TMErisk score: A tumor microenvironment-based model for predicting prognosis and immunotherapy in patients with head and neck squamous cell carcinoma.

Tumor microenvironment (TME) is closely associated with the progression and prognosis of head and neck squamous cell carcinoma (HNSCC). To investigate potential biomarkers for predicting therapeutic outcomes in HNSCC, we analyzed the immune and stromal status of HNSCC based on the genes associated with TME using the ESTIMATE algorithm. Immune and stromal genes were identified with differential gene expression and weighted gene co-expression network analysis (WGCNA). From these genes, 118 were initially selected through Cox univariate regression and then further input into least absolute shrinkage and selection operator (LASSO) regression analysis. As a result, 11 genes were screened out for the TME-related risk (TMErisk) score model which presented promising overall survival predictive potential. The TMErisk score was negatively associated with immune and stromal scores but positively associated with tumor purity. Individuals with high TMErisk scores exhibited decreased expression of most immune checkpoints and all human leukocyte antigen family genes, and reduced abundance of infiltrating immune cells. Divergent genes were mutated in HNSCC. In both high and low TMErisk score groups, the tumor protein P53 exhibited the highest mutation frequency. A higher TMErisk score was found to be associated with reduced overall survival probability and worse outcomes of immunotherapy. Therefore, the TMErisk score could serve as a valuable model for the outcome prediction of HNSCC in clinic.

On-chip integrated metasystem for spin-dependent multi-channel color holography.

On-chip integrated metasurface driven by in-plane guided waves is of great interests in various light-field manipulation applications such as colorful augmented reality and holographic display. However, it remains a challenge to design colorful multichannel holography by a single on-chip metasurface. Here we present metasurfaces integrated on top of a guided-wave photonic slab that achieves multi-channel colorful holographic light display. An end-to-end scheme is used to inverse design the metasurface for projecting off-chip preset multiple patterns. Particular examples are presented for customized patterns that were encoded into the metasurface with a single-cell meta-atom, working simultaneously at RGB color channels and for several different diffractive distances, with polarization dependence. Holographic images are generated at 18 independent channels with such a single-cell metasurface. The proposed design scheme is easy to implement, and the resulting device is viable for fabrication, promising plenty of applications in nanophotonics.

Continuous 3D Myocardial Motion Tracking via Echocardiography.

Myocardial motion tracking stands as an essential clinical tool in the prevention and detection of cardiovascular diseases (CVDs), the foremost cause of death globally. However, current techniques suffer from incomplete and inaccurate motion estimation of the myocardium in both spatial and temporal dimensions, hindering the early identification of myocardial dysfunction. To address these challenges, this paper introduces the Neural Cardiac Motion Field (NeuralCMF). NeuralCMF leverages implicit neural representation (INR) to model the 3D structure and the comprehensive 6D forward/backward motion of the heart. This method surpasses pixel-wise limitations by offering the capability to continuously query the precise shape and motion of the myocardium at any specific point throughout the cardiac cycle, enhancing the detailed analysis of cardiac dynamics beyond traditional speckle tracking. Notably, NeuralCMF operates without the need for paired datasets, and its optimization is self-supervised through the physics knowledge priors in both space and time dimensions, ensuring compatibility with both 2D and 3D echocardiogram video inputs. Experimental validations across three representative datasets support the robustness and innovative nature of the NeuralCMF, marking significant advantages over existing state-of-the-art methods in cardiac imaging and motion tracking. Code is available at: https://njuvision.github.io/NeuralCMF.

[Baduanjin improves sleep quality in patients with type 2 diabetes possibly via regulating Bmal1 gene].

The incidence of diabetes mellitus is increasing, and the sleep quality of patients with diabetes mellitus is often affected. Baduanjin may act on biological rhythm of the body, skeletal muscle glucose metabolism, skeletal muscle fibers and suprachiasmatic nucleus (SCN) by regulating the expression of Bmal1 gene, thus regulating the blood glucose level and circadian rhythm of patients with type 2 diabetes mellitus (T2DM) and improving their physiological functions. This article reviews the regulatory effect and mechanism of Baduanjin on Bmal1 gene expression in diabetes patients, and discusses the possibility of Baduanjin to improve the sleep quality of T2DM patients by regulating Bmal1 gene expression. This review can provide a new field for the clinical application of traditional Chinese Qigong Baduanjin, and provide a new scientific basis for exercise therapy of diabetes.

