Search Results (108,360)

Ophthalmic tacrolimus compounded formulations are usually made from the commercial intravenous presentation, which contains ethanol as a solubilizer due to the low solubility of tacrolimus. The use of cyclodextrins is presented as an alternative to ethanol, an ocular irritant excipient, to avoid its long-term irritant effects. Open-label, sequential, prospective study to compare effectiveness, safety, and adherence of a new formulation of 0.015% tacrolimus with cyclodextrins (TCD) versus 0.03% tacrolimus with ethanol (TE). The ocular evaluation was assessed by ocular signs, corneal staining, subjective questionnaires as Visual Function Questionnaire (VFQ-25) and Visual Analogue Scale (VAS) of symptoms, lacrimal stability, ocular redness, and intraocular pressure. Compliance was assessed by VAS of adherence and empirically (difference between theoretical and actual consumption). Clinical ocular signs and corneal staining score remained stable for most patients 3 months after switching formulations. The TCD formulation did not modify the tear stability and intraocular pressure of the treated patients compared to the TE formulation. TCD eye drops significantly decreased the subjective pain values on VFQ-25 scale and burning sensation on the VAS symptom scale in comparison to TE formulation after 3 months after the change to TCD formulation. The novel tacrolimus in cyclodextrins formulation is a promising alternative for treating inflammatory ocular pathologies refractory to first-line treatments. Full article

Squamous cell carcinoma of the head and neck (SCCHN) is among the ten most common cancers worldwide, with advanced SCCHN presenting with a 5-year survival of 34% in the case of nodal involvement and 8% in the case of metastatic disease. Disease-free survival at 2 years is 67% for stage II and 33% for stage III tumors, whereas 12–30% of patients undergo distant failures after curative treatment. Previous treatments often hinder the success of salvage surgery and/or reirradiation, while the standard of care for the majority of metastatic SCCHN remains palliative chemo- and immuno-therapy, with few patients eligible for locoregional treatments. The aim of this paper is to review the characteristics of recurrent SCCHN, based on different recurrence sites, and metastatic disease; we will also explore the possibilities not only of salvage surgery and reirradiation but also systemic therapy choices and locoregional treatment for metastatic SCCHN. Full article

Adeno-associated viral vectors (AAVs) are the predominant viral vectors used for gene therapy applications. A significant challenge in obtaining effective doses is removing non-therapeutic empty viral capsids lacking DNA cargo. Current methods for separating full (gene-containing) and empty capsids are challenging to scale, produce low product yields, are slow, and are difficult to operationalize for continuous biomanufacturing. This communication demonstrates the feasibility of separating full and empty capsids by ultrafiltration. Separation performance was quantified by measuring the sieving coefficients for full and empty capsids using ELISA, qPCR, and an infectivity assay based on the live cell imaging of green fluorescent protein expression. We demonstrated that polycarbonate track-etched membranes with a pore size of 30 nm selectively permeated empty capsids to full capsids, with a high recovery yield (89%) for full capsids. The average sieving coefficients of full and empty capsids obtained through ELISA/qPCR were calculated as 0.25 and 0.49, indicating that empty capsids were about twice as permeable as full capsids. Establishing ultrafiltration as a viable unit operation for separating full and empty AAV capsids has implications for developing the scale-free continuous purification of AAVs. Full article

Inter-Turn Short Circuit (ITSC) faults do not necessarily produce high temperatures but exhibit distinct heat distribution and characteristics. This paper proposes a novel fault diagnosis and identification scheme utilizing an improved You Look Only Once Vision 8 (YOLOv8) algorithm, enhanced with an infrared image slicing-aided hyper-inference (SAHI) technique, to automatically detect ITSC fault trajectories in dry-type transformers. The infrared image acquisition system gathers data on ITSC fault trajectories and captures images with varying contrast to enhance the robustness of the recognition model. Given that the fault trajectory constitutes a small portion of the overall infrared image and is subject to significant background interference, traditional recognition algorithms often misjudge or omit faults. To address this, a YOLOv8-based visual detection method incorporating Dynamic Snake Convolution (DSConv) and the Slicing-Aided Hyper-Inference algorithm is proposed. This method aims to improve recognition precision and accuracy for small targets in complex backgrounds, facilitating accurate detection of ITSC faults in dry-type transformers. Comparative tests with the YOLOv8 model, Fast Region-based Convolutional Neural Networks (Fast-RCNNs), and Residual Neural Networks (Retina-Nets) demonstrate that the enhancements significantly improve model convergence speed and fault trajectory detection accuracy. The approach offers valuable insights for advancing infrared image diagnostic technology in electrical power equipment. Full article

