Uncertainty was quantified from three complementary perspectives: ensemble spread, class ambiguity, and classification confidence. Our analysis revealed spatial collinearity between physical uncertainty and absolute residuals. This suggests that uncertainty is more closely related to the physical complexity of geological discontinuities and human-disturbed zones, rather than statistical noise. In the analysis of super-threshold probability, the threshold sensitivity exhibited by the mining area reflects the widespread yet moderate impact of mining activities. By contrast, the fault zone continues to exhibit distinct high-probability zones, even under extreme thresholds. It suggests that fault-controlled deformation is more physically intense and poses a greater risk of disaster than mining activities. Finally, we propose an engineering decision strategy that combines uncertainty and residual spatial patterns. This approach transforms statistical diagnostics into actionable, tiered control measures, thereby increasing the practical value of susceptibility mapping in the planning of natural resource extraction. Full article

Accurate segmentation of thyroid nodules in ultrasound images remains challenging due to low contrast, speckle noise, and inter-patient variability that disrupt the inherent spatial symmetry of thyroid anatomy. This study proposes a symmetry-aware SwinUNet framework with integrated spatial attention for thyroid nodule segmentation. The hierarchical window-based Swin Transformer encoder preserves spatial symmetry and scale consistency while capturing both global contextual information and fine-grained local features. Attention modules in the decoder emphasize symmetry consistent anatomical regions and asymmetric nodule boundaries, effectively suppressing irrelevant background responses. The proposed method was evaluated on the publicly available TN3K thyroid ultrasound dataset. Experimental results demonstrate strong performance, achieving a Dice Similarity Coefficient of 85.51%, precision of 87.05%, recall of 89.13%, an IoU of 78.00%, accuracy of 97.02%, and an AUC of 99.02%. Compared with the baseline model, the proposed approach improves the IoU and Dice score by 15.38% and 12.05%, respectively, confirming its ability to capture symmetry-preserving nodule morphology and boundary asymmetry. These findings indicate that the proposed symmetry-aware SwinUNet provides a robust and clinically promising solution for thyroid ultrasound image analysis and computer-aided diagnosis. Full article

Pediatric pneumonia detection faces the challenge of pathological asymmetry, where immature lung tissues present blurred boundaries and lesions exhibit extreme scale variations (e.g., small viral nodules vs. large bacterial consolidations). Conventional detectors often fail to address these imbalances. In this study, we propose YOLO-SMD, a detection framework built upon a symmetrical design philosophy to enforce balanced feature representation. We introduce three architectural innovations: (1) DySample (Content-Aware Upsampling): To address the blurred boundaries of pediatric lesions, this module replaces static interpolation with dynamic point sampling, effectively sharpening edge details that are typically smoothed out by standard upsamplers; (2) SAC2f (Cross-Dimensional Attention): To counteract background interference, this module enforces a symmetrical interaction between spatial and channel dimensions, allowing the model to suppress structural noise (e.g., rib overlaps) in low-contrast X-rays; (3) SDFM (Adaptive Gated Fusion): To resolve the extreme scale disparity, this unit employs a gated mechanism that symmetrically balances deep semantic features (crucial for large bacterial shapes) and shallow textural features (crucial for viral textures). Extensive experiments on a curated subset of 2611 images derived from the Chest X-ray Pneumonia Dataset demonstrate that YOLO-SMD achieves competitive performance with a focus on high sensitivity, attaining a Recall of 86.1% and an mAP@0.5 of 84.3%, thereby outperforming the state-of-the-art YOLOv12n by 2.4% in Recall under identical experimental conditions. The results validate that incorporating symmetry principles into feature modulation significantly enhances detection robustness in primary healthcare settings. Full article

Point cloud semantic segmentation remains challenging under extremely low annotation budgets due to inefficient utilization of sparse labels and sensitivity to data augmentation noise. To address this, we propose a dual-branch consistency learning (DBCL) framework featuring an EMA teacher for semi-supervised point cloud segmentation. Our core innovation lies in a unified consistency regularization scheme that enforces prediction-level alignment via JS divergence and feature-level contrastive learning, while a geometry-aware Laplacian smoothing term preserves local structural consistency. Extensive experiments demonstrate that DBCL achieves 68.56% mIoU on S3DIS with only 0.1% labels, outperforming existing semi-supervised methods and even matching some fully supervised baselines. Full article

Early and accurate diagnosis of lung diseases is essential for effective treatment and patient management. Conventional diagnostic models trained on a single data type often miss important clinical information. This study explored a multimodal deep learning framework that integrates cough sounds, chest radiograph (X-rays), and computed tomography (CT) scans to enhance disease classification performance. Two fusion strategies, early and intermediate fusion, were implemented and evaluated against three single-modality baselines. The dataset was collected from different sources. Each dataset underwent preprocessing steps, including noise removal, grayscale conversion, image cropping, and class balancing, to ensure data quality. Convolutional neural network (CNN) and Extreme Inception (Xception) architectures were used for feature extraction and classification. The results show that multimodal learning achieves superior performance compared with single models. The intermediate fusion model achieved 98% accuracy, while the early fusion model reached 97%. In contrast, single CXR and CT models achieved 94%, and the cough sound model achieved 79%. These results confirm that multimodal integration, particularly intermediate fusion, offers a more reliable framework for automated lung disease diagnosis. Full article