Effects of Cu2O nanoparticles on the kinetic characteristics of rapid chlorophyll fluorescence induction and related genes in wheat seedlings.

Metal nanoparticles could be accumulated in soils, which threatens the ecological stability of crops. Investigating the effects of cuprous oxide nanoparticles (Cu2O-NPs) on photosystem Ⅱ (PSⅡ) of wheat seedling leaves holds considerable importance in comprehending the implications of Cu2O-NPs on crop photosynthesis. Following the hydroponic method, we investigated the effects of 0, 10, 50, 100, and 200 mg·L-1 Cu2O-NPs on chlorophyll fluorescence induction kinetics and photosynthetic-related genes in wheat seedlings of "Zhoumai 18". The results showed that, with the increases of Cu2O-NPs concentrations, chlorophyll contents in wheat leaves decreased, and the standardization of the OJIP curve showed a clearly K-phase (ΔK＞0). Cu2O-NPs stress increased the parameters of active PSⅡ reaction centers, including the absorption flux per active RC (ABS/RC), the trapping flux per active RC (TRo/RC), the electron transport flux per active RC (ETo/RC), and the dissipation flux per active RC (DIo/RC). Cu2O-NPs stress decreased the parameters of PSⅡ energy distribution ratio including the maximum quantum yield of PSⅡ (φPo), the quantum yield of electron transport from QA (φEo), and the probability that a trapped exciton moved an electron further than QA (Ψo), while increased the quantum ratio for heat dissipation (φDo). Moreover, there was a decrease in photosynthetic quantum yield Y(Ⅱ), photochemical quenching coefficient (qP), net photosynthetic rate (Pn), stomatal conductance (gs), intercellular CO2 concentration (Ci), and transpiration rate (Tr) of leaves with the increases of Cu2O-NPs concentration. Under Cu2O-NPs stress, the expression levels of genes which included PSⅡ genes (PsbD, PsbP, Lhcb1), Rubisco large subunit genes (RbcL), cytochrome b6/f complex genes (PetD, Rieske), and ATP synthase genes (AtpA, AtpB, AtpE, AtpI) were downregulated. These results indicated that Cu2O-NPs stress altered the activity and structure of PSⅡ in wheat seedlings, affected the activity of PSⅡ reaction centers, performance parameters of PSⅡ donor and acceptor sides. PSⅡ related genes were downregulated and exhibited significant concentration effects.

Learning to Restore Compressed Point Cloud Attribute: A Fully Data-Driven Approach and A Rules-Unrolling-Based Optimization.

The emergence of holographic media drives the standardization of Geometry-based Point Cloud Compression (G-PCC) to sustain networked service provisioning. However, G-PCC inevitably introduces visually annoying artifacts, degrading the quality of experience (QoE). This work focuses on restoring G-PCC compressed point cloud attributes, e.g., RGB colors, to which fully data-driven and rules-unrolling-based post-processing filters are studied. At first, as compressed attributes exhibit nested blockiness, we develop a learning-based sample adaptive offset (NeuralSAO), which leverages a neural model using multiscale feature aggregation and embedding to characterize local correlations for quantization error compensation. Later, given statistically Gaussian distributed quantization noise, we suggest the utilization of a bilateral filter with Gaussian kernels to weigh neighbors by jointly considering their geometric and photometric contributions for restoration. Since local signals often present varying distributions, we propose estimating the smoothing parameters of the bilateral filter using an ultra-lightweight neural model. Such a bilateral filter with learnable parameters is called NeuralBF. The proposed NeuralSAO demonstrates the state-of-art restoration quality improvement, e.g., >20% BD-BR (Bjøntegaard delta rate) reduction over G-PCC on solid points clouds. However, NeuralSAO is computationally intensive and may suffer from poor generalization. On the other hand, although NeuralBF only achieves half of the gains of NeuralSAO, it is lightweight and exhibits impressive generalization across various samples. This comparative study between the data-driven large-scale NeuralSAO and the rules-unrolling-based small-scale NeuralBF helps to understand the capacity (i.e., performance, complexity, generalization) of underlying filters in terms of the quality restoration for compressed point cloud attribute.

