Search Results (18,143)

To enhance the accuracy and robustness of lithium-ion battery state-of-health (SOH) prediction, this study proposes a dual-mode intelligent optimization BP neural network model (AO–AVOA–BP) which integrates the Aquila Optimizer (AO) and the African Vulture Optimization Algorithm (AVOA). The model leverages the global search capabilities of AO and the local exploitation strengths of AVOA to achieve efficient and collaborative optimization of network parameters. In terms of feature construction, eight key health indicators are extracted from voltage, current, and temperature signals during the charging phase, and the optimal input set is selected using gray relational analysis. Experimental results demonstrate that the AO–AVOA–BP model significantly outperforms traditional BP and other improved models on both the NASA and CALCE datasets, with MAE, RMSE, and MAPE maintained within 0.0087, 0.0115, and 1.095%, respectively, indicating outstanding prediction accuracy and strong generalization performance. The proposed method demonstrates strong generalization capability and engineering adaptability, providing reliable support for lifetime prediction and safety warning in battery management systems (BMS). Moreover, it shows great potential for wide application in the health management of electric vehicles and energy storage systems. Full article

Objectives: There is a critical need for effective focal therapies for patients with inoperable or anatomically complex tumors where conventional ablation techniques pose high risk or are ineffective. Integrated Nanosecond Pulsed Irreversible Electroporation (INSPIRE) is a novel non-thermal ablation modality which uses real time temperature feedback during pulse delivery to safely treat tumors near critical structures. This study evaluated the impact of exposed electrode length on ablation zone size, reproducibility, and cardiac safety in a large animal model. Methods: INSPIRE treatments were performed in an in vivo healthy porcine liver model. All treatments administered 6000 V 1000 ns pulses with a 45 °C temperature set point. Treatments were administered percutaneously via an electrode and grounding pad approach using an internally cooled electrode applicator. The exposed electrode region at the distal end of the applicator was set to either 0.5, 1.0, 1.5, or 2.0 cm. Ablation zones were assessed via ultrasound, contrast-enhanced CT, and gross pathology one week post-treatment. Cardiac safety was evaluated by measuring pre- and post-treatment serum Troponin levels. Results: All treatments were completed without adverse events. Troponin levels remained stable (pre: 0.249 ng/mL; post: 0.224 ng/mL), indicating no measurable cardiac injury. The 1.5 cm exposure length produced the largest and most consistent ablation volumes, with a mean volume of 12.8 ± 2.6 cm³ and average dimensions of 3.7 × 2.7 cm in under 6 min. Increasing exposure length beyond 1.5 cm introduced greater variability and reduced treatment volumes. Conclusions: INSPIRE enables safe, large-volume, single-applicator ablation without a need for electrical pulse synchronization with R wave in cardiac rhythm. The 1.5 cm exposure length offers optimal balance between energy delivery and treatment consistency. These findings support further clinical investigation of INSPIRE for non-thermal ablation of inoperable tumors. Full article

Background: Increased hip-flexion gait (HFgait) has been shown to promote increased aerobic demands by increasing peak swing-phase hip-flexion angles while walking at comfortable speeds. Biomechanically, HFgait produces a gait pattern similar to walking, while removing the flight phase from running and reducing tibial accelerations. We sought to identify knee joint contact forces between HFgait and common exercise modalities, including running, walking, and cycling, across intensity levels. Methods: Ten healthy participants completed two bouts (low and high intensity) of four different exercises: treadmill running, walking, HFgait, and cycling. Tibiofemoral joint compressive force (TCF) was estimated using a static optimization-based approach. Results: Peak TCF was greater in running compared to HFgait, walking, and cycling; greater in HFgait compared to cycling; and greater in walking compared to cycling. The integral of TCF (iTCF) was greater in running compared to cycling, greater in HFgait compared to running, walking, and cycling, and greater in walking compared to running and cycling. Conclusions: HFgait produced lower knee joint loading than running, comparable joint loading to walking, and greater joint loading than cycling. Thus, HFgait may serve as an exercise modality for populations where joint loading is of particular concern, while achieving aerobic demands similar to running or increased functional demands compared to stationary cycling. Full article

