Search

Accurate in vivo imaging of G-quadruplexes (G4) is critical for understanding the emergence and progression of G4-associated diseases like cancer. However, existing in vivo G4 fluorescent probes primarily operate within the near-infrared region (NIR-I), which limits their application accuracy due to the short emission wavelength. The transition to second near-infrared (NIR-II) fluorescent imaging has been of significant interest, as it offers reduced autofluorescence and deeper tissue penetration, thereby facilitating more accurate in vivo imaging. Nonetheless, the advancement of NIR-II G4 probes has been impeded by the absence of effective probe design strategies. Herein, through a “step-by-step” rational design approach, we have successfully developed NIRG-2, the first small-molecule fluorescent probe with NIR-II emission tailored for in vivo G4 detection. Molecular docking calculations reveal that NIRG-2 forms stable hydrogen bonds and strong p–p interactions with G4 structures, which effectively inhibit twisted intramolecular charge transfer (TICT) and, thereby, selectively illuminate G4 structures. Due to its NIR-II emission (940 nm), large Stokes shift (90 nm), and high selectivity, NIRG-2 offers up to 47-fold fluorescence enhancement and a tissue imaging depth of 5 mm for in vivo G4 detection, significantly outperforming existing G4 probes. Utilizing NIRG-2, we have, for the first time, achieved high-contrast visualization of tumor metastasis through lymph nodes and precise tumor resection. Furthermore, NIRG-2 proves to be highly effective and reliable in evaluating surgical and drug treatment efficacy in cancer lymphatic metastasis models. We are optimistic that this study not only provides a crucial molecular tool for an in-depth understanding of G4-related diseases in vivo but also marks a promising strategy for the development of clinical NIR-II G4-activated probes.

Spectral reconstruction (SR) techniques can generate hyperspectral images (HSIs) from multispectral images (MSIs) with the same spatial resolution, thus alleviating the problem of limited availability and low spatial resolution of satellite HSIs. However, in scenarios where both HSIs and MSIs can be acquired simultaneously, spectral mapping relationship (SMR) among real images may not align with the sensor’s spectral response function (SRF), due to factors such as sensor noise and calibration errors. This mismatch can result in discrepancies in reflectivity between the reconstructed HSIs and the real HSIs. To solve the above problems, this article proposes a semi-supervised transfer learning SR (SSTSR) model based on gradient direction constraints. Through semi-supervised learning, SSTSR acquires precise SMRs in overlapping regions and extracts spectral trend information of ground objects from historical models in nonoverlapping regions. Experiments on two datasets demonstrate that the reconstructed HSIs closely resemble real HSIs, leading to impressive classification performance when employing a real HSI classifier.

Recently, deep learning models have dominated hyperspectral image (HSI) classification. Nowadays, deep learning is undergoing a paradigm shift with the rise of transformer-based foundation models. In this study, the potential of transformer-based foundation models, including the vision foundation model (VFM) and language foundation model (LFM), for HSI classification are investigated. First, to improve the performance of traditional HSI classification tasks, a spectral-spatial VFM-based transformer (SS-VFMT) is proposed, which inserts spectral-spatial information into the pretrained foundation transformer. Specifically, a given pretrained transformer receives HSI patch tokens for long-range feature extraction benefiting from the prelearned weights. Meanwhile, two enhancement modules, i.e., spatial and spectral enhancement modules (SpaEMs $\backslash $ SpeEMs), utilize spectral and spatial information for steering the behavior of the transformer. Besides, an additional patch relationship distillation strategy is designed for SS-VFMT to exploit the pretrained knowledge better, leading to the proposed SS-VFMT-D. Second, based on SS-VFMT, to address a new HSI classification task, i.e., generalized zero-shot classification, a spectral-spatial vision-language-based transformer (SS-VLFMT) is proposed. This task is to recognize novel classes not seen during training, which is more meaningful as the real world is usually open. The SS-VLFMT leverages SS-VFMT to extract spectral-spatial features and corresponding hash codes while integrating a pretrained language model to extract text features from class names. Experimental results on HSI datasets reveal that the proposed methods are competitive compared to the state-of-the-art methods. Moreover, the foundation model-based methods open a new window for HSI classification tasks, especially for HSI zero-shot classification.