Disorder-Invariant Implicit Neural Representation.

Implicit neural representation (INR) characterizes the attributes of a signal as a function of corresponding coordinates which emerges as a sharp weapon for solving inverse problems. However, the expressive power of INR is limited by the spectral bias in the network training. In this paper, we find that such a frequency-related problem could be greatly solved by re-arranging the coordinates of the input signal, for which we propose the disorder-invariant implicit neural representation (DINER) by augmenting a hash-table to a traditional INR backbone. Given discrete signals sharing the same histogram of attributes and different arrangement orders, the hash-table could project the coordinates into the same distribution for which the mapped signal can be better modeled using the subsequent INR network, leading to significantly alleviated spectral bias. Furthermore, the expressive power of the DINER is determined by the width of the hash-table. Different width corresponds to different geometrical elements in the attribute space, e.g., 1D curve, 2D curved-plane and 3D curved-volume when the width is set as 1, 2 and 3, respectively. More covered areas of the geometrical elements result in stronger expressive power. Experiments not only reveal the generalization of the DINER for different INR backbones (MLP versus SIREN) and various tasks (image/video representation, phase retrieval, refractive index recovery, and neural radiance field optimization) but also show the superiority over the state-of-the-art algorithms both in quality and speed.

Efficient Visual Computing With Camera RAW Snapshots.

Conventional cameras capture image irradiance (RAW) on a sensor and convert it to RGB images using an image signal processor (ISP). The images can then be used for photography or visual computing tasks in a variety of applications, such as public safety surveillance and autonomous driving. One can argue that since RAW images contain all the captured information, the conversion of RAW to RGB using an ISP is not necessary for visual computing. In this paper, we propose a novel ρ-Vision framework to perform high-level semantic understanding and low-level compression using RAW images without the ISP subsystem used for decades. Considering the scarcity of available RAW image datasets, we first develop an unpaired CycleR2R network based on unsupervised CycleGAN to train modular unrolled ISP and inverse ISP (invISP) models using unpaired RAW and RGB images. We can then flexibly generate simulated RAW images (simRAW) using any existing RGB image dataset and finetune different models originally trained in the RGB domain to process real-world camera RAW images. We demonstrate object detection and image compression capabilities in RAW-domain using RAW-domain YOLOv3 and RAW image compressor (RIC) on camera snapshots. Quantitative results reveal that RAW-domain task inference provides better detection accuracy and compression efficiency compared to that in the RGB domain. Furthermore, the proposed ρ-Vision generalizes across various camera sensors and different task-specific models. An added benefit of employing the ρ-Vision is the elimination of the need for ISP, leading to potential reductions in computations and processing times.

Genes that escape from X-chromosome inactivation and sexual dimorphism of systemic lupus erythematosus.

X chromosome inactivation can balance the effects of the two X chromosomes in females, and emerging evidence indicates that numerous genes on the inactivated X chromosome have the potential to evade inactivation. The mechanisms of escape include modification of DNA, RNA, histone, epitope, and various regulatory proteins, as well as the spatial structure of chromatin. The study of X chromosome inactivation escape has paved the way for investigating sex dimorphism in human diseases, particularly autoimmune diseases. It has been demonstrated that the presence of TLR7, CD40L, IRAK-1, CXCR3, and CXorf21 significantly contributes to the prevalence of SLE (systemic lupus erythematosus) in females. This article mainly reviews the molecular mechanisms underlying these genes that escape from X-chromosome inactivation and sexual dimorphism of systemic lupus erythematosus. Therefore, elucidating the molecular mechanisms underlying sexual dimorphism in SLE is not only crucial for diagnosing and treating the disease, but also holds theoretical significance in comprehensively understanding the development and regulatory mechanisms of the human immune system.