Ground engaging tools (GETs) are critical consumable components on mining excavators, and their timely replacement is essential to prevent risks and excessive downtime. This paper presents a monitoring method utilising the modal properties—natural frequencies and mode shapes. The method is applied in a test case to show how the GETs on an excavator bucket could be monitored. Modal analysis and dynamic analysis are conducted with ANSYS to verify the effectiveness of the proposed method. The finite element analysis models are validated by experimental vibration experiments. The results demonstrate a strong correlation between changes in natural frequencies and the conditions of the teeth on the excavator bucket, when comparing the intact to the worn-out condition. In conclusion, the presented method offers a promising approach for real-time monitoring of the GETs on mining excavators and similar equipment. It will contribute to efficient maintenance interventions and enhancing operational efficiency and safety. Full article

Land use/land cover (LULC) data serve as a critical information source for understanding the complex interactions between human activities and global environmental change. The subtropical karst region, characterized by fragmented terrain, spectral confusion, topographic shadowing, and frequent cloud cover, represents one of the most challenging natural scenes for remote sensing classification. This study reviews the evolution of multi-source data acquisition (optical, SAR, LiDAR, UAV) and preprocessing strategies tailored for subtropical regions. It evaluates the applicability and limitations of various methodological frameworks, ranging from traditional approaches and GEOBIA to machine learning and deep learning. The importance of uncertainty modeling and robust accuracy assessment systems is emphasized. The study identifies four major bottlenecks: scarcity of high-quality samples, lack of scale awareness, poor model generalization, and insufficient integration of geoscientific knowledge. It suggests that future breakthroughs lie in developing remote sensing intelligent models that are driven by few samples, integrate multi-modal data, and possess strong geoscientific interpretability. The findings provide a theoretical reference for LULC information extraction and ecological monitoring in heterogeneous geomorphic regions. Full article

The growing demand for functional bakery products necessitates research on the enrichment of wheat bread with pigweed (Amaranthus spp.) and purslane (Portulaca oleracea) flour. Although these plant-based raw materials offer nutritional and environmental benefits, their inclusion in wheat bread formulations poses challenges in the creation of formulations that may compromise the sensory and structural qualities of the final product. The main objective of this work is to systematically determine the optimal amounts of these alternative flour using multimodal bread characterization techniques that include physicochemical, organoleptic, geometric, and optical evaluations, supported by advanced data reduction techniques and regression models. A total of 70 features were analyzed and reduced to 22 for pigweed flour and 15 for purslane flour informative features. Predictive models (R² = 0.85 for pigweed flour, R² = 0.84 for purslane flour) were developed to optimize the inclusion of alternative flour, resulting in appropriate concentrations of 3.69% for pigweed flour and 7.13% for purslane flour. These formulations balance improved nutritional profiles with acceptable sensory and structural properties. The results obtained not only complement the potential of pigweed and purslane as sustainable functional raw materials but also demonstrate the efficacy of an automated, image-based approach to formulating recipes in food manufacturing. Full article