We compared scatter correction (SC) in single-photon emission computed tomography (SPECT) images using effective scatter source estimation (ESSE) and the triple-energy window (TEW) method. We acquired ^99mTc and ¹²³I images of brain, myocardial, and performance phantoms containing rods with different diameters. We assessed contrast ratios (CRs) and ROI-based noise metrics (coefficient of variation, signal-to-noise ratio, and contrast-to-noise ratio [CNR] ). Under ^99mTc, ESSE yielded higher CRs than TEW across all phantoms (mean difference 0.04, range 0.01–0.05) and produced the highest CNR in the myocardial phantom, improving the conspicuousness of the simulated defect. Under ¹²³I, CR differences between ESSE and TEW were small and inconsistent (performance phantom: −0.04 to 0.06; brain phantom: −0.01 to 0.00). A Monte Carlo simulation (point source in air) showed substantial photopeak window penetration for cardiac high-resolution collimators (40.0%) but low penetration for medium-energy general-purpose collimators (5.1%), supporting photopeak contamination as a contributor to the ¹²³I findings and potentially attenuating the apparent advantage of model-based SC that does not explicitly account for penetration photons. These findings suggest that SC selection should consider the radionuclide and imaging target and that ESSE might be a reasonable option for ^99mTc myocardial imaging under the settings examined. Full article

Background/Objectives: Super-resolution deep-learning reconstruction (SR-DLR) is an advanced image reconstruction technique, but its effect on dynamic myocardial computed tomography perfusion (CTP) imaging has not been evaluated. This study aimed to examine the impact of SR-DLR on image quality and perfusion parameters in dynamic myocardial CTP. Methods: Thirty-five patients who underwent dynamic myocardial CTP for coronary artery disease assessment were retrospectively analyzed. Two CTP datasets were reconstructed using hybrid iterative reconstruction (HIR) and SR-DLR. Image quality was compared qualitatively and quantitatively, including image noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and edge rise slope (ERS). Equivalence of CT-derived myocardial blood flow (CT-MBF) between two reconstructions was tested using a previously reported 15% equivalence margin. Intra-patient variability of CT-MBF was evaluated using the robust coefficient of variation (rCV). Results: In the qualitative assessment, SR-DLR had significantly higher scores in contrast (4.0 vs. 2.0) and sharpness (4.5 vs. 2.5) compared with HIR (p < 0.001), while contrast scores were similar. In the quantitative assessment, SR-DLR demonstrated significantly lower image noise (19.4 vs. 29.4 HU), and improved SNR (6.1 vs. 4.1), CNR (13.7 vs. 10.9), and ERS (171.0 vs. 135.1 HU/mm) (all p < 0.001). Mean global CT-MBF was comparable (3.15 ± 0.91 mL/g/min for HIR vs. 3.18 ± 0.97 mL/g/min for SR-DLR) and equivalence was confirmed (p = 0.022). SR-DLR significantly reduced rCV compared with HIR (36.0% vs. 41.0%, p < 0.001). Conclusions: SR-DLR enhances image quality in dynamic myocardial CTP while maintaining mean global CT-MBF and reducing intra-patient variability. Full article

Although fine-tuning pretrained language models has brought remarkable progress to zero-shot named entity recognition (NER), current generative approaches still suffer from inherent limitations. Their autoregressive decoding mechanism requires token-by-token generation, resulting in low inference efficiency, while the massive parameter scale leads to high computational and deployment costs. In contrast, span-based methods avoid autoregressive decoding but often face large candidate spaces and severe noise redundancy, which hinder efficient entity localization in long-text scenarios. To overcome these challenges, we propose an efficient Embedding-based NER framework that achieves an optimal balance between performance and efficiency. Specifically, the framework first introduces a lightweight dynamic feature matching module for coarse-grained entity localization, enabling rapid filtering of potential entity regions. Then, a hierarchical progressive entity filtering mechanism is applied for fine-grained recognition and noise suppression. Experimental results demonstrate that the proposed model, which is trained on a single RTX 5090 GPU for only 24 h, attains approximately 90% of the performance of the SOTA GNER-T5 11B model while using only one-seventh of its parameters. Moreover, by eliminating the redundancy of autoregressive decoding, the proposed framework achieves a 17× faster inference speed compared to GNER-T5 11B and significantly surpasses traditional span-based approaches in efficiency. Full article