Several fundamental remote sensing (RS) image processing tasks, including classification, segmentation, and detection, have been set to serve for manifold applications. In the RS community, the individual tasks have been studied separately for many years. However, the specialized models were only capable of a single task. They lacked the adaptability for generalizing to the other tasks. Moreover, transformer exhibits a powerful generalization capacity because it has the property of dynamic feature weighting. Hence, there is a large potential of a uniform transformer to learn multiple tasks simultaneously, i.e., multitask learning (MTL). An MTL transformer can combine knowledge from different tasks by sharing a uniform network. In this study, a general-purpose transformer, which simultaneously processes the three tasks, is investigated for RS MTL. To build a transformer capable of the three tasks, an MTL framework named RSCoTr is proposed. The framework uses a shared encoder to extract multiscale features efficiently and three task-specific decoders to obtain different results. Moreover, a flexible training procedure named co-training is proposed. The MTL model is trained with multiple general datasets annotated for individual tasks. The co-training is as easy as training a specialized model for a single task. It can be developed into different learning strategies to meet various requirements. The proposed RSCoTr is trained jointly with various strategies on three challenging datasets of the three tasks. And the results demonstrate that the proposed MTL method achieves state-of-the-art performance in comparison with other competitive approaches. Code will be available at https://github.com/Li-Qingyun/RSCoTr.

Existing oriented object detection in aerial images has progressed a lot in recent years and achieved favorable success. However, high-precision oriented object detection in aerial images remains a challenging task. Some recent works have adopted the classification-based method to predict the angle to address boundary problems in angle. However, we have found that these works often neglect the sensitivity of objects with different aspect ratios to angle. At the same time, it is worth exploring a suitable way to improve the emerging transformer-based approaches to adapt them to oriented object detection. In this article, we propose an Aspect Ratio-Sensitive DEtection TRansformer, termed ARS-DETR, for oriented object detection in aerial images. Specifically, a new angle classification method, called aspect ratio-aware circle smooth label (AR-CSL), is proposed to smooth the angle label more reasonably and discard the hyperparameter introduced by previous work [e.g., circular smooth label (CSL)]. Then, a rotated deformable attention (RDA) module is designed to rotate the sampling points with the corresponding angles and eliminate the misalignment between region features and sampling points. Moreover, a dynamic weight coefficient according to the aspect ratio is adopted to calculate the angle loss. Comprehensive experiments on several challenging datasets demonstrate that our method achieves a competitive performance in the high-precision oriented object detection task.

With the increasing availability and openness of remote sensing (RS) data collected from diverse sensors, there has been a growing interest in multimodal RS data classification. Nowadays, in the area of deep learning, there is a paradigm shift with the rise of foundation models, which are trained on large-scale datasets and are adaptable to a wide range of downstream tasks. In this study, the potential and effectiveness of foundation models for multimodal RS data classification is investigated. The training datasets of foundation models and multimodal RS datasets are quite different, and therefore, it is difficult to use a pretrained foundation model for multimodal RS data classification directly. To mitigate this difficulty, this article proposes a foundation model adaptation (FMA) framework for multimodal RS data classification without fine-tuning the parameters. Specifically, two learnable modules, i.e., cross-spatial interaction module and cross-channel interaction module, are proposed to add to the foundation model for extracting multimodal-specific representations. The cross-spatial and cross-channel interaction modules extract the characteristics of unimodal features along the spatial dimension and channel dimension, respectively. To effectively tackle the disparities among various RS modalities, an alignment approach (FMA2) is further explored based on the FMA. The FMA2 describes dependencies between different modalities by establishing a coupling score function, which can further enhance classification performance. To demonstrate the effectiveness and superiority of the FMA framework, comprehensive experiments are conducted on three multimodal RS datasets, showing improvement over the advanced multimodal RS data classification image methods.