Background. Severe left ventricular hypertrophy (SLVH) is a high-risk structural cardiac abnormality associated with increased risk of heart failure. It is typically assessed using echocardiography or cardiac magnetic resonance imaging, but these modalities are limited by cost, accessibility, and workflow burden. We introduce a deep learning framework that classifies SLVH directly from chest radiographs, without intermediate anatomical estimation models or demographic inputs. A key contribution of this work lies in interpretability. We quantify how clinically relevant attributes are encoded within internal representations, enabling transparent model evaluation and integration into AI-assisted workflows. Methods. We construct class-balanced subsets from the CheXchoNet dataset with equal numbers of SLVH-positive and negative cases while preserving the original train, validation, and test proportions. ResNet-18 is fine-tuned from ImageNet weights, and a Vision Transformer (ViT) encoder is pretrained via masked autoencoding with a trainable classification head. No anatomical or demographic inputs are used during training. We apply Mutual Information Neural Estimation (MINE) to quantify dependence between learned features and five attributes: age, sex, interventricular septal diameter (IVSDd), posterior wall diameter (LVPWDd), and internal diameter (LVIDd). Results. ViT achieves an AUROC of 0.82 [95% CI: 0.78–0.85] and an AUPRC of 0.80 [95% CI: 0.76–0.85], indicating strong performance in SLVH detection from chest radiographs. MINE reveals clinically coherent attribute encoding in learned features: age > sex > IVSDd > LVPWDd > LVIDd. Conclusions. This study shows that SLVH can be accurately classified from chest radiographs alone. The framework combines diagnostic performance with quantitative interpretability, supporting reliable deployment in triage and decision support. Full article

The diagnosis and monitoring of progressive multiple sclerosis (PMS) require further development of fast and effective clinical tools. Relations between MRI-based brain atrophy measures and cognitive impairment in people with primary progressive and secondary progressive MS (PwPPMS, n = 20 and PwSPMS, n = 19, respectively) were investigated in a prospective study with follow-up after a mean 14.97 ± 4.67 months. MRI analysis showed that at baseline and follow-up in PwSPMS, the left thalamic fraction and corpus callosum fraction were significantly lower than in PwPPMS (baseline: 0.39 ± 0.04 vs. 0.44 ± 0.06, p = 0.0203 and 0.26 ± 0.05 vs. 0.30 ± 0.05, p = 0.0097; respectively and follow-up: 0.40 ± 0.04 vs. 0.44 ± 0.07, p = 0.0443 and 0.25 ± 0.06 vs. 0.30 ± 0.05, p = 0.0103, respectively). In contrast, only at baseline, PwPPMS had a significantly lower cerebellar white matter fraction (CWMF) than PwSPMS (1.83 ± 0.20 vs. 2.01 ± 0.24, p = 0.0132). No other significant differences were observed in the MRI fractions at either study time point or in the changes of the MRI fractions between the PwPPMS and PwSPMS. However, a significant decline in the right putaminal fraction was found during observation in PwSPMS (0.332% ± 0.05% vs. 0.328% ± 0.05%, p = 0.0479). Cognitive test scores and their changes did not differ significantly between the subgroups. Declines in the Brief Visuospatial Memory Test Revised in the whole PMS group (18.74 ± 7.43 vs. 17.03 ± 7.61, p = 0.0209) and in PwPPMS (19.50 ± 8.29 vs. 17.20 ± 7.72, p = 0.0338), as well as in the Brief International Cognitive Assessment for Multiple Sclerosis in PwPPMS (1.05 ± 0.89 vs. 1.25 ± 1.02, p = 0.0421), were observed. In both PwPMS and PwPPMS, a worsening on the Symbol Digit Modalities Test (SDMT) was associated with the reduction of fractions of white matter, cerebellum and right thalamus. SDMT performance also correlated with both gray matter fraction (GMF) and CWMF in the whole group, and with cerebellar gray matter fraction (CGMF) in PwPPMS. In PwSPMS, only Stroop Color and Word Test scores correlated with GMF and CGMF. In conclusion, subtle differences between PwPPMS and PwSPMS were detected both in MRI and neuropsychological parameters. Thus, our results indicate the need for a multicomponent attempt in characterizing progression in different clinical courses of MS. Full article

