Explore the vast range of titles available.

HSI classification is essential for monitoring and analyzing the Earth’s surface, with methods utilizing convolutional neural networks (CNNs) and transformers rapidly gaining prominence and advancing in recent years. However, CNNs are limited by their restricted receptive fields and can only process local information. Although transformers excel at establishing long-range dependencies, they underutilize the spatial information of HSIs. To tackle these challenges, we present the multi-frequency feature enhancement network (DB-MFENet) for HSI classification. First, orthogonal position encoding (OPE) is employed to map image coordinates into a high-dimensional space, which is then combined with corresponding spectral values to compute a multi-frequency feature. Next, the multi-frequency feature is divided into low-frequency and high-frequency components, which are separately enhanced through a dual-branch structure and then fused. Finally, a transformer encoder and a linear layer are employed to encode and classify the enhanced multi-frequency feature. The experimental results demonstrate that our method is efficient and robust for HSIs classification, achieving overall accuracies of 97.05%, 91.92%, 98.72%, and 96.31% on Indian Pines, Salinas, Pavia University, and WHU-Hi-HanChuan datasets, respectively.

Indole-3-carbinol(I3C) is a tumor chemopreventive substance that can be extracted from cruciferous vegetables. Indole-3-carbinol (I3C) has been shown to have antioxidant and anti-inflammatory effects. In this study, we investigated the cerebral protective effects of I3C in an in vivo rats model of middle cerebral artery occlusion (MCAO). 8–10 Week-Old male SD rat received I3C (150 mg/kg, once daily) for 3 days and underwent 3 h of middle cerebral artery occlusion (MCAO) followed by reperfusion. The results showed that I3C pretreatment (150 mg/kg, once daily) prevented CIRI-induced cerebral infarction in rats. I3C pretreatment also decreased the mRNA expression levels of several apoptotic proteins, including Bax, caspase-3 and caspase-9, by increasing the mRNA expression levels of the anti-apoptotic protein Bcl-2. Inhibited apoptosis in the brain cells of MCAO rats. In addition, we found that I3C pretreatment reduced neuronal loss, promoted neurological recovery after ischemia–reperfusion injury and increased seven-day survival in MCAO rats. I3C pretreatment also significantly reduced the expression of inducible nitric oxide synthase (INOS), interleukin-1β (IL-1β) and interleukin-6 (IL-6) mRNA in ischemic brain tissue; Increased expression of interleukin-4 (IL-4) and interleukin-10 (IL-10) mRNA. At the same time, I3C pretreatment significantly decreased the expression of the M1 microglial marker IBA1 after cerebral ischemia–reperfusion injury and increased the expression of these results in the M2 microglial marker CD206. I3C pretreatment also significantly decreased apoptosis and death of HAPI microglial cells after hypoxia induction, decreased interleukin-1β (IL-1β) and interleukin-6 (IL-6) mRNA The expression of interleukin-4 (IL-4) and interleukin-10 (IL-10) mRNAs was increased. These results suggest that I3C protects the brain from CIRI by regulating the anti-inflammatory and anti-apoptotic effects of microglia.

In this paper we present a novel method for efficient and effective 3D surface reconstruction in open scenes. Existing Neural Radiance Fields (NeRF) based works typically require extensive training and rendering time due to the adopted implicit representations. In contrast, 3D Gaussian splatting (3DGS) uses an explicit and discrete representation, hence the reconstructed surface is built by the huge number of Gaussian primitives, which leads to excessive memory consumption and rough surface details in sparse Gaussian areas. To address these issues, we propose Gaussian Voxel Kernel Functions (GVKF), which establish a continuous scene representation based on discrete 3DGS through kernel regression. The GVKF integrates fast 3DGS rasterization and highly effective scene implicit representations, achieving high-fidelity open scene surface reconstruction. Experiments on challenging scene datasets demonstrate the efficiency and effectiveness of our proposed GVKF, featuring with high reconstruction quality, real-time rendering speed, significant savings in storage and training memory consumption.