QARV: Quantization-Aware ResNet VAE for Lossy Image Compression.

This paper addresses the problem of lossy image compression, a fundamental problem in image processing and information theory that is involved in many real-world applications. We start by reviewing the framework of variational autoencoders (VAEs), a powerful class of generative probabilistic models that has a deep connection to lossy compression. Based on VAEs, we develop a new scheme for lossy image compression, which we name quantization-aware ResNet VAE (QARV). Our method incorporates a hierarchical VAE architecture integrated with test-time quantization and quantization-aware training, without which efficient entropy coding would not be possible. In addition, we design the neural network architecture of QARV specifically for fast decoding and propose an adaptive normalization operation for variable-rate compression. Extensive experiments are conducted, and results show that QARV achieves variable-rate compression, high-speed decoding, and better rate-distortion performance than existing baseline methods.

Effect of spinal orthosis on clinical outcomes of patients after oblique lumbar interbody fusion: a randomized controlled trial study protocol.

BACKGROUND: Oblique lumbar interbody fusion (OLIF) is an internationally popular minimally invasive technology for the treatment of various lumbar diseases. Since its introduction to China in 2014, OLIF technology has clearly shown its superiority in reconstructing intervertebral stability, restoring intervertebral space height, achieving indirect decompression, and restoring normal lumbar sequence. However, some patients still suffer from persistent symptoms after OLIF, including low back pain and soreness, which indirectly affect the overall surgical efficacy and patient satisfaction. Therefore, some clinicians recommend that patients routinely use spinal orthoses after OLIF to reduce the stress on the lower back muscles and ligaments, thereby relieving or avoiding postoperative residual symptoms or new symptoms. Accordingly, spinal orthosis use after OLIF has emerged as an essential option. However, the role of spinal orthoses in OLIF and their specific impact on postoperative patient clinical outcomes have remained unclear, and there is a lack of strong clinical evidence to indirectly or directly support the role of spinal orthoses in OLIF and demonstrate their impact on patient clinical outcomes. This study aims to investigate the role of spinal orthoses in OLIF by grouping patients based on the use or nonuse of spinal orthosis after OLIF, thus providing a better basis for the majority of patients and physicians. METHODS/DESIGN: We plan to conduct a 1-year randomized controlled trial involving 60 subjects. The subjects will be randomized into two groups: group A (those wearing spinal orthoses after surgery) and group B (those not wearing spinal orthoses after surgery). The clinical outcomes of these patients will be evaluated using the Oswestry disability index, visual analog scale, and Brantigan, Steffee, Fraser 1 day before surgery and 2 weeks and 1, 6, and 12 months after surgery. DISCUSSION: This randomized controlled trial aims to provide a reference for further comprehensive trial design. The findings of this study will provide a better and more scientific basis for the choice of postoperative rehabilitation and treatment for patients undergoing such a procedure. TRIAL REGISTRATION: This study has been registered in the Chinese Clinical Trial Registry (Registration No.: ChiCTR2200059000). Registration date: April 22, 2022. Registration website: http://www.chictr.org.cn/showproj.aspx?proj=166310.

Inverse design of a light nanorouter for a spatially multiplexed optical filter.

It is attractive to use an optical nanorouter by artificial nanostructures to substitute the traditional Bayer filter for an image array sensor, which, however, poses great challenges in balancing the design strategy and the ease of fabrication. Here, we implement and compare two inverse design schemes for rapid optimization of RGGB Bayer-type optical nanorouter. One is based on the multiple Mie scattering theory and the adjoint gradient that is applicable to arrays of nanospheres with varying sizes, and the other is based on the rigorous coupled wave analysis and the genetic algorithm. In both cases, we study layered nanostructures that can be efficiently modeled respectively which greatly accelerates the inverse design. It is shown that the color-dependent peak collection efficiencies of nanorouters designed in the two methods for red, green, and blue wavelengths reach 37%, 44%, and 45% and 52%, 50%, and 66%, respectively. We further demonstrate color nanorouters that provide light focusing to four quadrants working in both the visible and infrared bands, which promises multispectral imaging applications.