Background/Objectives: Pemphigus is a rare blistering disease characterized by a chronic course, associated with significant mortality and morbidity. This review article aims to delve into three Pemphigus subtypes: Pemphigus Vulgaris, Pemphigus Foliaceus and Paraneoplastic Pemphigus, including the safety and efficacy of their treatment options. Methods: A thorough literature search was conducted using PubMed, EMBASE, Medline and Cochrane Library to collate data on pharmaceutical treatments of Pemphigus. Studies were selected based on predefined inclusion criteria, which included English language, peer-reviewed articles published in the date range January 2000 to May 2025. Eligible studies involved adults diagnosed with Pemphigus Vulgaris, Pemphigus Foliaceus or Paraneoplastic Pemphigus who were treated with Glucocorticoids, Mycophenolate mofetil, azathioprine or rituximab. The focus was on identifying adverse effects, complete remission and relapse rates. Results: The analysis revealed that glucocorticoid is the first-line treatment for Pemphigus. However, low remission rates of 34% along with steroid-related adverse effects indicate the use of Mycophenolate and azathioprine as steroid-sparing adjuvant therapies. Emerging treatments with rituximab have demonstrated 90% remission rates, indicating promising results with a comparatively mild side effect profile. Conclusions: The findings highlight the importance of ongoing evaluation of treatment modalities for Pemphigus subtypes to optimise remission rates and minimise adverse effects. Ultimately, studies often fail to isolate specific Pemphigus subtypes owing to the scarcity of literature. There is also a crucial need to address the lack of a standardised grading system for the side effects of glucocorticoids and long-term safety data for rituximab in further longitudinal research. Full article

As network infrastructure and Internet of Things (IoT) technologies continue to evolve, immersive systems such as virtual reality (VR) are becoming increasingly integrated into interconnected environments. These advancements allow real-time processing of multi-modal data, improving user experiences with rich visual and three-dimensional interactions. However, ensuring continuous user authentication in VR environments remains a significant challenge. To address this issue, an effective user monitoring system is required to track VR users in real time and trigger re-authentication when necessary. Based on this premise, we propose a multi-modal authentication framework that uses eye-tracking data for authentication, named MobileNetV3pro. The framework applies a transfer learning approach by adapting the MobileNetV3Large architecture (pretrained on ImageNet) as a feature extractor. Its pre-trained convolutional layers are used to obtain high-level image representations, while a custom fully connected classification is added to perform binary classification. Authentication performance is evaluated using Equal Error Rate (EER), accuracy, F1-score, model size, and inference time. Experimental results show that eye-based authentication with MobileNetV3pro achieves a lower EER (3.00%) than baseline models, demonstrating its effectiveness in VR environments. Full article

Vibration control systems are extensively utilized in structures to enhance their resilience against earthquakes and wind forces. However, structures with significant damping exhibit atypical damping behaviors, which impose constraints on the effectiveness of traditional modal analysis methods for discerning modal responses and estimating properties. To surmount this challenge, a novel State-Space-Based Modal Decomposition approach is proposed in this study. The State-Space-Based Modal Decomposition technique adeptly extracts modal responses and identifies modal attributes from acquired data of highly damped structures. The approach accurately calculates damping ratios and natural frequencies by scrutinizing the power spectrum within the deconstructed modal response. The validity of this method is confirmed through a numerical simulation with a three-degree-of-freedom system equipped with oil dampers and experimentation of a structure outfitted with a tuned mass damper system. The findings underscore that the transfer function of the modal response in state-space encompasses both displacement and velocity transfer functions. The results demonstrate that precise estimation of modal parameters can be accomplished by suitably evaluating the participation ratio of the two response components. Full article

Deformable object manipulation (DOM) is a primary bottleneck for the real-world application of autonomous robots, requiring advanced frameworks for sensing, perception, modeling, planning, and control. When fragile objects such as soft tissues or fruits are involved, ensuring safety becomes the paramount concern, fundamentally altering the manipulation problem from one of pure trajectory optimization to one of constrained optimization and real-time adaptive control. Existing DOM methodologies, however, often fall short of addressing fragility constraints as a core design feature, leading to significant gaps in real-time adaptiveness and generalization. This review systematically examines individual components in DOM with a focus on their effectiveness in handling fragile objects. We identified key limitations in current approaches and, based on this analysis, discussed a promising framework that utilizes both low-latency reflexive mechanisms and global optimization to dynamically adapt to specific object instances. Full article