Recently, metric-based few-shot learning (FSL) methods have achieved good performance in the hyperspectral image (HSI) classification. However, existing methods suffer from two problems: over-reliance on image modality information leads to inaccurate prototype representation, where a prototype refers to the centroid of each class in the dataset, and the impact of redundant and noisy pixels on model discriminability is rarely considered. These problems result in insufficient discriminability of the model for the target domain. To address the above issues, we propose a cross-domain few-shot HSI classification framework with cross-modal alignment and supervised contrastive learning (CDFS-CASCL). It is well known that human visual learning greatly benefits from the input of various modal information such as vision, language, and video. Inspired by the way humans abstract image class concepts in language form and understand the essence of classes, we perform cross-modal alignment (CA) between similar image and text prototypes and use abstract text semantics to guide the model to learn semantic-related features with good generalization ability in images, so as to improve the accuracy of image prototypes representation of the prototypes. In addition, through supervised contrastive learning (SCL) based on neighborhood pixel mask in the target domain, the enhanced sample features belonging to the same class are closer, while the enhanced sample features belonging to different classes are pulled further, enabling the model to learn mask-robust discriminative feature representations, suppressing the negative impact of redundant and noisy pixels, and improving the model’s discriminability. The experimental results demonstrate the superiority of the proposed CDFS-CASCL. The code is available at https://github.com/Li-ZK/CDFS-CASCL-2024.

CCR (catalyst continuous regeneration) catalytic reforming is one of the main processes in refineries in which high-octane gasoline and aromatics such as benzene, toluene, and xylene are produced. To meet the need of refineries to improve the efficiency of the reforming processes, a new kinetic model of the CCR catalytic reforming process based on the data of commercial catalytic reforming units for aromatics was built, which contains 36 lumps and 73 reactions in the network, and considers the effect of coke deposition on catalyst activity. The model parameters were validated by comparing the model simulated results and the measured values of the commercial reforming unit, and its results had a good agreement. The composition, temperature and pressure along the catalyst bed within the reactors as well as the effect of operational variables on the product distribution of CCR reforming were predicted with the developed model, providing guidance for the operation optimization of commercial CCR reforming units.

The purpose of this study was to explore the potential of the development and application of Paecilomyces hepiali, a fungus with edible and medicinal value, as a foodborne antioxidant and anti-aging agent. Its n-butanol extract (PHE) from rice cultures was selected for anti-aging experiment because of significant free radical scavenging activity in vitro. In vivo, PHE could significantly prolong the mean lifespan, 50 % survival days, and the maximum lifespan of Drosophilaon a high-fat diet.It is amazing that the mean lifespan increased from 19.1 days to 32.9 days, 50 % survival days increased from 15.7 days to 34.3 days, and the maximum lifespan extended from 44.7 days to 52.7 days, when the high-fat female Drosophilamodel was fed with 10 μg/mL PHE. Further research showed that PHE reduced the accumulation of peroxide products and increased the activity of antioxidant enzymes. Then, through antioxidant activity tracking, dimerumic acid (compound 1,the IC50value of 3.4 μg/mL on DPPH free radicals scavenging activity), 4,5-dihydroxy-3-methoxypentanoic acid (compound 2, new compound), and thymidine (compound 3) were isolated from PHE. It is worth mentioning that dimerumic acid, the major antioxidant compound of PHE (content up to 3 %), was discovered in P. hepialifor the first time. It was concluded that PHE showed excellent anti-aging activity at a very low concentration on fat diet-induced high fat Drosophila melanogaster, and dimerumic acid may be its main material basis. These results indicated that PHE had the potential to be developed as antioxidant and anti-aging agent in the healthcare industry.

[Display omitted]

Recurrent events after a first symptomatic deep venous thrombosis (DVT) are relatively frequent, but little is known about contralateral recurrent DVT (RDVT).