Ultrafast lasers, characterized by femtosecond and picosecond pulse durations, have revolutionized material processing due to their high energy density and minimal thermal diffusion, and have played a transformative role in precision manufacturing. This review first traces the progression from early ruby lasers to modern titanium–sapphire lasers, highlighting breakthroughs like Kerr-lens mode-locking and chirped pulse amplification. It also examines the interaction mechanisms between ultrafast pulses and various materials, including metals, dielectrics, and semiconductors. Applications of ultrafast lasers in microstructure processing techniques are detailed, such as drilling, cutting, surface ablation, and nano welding, demonstrating the versatility and precision of the technology. Additionally, it covers femtosecond laser direct writing for optical waveguides and the significant advancements in imaging and precision measurement. This review concludes by discussing potential future advancements and industrial applications of ultrafast lasers. Full article

Large language models (LLMs) are widely integrated into autonomous driving systems to enhance their operational intelligence and responsiveness and improve self-driving vehicles’ overall performance. Despite these advances, LLMs still struggle between hallucinations—when models either misinterpret the environment or generate imaginary parts for downstream use cases—and taxing computational overhead that relegates their performance to strictly non-real-time operations. These are essential problems to solve to make autonomous driving as safe and efficient as possible. This work is thus focused on symmetrical trade-offs between the reduction of hallucination and optimization, leading to a framework for these two combined and at least specifically motivated by these limitations. This framework intends to generate a symmetry of mapping between real and virtual worlds. It helps in minimizing hallucinations and optimizing computational resource consumption reasonably. In autonomous driving tasks, we use multimodal LLMs that combine an image-encoding Visual Transformer (ViT) and a decoding GPT-2 with responses generated by the powerful new sequence generator from OpenAI known as GPT4. Our hallucination reduction and optimization framework leverages iterative refinement loops, RLHF—reinforcement learning from human feedback (RLHF)—along with symmetric performance metrics, e.g., BLEU, ROUGE, and CIDEr similarity scores between machine-generated answers specific to other human reference answers. This ensures that improvements in model accuracy are not overused to the detriment of increased computational overhead. Experimental results show a twofold improvement in decision-maker error rate and processing efficiency, resulting in an overall decrease of 30% for the model and a 25% improvement in processing efficiency across diverse driving scenarios. Not only does this symmetrical approach reduce hallucination, but it also better aligns the virtual and real-world representations. Full article

The influence of fiber orientation on the mechanical behavior of a polymer matrix composite reinforced with natural jute fibers is investigated in this study. Two fiber orientation configurations are examined: the first involves woven fibers aligned in the direction of testing, while the second considers a 45° orientation. The research involves manufacturing composite plates using jute fabric with the mentioned orientations, followed by cutting rectangular specimens for tensile testing to determine which orientation yields superior properties. Displacement fields are measured using a digital image correlation technique, synchronized with load data obtained from a universal testing machine equipped with a load cell to obtain stress–strain curves for each configuration. Results indicate that 0° specimens achieve higher stress but lower strain compared to 45° specimens. This research contributes to understanding the optimal fiber alignment for enhancing the mechanical performance of fiber-reinforced polymer composites. Full article

Intervertebral disc (IVD) degeneration is one of the primary causes of low back pain, causing disability; hence, there is no regenerative nature of the current treatments. Hyaluronic acid (HA) was reported to facilitate tissue repair and alleviate pain. Herein, we determined the therapeutic effect of HA and type II collagen (COLII) hydrogel for tissue repair targeting pain in IVD degeneration. We implanted HA/COLII hydrogel following surgically induced disc injury at coccygeal levels in the rat tail model of pain. We assessed the efficacy of the HA/COLII hydrogel in reducing pain behaviour by using the von Frey assessment, protein expression of growth-associated protein (GAP) 43 for sensory nerve innervation, and disc hydration by magnetic resonance imaging (MRI). We observed the anti-nociceptive effect of the HA/COLII hydrogel in alleviating mechanical allodynia in rats. There was an inhibition of sensory hyperinnervation indicated by the GAP43 protein in the treatment group. We revealed an increase in T1ρ mapping of MRI, indicating that the hydrogel restored disc hydration in vivo. Our findings suggest the HA/COLII hydrogel alleviates pain behaviour, inhibits hyperinnervation and promotes disc hydration for tissue repair, implying that it is a potential candidate for the treatment of degenerative disc-associated low back pain. Full article