Gait recognition methods based on silhouette templates, such as the Gait Energy Image (GEI), achieve high accuracy under controlled conditions but often degrade when appearance varies due to viewpoint, clothing, or carried objects. In contrast, skeleton-based approaches provide interpretable motion cues but remain sensitive to pose-estimation noise. This work proposes two compact 2D skeletal descriptors—Gait Skeleton Images (GSIs)—that encode 3D joint trajectories into line-based and joint-based static templates compatible with standard 2D CNN architectures. A unified processing pipeline is introduced, including skeletal topology normalization, rigid view alignment, orthographic projection, and pixel-level rendering. Core design factors are analyzed on the GRIDDS dataset, where depth-based 3D coordinates provide stable ground truth for evaluating structural choices and rendering parameters. An extensive evaluation is then conducted on the widely used CASIA-B dataset, using 3D coordinates estimated via human pose estimation, to assess robustness under viewpoint, clothing, and carrying covariates. Results show that although GEIs achieve the highest same-view accuracy, GSI variants exhibit reduced degradation under appearance changes and demonstrate greater stability under severe cross-view conditions. These findings indicate that compact skeletal templates can complement appearance-based descriptors and may benefit further from continued advances in 3D human pose estimation. Full article

To enhance imaging quality, singular value decomposition (SVD) has been applied to single-wavelength ghost imaging (GI) or color GI. In this paper, we extend the application of SVD to wavelength-multiplexing ghost imaging (WMGI) for reducing the redundant information in the random measurement matrix corresponding to multi-wavelength modulated speckle fields. The feasibility of this method is demonstrated through numerical simulations and optical experiments. Based on the intensity statistical properties of multi-wavelength speckle fields, we derived an expression for the contrast-to-noise ratio (CNR) to characterize imaging quality and conducted a corresponding analysis. The theoretical results indicate that in SVDWMGI, for the m-wavelength case, the CNR of the reconstructed image is $\sqrt{m}$ times that of single-wavelength GI. Moreover, we carried out an optical experiment with a three-wavelength speckle-modulated light source to verify the method. This approach integrates the advantages of both SVD and wavelength division multiplexing, potentially facilitating the application of GI in long-distance imaging fields such as remote sensing. Full article

This study evaluated the feasibility, image quality, and referral accuracy of a patient-operated optical coherence tomography (OCT) device (SightSync) compared with technician-acquired Heidelberg OCT. This study was conducted in a Veterans Affairs retina clinic (Cleveland, Ohio), resulting in a predominantly male (98%) study population representative of the local veteran demographics. One hundred patients attempted self-administered OCT imaging after brief instruction, yielding 118 successful scans (59% of eyes) with no significant association between scan success and age, visual acuity, or diagnosis. Quantitative analysis of 142 paired images showed that SightSync produced interpretable scans with comparable sharpness to Heidelberg OCT, though signal- and intensity-based metrics (signal-to-noise ratio; SNR, contrast-to-noise ratio; CNR, entropy, pixel intensity; p90) were lower, consistent with hardware differences between a compact patient-operated prototype and a clinical-grade system. Among 121 high-quality SightSync scans, referral decisions demonstrated strong agreement with Heidelberg OCT, with a sensitivity of 83.9%, specificity of 75.6%, and a negative predictive value of 93.2%, indicating reliable exclusion of clinically significant pathology. These findings demonstrate that patients can independently acquire clinically interpretable OCT images and that SightSync provides safe, conservative triage performance—supporting its potential as a scalable community-based retinal imaging solution—while a review of unsuccessful scans has identified prototype modifications expected to further improve device feasibility. Full article

Coded Aperture Snapshot Spectral Imaging (CASSI) is a rapid hyperspectral imaging technique with broad application prospects. Due to limitations in three-dimensional compressed data acquisition modes and hardware constraints, the compressed measurements output by actual CASSI systems have a finite dynamic range, leading to degraded hyperspectral reconstruction quality. To address this issue, a high-quality hyperspectral reconstruction method based on multi-exposure fusion is proposed. A multi-exposure data acquisition strategy is established to capture low-, medium-, and high-exposure low-dynamic-range (LDR) measurements. A multi-exposure fusion-based high-dynamic-range (HDR) CASSI measurement reconstruction network (HCNet) is designed to reconstruct physically consistent HDR measurement images. Unlike traditional HDR networks for visual enhancement, HCNet employs a multiscale feature fusion architecture and combines local–global convolutional joint attention with residual enhancement mechanisms to efficiently fuse complementary information from multiple exposures. This makes it more suitable for CASSI systems, ensuring high-fidelity reconstruction of hyperspectral data in both spatial and spectral dimensions. A multi-exposure fusion CASSI mathematical model is constructed, and a CASSI experimental system is established. Simulation and real-world experimental results demonstrate that the proposed method significantly improves hyperspectral image reconstruction quality compared to traditional single-exposure strategies, exhibiting high robustness against multi-exposure interval jitters and shot noise in practical systems. Leveraging the higher-dynamic-range target information acquired through multiple exposures, especially in HDR scenes, the method enables reconstruction with enhanced contrast in both bright and dark details and also demonstrates higher spectral correlation, validating the enhancement of CASSI reconstruction and effective measurement capability in HDR scenarios. Full article