Hands-free computer interaction is a key topic in assistive technology, with camera-based and voice-based systems being the most common methods. Recent camera-based solutions leverage facial expressions or head movements to simulate mouse clicks or key presses, while voice-based systems enable control via speech commands, wake-word detection, and vocal gestures. However, existing systems often suffer from limitations in responsiveness and accuracy, especially under real-world conditions. In this paper, we present 3-Modal Human-Computer Interaction (3M-HCI), a novel interaction system that dynamically integrates facial, vocal, and eye-based inputs through a new signal processing pipeline and a cross-modal coordination mechanism. This approach not only enhances recognition accuracy but also reduces interaction latency. Experimental results demonstrate that 3M-HCI outperforms several recent hands-free interaction solutions in both speed and precision, highlighting its potential as a robust assistive interface. Full article

Indoor positioning systems based on received signal strength (RSS) achieve indoor positioning by leveraging the position-related features inherent in spatial RSS fingerprint images. Their positioning accuracy and robustness are directly influenced by the quality of fingerprint features. However, the inherent spatial low-resolution characteristic of spatial RSS fingerprint images makes it challenging to effectively extract subtle fingerprint features. To address this issue, this paper proposes an RSS-based indoor positioning method that combines enhanced spatial frequency fingerprint representation with fusion learning. First, bicubic interpolation is applied to improve image resolution and reveal finer spatial details. Then, a 2D fast Fourier transform (2D FFT) converts the enhanced spatial images into frequency domain representations to supplement spectral features. These spatial and frequency fingerprints are used as dual-modality inputs for a parallel convolutional neural network (CNN) model with efficient multi-scale attention (EMA) modules. The model extracts modality-specific features and fuses them to generate enriched representations. Each modality—spatial, frequency, and fused—is passed through a dedicated fully connected network to predict 3D coordinates. A coordinate optimization strategy is introduced to select the two most reliable outputs for each axis (x, y, z), and their average is used as the final estimate. Experiments on seven public datasets show that the proposed method significantly improves positioning accuracy, reducing the mean positioning error by up to 47.1% and root mean square error (RMSE) by up to 54.4% compared with traditional and advanced time–frequency methods. Full article

Background: Timely and accurate identification of wound infections is essential for effective management, yet remains clinically challenging. This study evaluated the utility of a near-infrared autofluorescence imaging system (Fluobeam^®, Fluoptics, Grenoble, France) and a thermal imaging system (FLIR^®, Teledyne LLC, Thousand Oaks, CA, USA) for detecting bacterial and fungal infections in chronic wounds. Fluobeam^® enables real-time visualization of microbial autofluorescence without exogenous contrast agents, whereas FLIR^® detects localized thermal changes associated with infection-related inflammation. Methods: This retrospective clinical study included 33 patients with suspected wound infections. All patients underwent autofluorescence imaging using Fluobeam^® and concurrent thermal imaging with FLIR^®. Imaging findings were compared with microbiological culture results, clinical signs of infection, and semi-quantitative microbial burdens. Results: Fluobeam^® achieved a sensitivity of 78.3% and specificity of 80.0% in detecting culture-positive infections. Fluorescence signal intensity correlated strongly with microbial burden (r = 0.76, p < 0.01) and clinical indicators, such as exudate, swelling, and malodor. Pathogens with high metabolic fluorescence, including Pseudomonas aeruginosa and Candida spp., were consistently identified. Representative cases demonstrate the utility of fluorescence imaging in guiding targeted debridement and enhancing intraoperative decision-making. Conclusions: Near-infrared autofluorescence imaging with Fluobeam^® and thermal imaging with FLIR^® offer complementary, noninvasive diagnostic insights into microbial burden and host inflammatory response. The combined use of these modalities may improve infection detection, support clinical decision-making, and enhance wound care outcomes. Full article