Recently, few-shot learning (FSL) has been introduced for hyperspectral image (HSI) classification with few labeled samples. However, existing FSL-based HSI classification methods mainly focus on the meta-knowledge transfer between HSIs. Compared with HSIs, natural images have sufficient annotated data. To utilize natural images (base class data) to achieve accurate classification of HSIs (novel class data), we propose a novel few-shot classification framework with SSL (FSCF-SSL) for HSIs in this article. The orientation of objects in natural images is relatively unitary, whereas the objects of image patches for each pixel in HSIs have diverse orientations in the spatial domain. To make better use of base classes, we design an SSL with geometric transformations (SSLGTs), which sets rotation labels as supervision to extract low-level features that can better represent diverse orientations, and then conduct SSLGT and FSL on base classes to learn transferable spatial meta-knowledge. Next, a spectral-spatial feature extraction network is carefully designed to better utilize the spatial and spectral information of HSIs, where the weights of the first seven layers of the spatial part are initialized by the weights of the corresponding layers trained on base classes. Finally, to fully explore the few annotated data from novel classes, we design an SSL with contrastive learning (SSLCL) that can mine the category-invariant features contained in the novel class data itself, and then perform SSLCL and FSL on novel classes to learn more discriminative individual knowledge. Experimental results on four HSI datasets show that FSCF-SSL offers a significant improvement over state-of-the-art methods. The code is available at https://github.com/Li-ZK/FSCF-SSL-2023.

Recently, due to the powerful capability at modeling the long-range relationships, Transformer-based methods have been widely explored in many research areas, including hyperspectral image (HSI) classification. However, because of lots of trainable parameters and the lack of inductive bias, it is difficult to train a Transformer-based HSI classifier, especially when the number of training samples is limited. To address this issue, in this study, spectral–spatial masked Transformer (SS-MTr) is explored for HSI classification, which uses a two-stage training strategy. In the first stage, SS-MTr pretrains a vanilla Transformer via reconstruction from masked HSI inputs, which embeds the local inductive bias into the Transformer. In the second stage, the well pretrained Transformer is cooperated with a fully connected layer and is then fine-tuned for the HSI classification. Furthermore, in order to incorporate discriminative feature learning into the SS-MTr, three SS-MTr-based methods, including contrastive SS-MTr (C-SS-MTr), supervised SS-MTr (S-SS-MTr), and supervised C-SS-MTr (SC-SS-MTr), are proposed by adding extra branches for specific tasks in parallel with the existing reconstruction task. Specifically, the proposed C-SS-MTr adds a contrastive loss, which brings instance discriminability. Besides, the proposed S-SS-MTr builds an extra classification branch for embracing interclass discriminability and intraclass similarity. Moreover, the proposed SC-SS-MTr combines C-SS-MTr and S-SS-MTr for better generalization. The proposed SS-MTr, C-SS-MTr, S-SS-MTr, and SC-SS-MTr are tested on three popular hyperspectral datasets (i.e., Indian Pines, Pavia University, and Houston). The obtained results reveal that the proposed models achieve competitive results compared with the state-of-the-art HSI classification methods. The code is available at https://github.com/mengduanjinghua/SS-MTr.

In recent years, deep learning models have been widely used for hyperspectral image (HSI) classification, and most of the existing deep-learning-based methods merely focused on high classification accuracy. However, in real applications, classification with low uncertainty matters as much as accurate classification. Unfortunately, the existing methods fail to consider uncertainty. To tackle this challenge, for the first time, Bayesian deep learning (BDL) is investigated to analyze the model uncertainty for HSI classification. Specifically, first, at the feature extraction (FE) stage, an HSI classification framework based on BDL, which contains two Bayesian Gabor layers and a global pooling layer (i.e., BDL-G2), is proposed. In BDL-G2, parameters in Gabor layers are sampled from the Gaussian distribution. The proposed BDL-G2 not only provides the uncertainty estimation but also strengthens the structure characteristic (i.e., texture) of HSI. Second, to model the uncertainty at the final classification stage, BDL-G2 is combined with a Bayesian fully connected layer (BFL) (i.e., BDL-G2-BFL), where the parameters’ distribution is adjusted adaptively. In the proposed BDL-G2-BFL, the uncertainty at FE and classification stages is captured, and a whole uncertainty estimation framework is established. Experimental results on the three public HSI datasets demonstrate the superiority in both accuracy and uncertainty. The proposed BDL-based methods pioneer a new direction and provide useful inspiration and experience for practical applications.