Existing auto-regressive mesh generation approaches suffer from ineffective topology preservation, which is crucial for practical applications. This limitation stems from previous mesh tokenization methods treating meshes as simple collections of equivalent triangles, lacking awareness of the overall topological structure during generation. To address this issue, we propose a novel mesh tokenization algorithm that provides a canonical topological framework through vertex layering and ordering, ensuring critical geometric properties including manifoldness, watertightness, face normal consistency, and part awareness in the generated meshes. Measured by Compression Ratio and Bits-per-face, we also achieved state-of-the-art compression efficiency. Furthermore, we introduce an online non-manifold data processing algorithm and a training resampling strategy to expand the scale of trainable dataset and avoid costly manual data curation. Experimental results demonstrate the effectiveness of our approach, showcasing not only intricate mesh generation but also significantly improved geometric integrity.

We introduce Mesh Silksong, a compact and efficient mesh representation tailored to generate the polygon mesh in an auto-regressive manner akin to silk weaving. Existing mesh tokenization methods always produce token sequences with repeated vertex tokens, wasting the network capability. Therefore, our approach tokenizes mesh vertices by accessing each mesh vertice only once, reduces the token sequence's redundancy by 50\\%, and achieves a state-of-the-art compression rate of approximately 22\\%. Furthermore, Mesh Silksong produces polygon meshes with superior geometric properties, including manifold topology, watertight detection, and consistent face normals, which are critical for practical applications. Experimental results demonstrate the effectiveness of our approach, showcasing not only intricate mesh generation but also significantly improved geometric integrity.

This paper investigates an open research challenge of reconstructing high-quality, large 3D open scenes from images. It is observed existing methods have various limitations, such as requiring precise camera poses for input and dense viewpoints for supervision. To perform effective and efficient 3D scene reconstruction, we propose a novel graph-guided 3D scene reconstruction framework, GraphGS. Specifically, given a set of images captured by RGB cameras on a scene, we first design a spatial prior-based scene structure estimation method. This is then used to create a camera graph that includes information about the camera topology. Further, we propose to apply the graph-guided multi-view consistency constraint and adaptive sampling strategy to the 3D Gaussian Splatting optimization process. This greatly alleviates the issue of Gaussian points overfitting to specific sparse viewpoints and expedites the 3D reconstruction process. We demonstrate GraphGS achieves high-fidelity 3D reconstruction from images, which presents state-of-the-art performance through quantitative and qualitative evaluation across multiple datasets. Project Page: https://3dagentworld.github.io/graphgs.

Implicit neural representation (INR) is a popular approach for arbitrary-scale image super-resolution (SR), as a key component of INR, position encoding improves its representation ability. Motivated by position encoding, we propose orthogonal position encoding (OPE) - an extension of position encoding - and an OPE-Upscale module to replace the INR-based upsampling module for arbitrary-scale image super-resolution. Same as INR, our OPE-Upscale Module takes 2D coordinates and latent code as inputs; however it does not require training parameters. This parameter-free feature allows the OPE-Upscale Module to directly perform linear combination operations to reconstruct an image in a continuous manner, achieving an arbitrary-scale image reconstruction. As a concise SR framework, our method has high computing efficiency and consumes less memory comparing to the state-of-the-art (SOTA), which has been confirmed by extensive experiments and evaluations. In addition, our method has comparable results with SOTA in arbitrary scale image super-resolution. Last but not the least, we show that OPE corresponds to a set of orthogonal basis, justifying our design principle.

Osimertinib (Osi) is a widely used epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI). However, the emergence of resistance is inevitable, partly due to the gradual evolution of adaptive resistant cells during initial treatment. Here, we find that Osi treatment rapidly triggers adaptive resistance in tumor cells. Metabolomics analysis reveals a significant enhancement of oxidative phosphorylation (OXPHOS) in Osi adaptive-resistant cells. Mechanically, Osi treatment induces an elevation of NCOA4, a key protein of ferritinophagy, which maintains the synthesis of iron-sulfur cluster (ISC) proteins of electron transport chain and OXPHOS. Additionally, active ISC protein synthesis in adaptive-resistant cells significantly increases the sensitivity to copper ions. Combining Osi with elesclomol, a copper ion ionophore, significantly increases the efficacy of Osi, with no additional toxicity. Altogether, this study reveals the mechanisms of NCOA4-mediated ferritinophagy in Osi adaptive resistance and introduces a promising new therapy of combining copper ionophores to improve its initial efficacy. The mechanisms leading to acquired resistance to targeted therapy in cancer are not completely understood. Here, the authors show that ferritinophagy mediates adaptive resistance to Osimertinib, and that combining this EGFR tyrosine kinase inhibitor with copper ionophores improves its therapeutic efficacy in preclinical models of non-small cell lung cancer.