The development of methods to detect plant stress is not only a scientific challenge, but is also of great importance for agriculture and forestry. However, at present, stress diagnostics based on plant spectral characteristics has several limitations: (1) the high dependence of stress assessment on plant species identity; (2) the poor differentiation of different types of stress; and (3) the difficulty of detecting stress before visible symptoms appear. In this regard, the development of plant spectral metrics represents a significant area of research. Ficus elastica plants were exposed under the photosynthetic photon flux density (PPFD) from 0 to 1200 μmol photons m⁻²s⁻¹. Exposure of F. elastica leaves to excess light (EL) (≥400 μmol photons m⁻²s⁻¹) resulted in an increase in reflectance in the yellow-green region (522–594 nm) and a decrease in reflectance in the red region (666–682 nm) of the spectrum, accompanied by a shift of the red edge point toward the longer wavelength. These changes were revealed using the previously proposed light stress index (LSI = mean(R_666:682)/mean(R_522:594)). Based on the results obtained, two new vegetation indices (VIs) were proposed: LSI_Red = R₆₇₄/R₆₅₄ and LSI_Norm = (R₆₇₄ − R₆₅₄)/(R₆₇₄ + R₆₅₄), indicating light stress by changes in the red region of the spectrum. The results of the study showed that LSI, LSI_Red, and LSI_Norm have a high degree of coupling strength with maximal quantum yields of photosystem II values. The plant response to EL exposure, as assessed by the values of these three VIs, was well expressed regardless of the PPFD levels. The effect of EL at non-stressful PPFDs (50–200 μmol photons m⁻²s⁻¹) was found to disappear within one hour after cessation of exposure. In contrast, the effect of the stressful PPFD (800 μmol photons m⁻²s⁻¹) was found to persist for at least 80 h after cessation of exposure. The results of the study indicate the need to consider light history in spectral monitoring of vegetation. Full article

Detecting objects in images is crucial for several applications, including surveillance, autonomous navigation, augmented reality, and so on. Although AI-based approaches such as Convolutional Neural Networks (CNNs) have proven highly effective in object detection, in scenarios where the objects being recognized are unknow, it is difficult to generalize an AI model for such tasks. In another trend, feature-based approaches like SIFT, SURF, and ORB offer the capability to search any object but have limitations under complex visual variations. In this work, we introduce a novel edge-based object/scene recognition method. We propose that utilizing feature edges, instead of feature points, offers high performance under complex visual variations. Our primary contribution is a directional pixel voting descriptor based on image segments. Experimental results are promising; compared to previous approaches, ours demonstrates superior performance under complex visual variations and high processing speed. Full article

Fire detection is crucial due to the exorbitant annual toll on both human lives and the economy resulting from fire-related incidents. To enhance forest fire detection in complex environments, we propose a new algorithm called FD-Net for various environments. Firstly, to improve detection performance, we introduce a Fire Attention (FA) mechanism that utilizes the position information from feature maps. Secondly, to prevent geometric distortion during image cropping, we propose a Three-Scale Pooling (TSP) module. Lastly, we fine-tune the YOLOv5 network and incorporate a new Fire Fusion (FF) module to enhance the network’s precision in identifying fire targets. Through qualitative and quantitative comparisons, we found that FD-Net outperforms current state-of-the-art algorithms in performance on both fire and fire-and-smoke datasets. This further demonstrates FD-Net’s effectiveness for application in fire detection. Full article

Ischemic stroke lesion segmentation in MRI images represents significant challenges, particularly due to class imbalance between foreground and background pixels. Several approaches have been developed to achieve higher F1-Scores in stroke lesion segmentation under this challenge. These strategies include convolutional neural networks (CNN) and models that represent a large number of parameters, which can only be trained on specialized computational architectures that are explicitly oriented to data processing. This paper proposes a lightweight model based on the U-Net architecture that handles an attention module and the Generalized Dice Focal loss function to enhance the segmentation accuracy in the class imbalance environment, characteristic of stroke lesions in MRI images. This study also analyzes the segmentation performance according to the pixel size of stroke lesions, giving insights into the loss function behavior using the public ISLES 2015 and ISLES 2022 MRI datasets. The proposed model can effectively segment small stroke lesions with F1-Scores over 0.7, particularly in FLAIR, DWI, and T2 sequences. Furthermore, the model shows reasonable convergence with their 7.9 million parameters at 200 epochs, making it suitable for practical implementation on mid and high-end general-purpose graphic processing units. Full article

Reversible data-hiding in encrypted images (RDHEI) plays a pivotal role in preserving privacy within images stored on cloud platforms. Recently, Wang et al. introduced a dual-stage permutation encryption scheme, which is highly compatible with RDHEI techniques. In this study, we undertake an exhaustive examination of the characteristics inherent to the dual-stage permutation scheme and propose two cryptanalysis schemes leveraging a large-kernel convolutional neural network (LKCNN) and a known plaintext attack (KPA) scheme, respectively. Our experimental findings demonstrate the effectiveness of our cryptanalysis schemes in breaking the dual-stage permutation encryption scheme. Based on our investigation, we highlight significant security vulnerabilities in the dual-stage permutation encryption scheme, raising concerns about its suitability for secure image storage and privacy protection in cloud environments. Full article