ABSTRACTConvolutional Neural Network (CNN) has been proved excellent for hyperspectral image(HSI) classification on the basis of large training set. However, the lack of available samples degrades the performance of CNNs in practical HSI classification. Although transfer learning is a typical strategy to address the issue of limited training samples, such methods tend to reload source model’s feature extractors directly ignoring their generalization. Furthermore, most existing adaptive transfer learning algorithms pay attention to spectral features of HSI while the obvious difference between RGB image and HSI’s spatial features is overlooked. In this paper, the proposed method can solve the mentioned problems to some extent. Firstly, the proposed adaptive heterogeneous transfer (AHT) can measure their transferability and find appropriate solution to transfer knowledge. It computes feature extractors’ transferability based on feature maps and influences the learning process of target model through backpropagation, therefore, target model is prone to receive knowledge from feature extractors with high generalization. Secondly, the proposed method lays emphasis on HSI’s spatial features. It adopts the methodology of spatial attention and calculates the regularization loss based on spatial features. Experiments were conducted on two widely used HSI datasets. The results demonstrate the method’s effectiveness when few samples are available.

The study of motor muscle control mechanisms can improve rehabilitation therapy and human-computer interaction technology. The limitations of traditional electroencephalography (EEG) limit the comprehensive understanding of motor muscle control mechanisms. Therefore, this study aims to explore the mechanism of motor muscle control based on optical EEG images, in order to expand the understanding of the process of motor control. The study selected optical EEG imaging technology as the main data acquisition tool. Optical EEG images have higher spatiotemporal resolution and can provide more detailed neural activity information. This technology combines optical imaging with EEG images to obtain spatiotemporal information of brain activity in a short period of time. The device is composed of multiple optical sensors and can measure blood oxygen concentration and neuronal activity in the cerebral cortex. Preprocess EEG image data using image processing and signal processing techniques, then use computational methods and algorithms to detect activated regions, and evaluate their relationships using correlation analysis and statistical methods. By comparing EEG image data and motor muscle activity data under different motor tasks. The research results show that optical EEG imaging technology can provide more detailed information on brain neural activity and accurately capture the activity patterns of different motor muscles. These results provide new perspectives and methods for further studying the control mechanisms of motor muscles.

Autophagy classically functions as a physiological process to degrade cytoplasmic components, protein aggregates, and/or organelles, as a mechanism for nutrient breakdown, and as a regulator of cellular architecture. Its biological functions include metabolic stress adaptation, stem cell differentiation, immunomodulation and diseases regulation, and so on. Current researches have proved that autophagy dysfunction may contribute to the pathogenesis of some myopathies through impairment of myofibres regeneration. Studies of autophagy inhibition also indicate the importance of autophagy in muscle regeneration, while activation of autophagy can restore muscle function in some myopathies. In this review, we aim to report the mechanisms of action of autophagy on muscle regeneration to provide relevant references for the treatment of regenerating defective myopathies by regulating autophagy. Results have shown that one key mechanism of autophagy regulating the muscle regeneration is to affect the differentiation fate of muscle stem cells (MuSCs), including quiescence maintenance, activation and differentiation. The roles of autophagy (organelle/protein degradation, energy facilitation, and/or other) vary at different myogenic stages of the repair process. When the muscle is in homeostasis, basal autophagy can maintain the quiescence state and stemness of MuSCs by renewing organelle and protein. After injury, the increased autophagy flux contributes to meet biological energy demand of MuSCs during activation and proliferation. By mitochondrial remodelling, autophagy during differentiation can promote the metabolic transformation and balance mitochondrial‐mediated apoptosis signals in myoblasts. Autophagy in mature myofibres is also essential for the degradation of necrotic myofibres, and may affect the dynamics of MuSCs by affecting the secretion spectrum of myofibres or the recruitment of supporting cells. Except for myogenic cells, autophagy also plays an important role in regulating the function of non‐myogenic cells in the muscle microenvironment, which is also essential for successful muscle recovery. Autophagy can regulate the immune microenvironment during muscle regeneration through the recruitment and polarization of macrophages, while autophagy in endothelial cells can regulate muscle regeneration in an angiogenic or angiogenesis‐independent manner. Drug or nutrition targeted autophagy has been preliminarily proved to restore muscle function in myopathies by promoting muscle regeneration, and further understanding the role and mechanism of autophagy in various cell types during muscle regeneration will enable more effective combinatorial therapeutic strategies.