Background Lung adenocarcinoma (LUAD) is the leading cause of death worldwide. However, the roles of long noncoding RNAs (lncRNAs) hijacked by super-enhancers (SEs), vital regulatory elements of the epigenome, remain elusive in the progression of LUAD metastasis. Methods SE-associated lncRNA microarrays were used to identify the dysregulated lncRNAs in LUAD. ChIP-seq, Hi-C data analysis, and luciferase reporter assays were utilized to confirm the hijacking of LINC01977 by SE. The functions and mechanisms of LINC01977 in LUAD were explored by a series of in vitro and in vivo assays. Results We found that LINC01977 , a cancer-testis lncRNA, was hijacked by SE, which promoted proliferation and invasion both in vitro and in vivo. LINC01977 interacted with SMAD3 to induce its nuclear transport, which facilitated the interaction between SMAD3 and CBP/P300, thereby regulating the downstream target gene ZEB1. Additionally, SMAD3 up-regulated LINC09177 transcription by simultaneously binding the promoter and SE, which was induced by the infiltration of M2-like tumor-associated macrophages (TAM2), subsequently activating the TGF-β/SMAD3 pathway. Moreover, LINC01977 expression was positively correlated with TAM2 infiltration and SMAD3 expression, especially in early-stage LUAD. Higher chromatin accessibility in the SE region of LINC01977 was observed with high expression of TGF-β. Early-stage LUAD patients with high LIN01977 expression had a shorter disease-free survival. Conclusions TAM2 infiltration induced a rich TGF-β microenvironment, activating SMAD3 to bind the promoter and the SE of LINC01977 , which up-regulated LINC01977 expression. LINC01977 also promoted malignancy via the canonical TGF-β/SMAD3 pathway. LINC01977 hijacked by SE could be a valuable therapeutic target, especially for the treatment of early-stage LUAD.

Evaluating the accuracy of pulmonary nodule diagnosis avoids repeated low‐dose computed tomography (LDCT)/CT scans or invasive examination, yet remains a main clinical challenge. Screening for new diagnostic tools is urgent. Herein, we established a nomogram based on the diagnostic signature of five circulating tsRNAs and CT information to predict malignant pulmonary nodules. In total, 249 blood samples of patients with pulmonary nodules were selected from three different lung cancer centers. Five tsRNAs were identified in the discovery and training cohorts and the diagnostic signature was established by the randomForest algorithm (tRF‐Ser‐TGA‐003, tRF‐Val‐CAC‐005, tRF‐Ala‐AGC‐060, tRF‐Val‐CAC‐024, and tiRNA‐Gln‐TTG‐001). A nomogram was developed by combining tsRNA signature and CT information. The high level of accuracy was identified in an internal validation cohort (n = 83, area under the receiver operating characteristic curve [AUC] = 0.930, sensitivity 100.0%, specificity 73.8%) and external validation cohort (n = 66, AUC = 0.943, sensitivity 100.0%, specificity 86.8%). Furthermore, the diagnostic ability of our model discriminating invasive malignant ones from noninvasive lesions was assessed. A robust performance was achieved in the diagnosis of invasive malignant lesions in both training and validation cohorts (discovery cohort: AUC = 0.850, sensitivity 86.0%, specificity 81.4%; internal validation cohort: AUC = 0.784, sensitivity 78.8%, specificity 78.1%; and external validation cohort: AUC = 0.837, sensitivity 85.7%, specificity 84.0%). This novel circulating tsRNA‐based diagnostic model has potential significance in predicting malignant pulmonary nodules. Application of the model could improve the accuracy of pulmonary nodule diagnosis and optimize surgical plans. Low‐dose computed tomography screening resulted in high detection rates of pulmonary nodules, which warranted invasive diagnostic procedures with high risk. We show that combining novel tsRNAs biomarkers in liquid biopsy with computed tomography information provides improved diagnostic accuracy. The model achieved robust accuracy in internal and external validation cohorts. In addition, this model provided fair results in discriminating invasive nodules from noninvasive nodules.