ABSTRACT5-Methylcytosine (m5C) is a type of RNA modification that exists in tRNAs and rRNAs and was recently found in mRNA. Although mRNA m5C modification has been reported to regulate diverse biological process, its function in adipogenesis remains unknown. Here, we demonstrated that knockdown of NOL1/NOP2/Sun domain family member 2 (NSUN2), a m5C methyltransferase, increased lipid accumulation of 3T3-L1 preadipocytes through accelerating cell cycle progression during mitotic clonal expansion (MCE) at the early stage of adipogenesis. Mechanistically, we proved that NSUN2 directly targeted cyclin-dependent kinase inhibitor 1A (CDKN1A) mRNA, a key inhibitory regulator of cell cycle progression, and upregulated its protein expression in an m5C-dependent manner. Further study identified that CDKN1Awas the target of Aly/REF export factor (ALYREF), a reader of m5C modified mRNA. Upon NSUN2 deficiency, the recognition of CDKN1AmRNA by ALYREF was suppressed, resulting in the decrease of CDKN1AmRNA shuttling from nucleus to cytoplasm. Thereby, the translation of CDKN1A was reduced, leading to the acceleration of cell cycle and the promotion of adipogenesis. Together, these findings unveiled an important function and mechanism of the m5C modification on adipogenesis by controlling cell cycle progression, providing a potential therapeutic target to prevent obesity.

The scarcity and low spatial resolution of hyperspectral images (HSIs) have become a major problem limiting the application of the images. In recent years, spectral reconstruction (SR) has been applied to convert multispectral images (MSIs) with abundant quantities and high spatial resolution into HSIs. With the launch of several new multispectral (MS) satellites with a short repeat period, the simultaneous acquisition of images from multiple MS sensors in the same area is gradually becoming feasible. Unfortunately, existing SR methods only consider the reconstruction of the MSIs of a single sensor without considering using MSIs from different MS sensors to obtain a better construction effect through their complementary bands. However, multi-sensor SR is characterized by two problems: inconsistency in the amplitude information of real multisensor imaging and difficulty in the extraction of the complex correlations of bands from different sensors. To solve these problems, this paper proposes a multi-sensor SR framework based on a two-step approach in which the problems of amplitude inconsistency and band information extraction are solved using an ideal projection network and an ideal multi-sensor SR network, respectively. The effectiveness of the proposed method is verified by experiments on three datasets.

ABSTRACTDeep learning-based methods, especially deep convolutional neural network (CNN), have proven their powerfulness in hyperspectral image (HSI) classification. On the other hand, ensemble learning is a useful method for classification task. In this letter, in order to further improve the classification accuracy, the combination of CNN and random forest (RF) is proposed for HSI classification. The well-designed CNN is used as individual classifier to extract the discriminant features of HSI and RF randomly selects the extracted features and training samples to formulate a multiple classifier system. Furthermore, the learned weights of CNN are adopted to initialize other individual CNN. Experimental results with two hyperspectral data sets indicate that the proposed method provides competitive classification results compared with state-of-the-art methods.

Abstract

Abstract

Abstract

Abstract

ABSTRACTIn this letter, we present a new hyperspectral data-clustering method, named density analysis ensemble, from a different perspective. Instead of distance-based metrics in traditional clustering methods, we use density analysis for hyperspectral data clustering. Moreover, in order to improve the performance, we use the random subspace ensemble method to formulate a set of clustering systems. The final results are retrieved through majority voting. Compared to the k-means method, the overall accuracies have been improved by 7.05% and 6.93% for the Salinas and Pavia University data sets, respectively.

The influences of reheating duration on microstructures and tensile properties of thixoforged in situ Mg2Sip/AM60B composites have been investigated. The results indicate that the reheating time affects the volume fraction, composition and distribution of the liquid phase, the evolution of the liquid pools within the primary particles, and the morphology and distribution of the Mg2Si particles of the semisolid feedstock. These changes then affect the subsequent solidification behavior and deformation mechanism during thixoforging, which are responsible for the resulting microstructure and tensile properties. Mg2Si particles play a strengthening role for the matrix through the load transfer mechanism. The appropriate reheating time is 60 min at 600 °C when the mold preheating temperature is at 300 °C. The ultimate tensile strength and elongation of the resulting composite are up to 209 MPa and 11.9%, respectively.