[Dual-Perspective Analysis of the Warming Effect of the Methane Emissions from Animal Husbandry in China].

Accurate quantitative evaluation of the greenhouse effects of methane(CH4) is the foundation for developing effective mitigation strategies. This study was the first to quantitatively evaluate the warming effects of the CH4 emissions from animal husbandry in China using the recently proposed climate metric GWP-star(GWP*), which is designed for short-lived climate pollutants(SLCP), and to compare the results with the commonly used climate metric global warming potential(GWP). The results showed:CH4emissions from animal husbandry in China decreased from 957.0×105 t in 2000 to 764.0×105 t. The GWP results showed that the greenhouse effect of CH4 emissions from animal husbandry in China was increasing between 2015 and 2019, and the GWP* results showed that it decreased compared to that 20 years ago. The amount of reduction was equivalent to removing the warming of 2.1×108 t of carbon dioxide. Under the GWP evaluation system, achieving carbon neutrality in the livestock industry in China requires eliminating or offsetting stable annual CH4 emissions from increased carbon sinks. Instead, under the GWP* evaluation system, China's livestock industry could achieve its carbon neutrality in the short term by effectively reducing CH4 emissions by only 0.3% per year. In the case that the livestock industry in China continues to take effective emission reduction measures, the reduction target under the GWP* metric will be reached earlier than that under GWP. Still, the choice of GWP or GWP* requires careful consideration of the objectives of evaluation, the time scale of assessment, and practical operability.

Bu-Fei Yi-Shen Granules Reduce Acute Exacerbations in Patients with GOLD 3-4 COPD: A Randomized Controlled Trial.

Purpose: Chronic obstructive pulmonary disease (COPD) is a disease characterized by frequent acute exacerbations (AEs), especially in severe and very severe cases. We aimed to evaluate the efficacy and safety of Bu-fei Yi-shen granules (BYGs) for COPD. Patients and Methods: We conducted a multicenter, randomized, double-blinded, placebo-controlled trial of 348 COPD patients with GOLD 3-4 COPD. The patients were randomly assigned into experimental or control groups in a 1:1 ratio. Patients in the experimental group were prescribed BYG, while those in the control group were administered a placebo, orally, twice daily, with 5 days on and 2 days off per week for 52 weeks. The outcomes included AEs, pulmonary function, clinical signs and symptoms, dyspnea scores (mMRC), quality of life scores, and a 6-minute walk test (6MWT). Results: A total of 280 patients completed the trial, including 135 patients in the experimental group and 145 in the control group. Compared to the control group, significant differences were observed in frequencies of AEs (mean difference: -0.35; 95% CI: -0.61, -0.10; P = 0.006) and AE-related hospitalizations (-0.18; 95% CI: -0.36, -0.01; P = 0.04), 6MWD (40.93 m; 95% CI: 32.03, 49.83; P < 0.001), mMRC (-0.57; 95% CI: -0.76, -0.37; P < 0.001), total symptoms (-2.18; 95% CI: -2.84, -1.53; P < 0.001), SF-36 (11.60; 95% CI: 8.23, 14.97; P < 0.001), and mCOPD-PRO (-0.45; 95% CI: -0.57, -0.33; P < 0.001) after treatment. However, there were no significant differences in mortality, pulmonary function, and mESQ-PRO scores (P > 0.05). No obvious adverse events were observed. Conclusion: BYG, as compared to a placebo, could significantly reduce the frequencies of AEs and AE-related hospitalizations for GOLD 3-4 COPD patients. Clinical symptoms, treatment satisfaction, quality of life, and exercise capacity improved. There was no significant improvement in mortality and pulmonary function.

GRNet: Geometry Restoration for G-PCC Compressed Point Clouds Using Auxiliary Density Signaling.

The lossy Geometry-based Point Cloud Compression (G-PCC) inevitably impairs the geometry information of point clouds, which deteriorates the quality of experience (QoE) in reconstruction and/or misleads decisions in tasks such as classification. To tackle it, this work proposes GRNet for the geometry restoration of G-PCC compressed large-scale point clouds. By analyzing the content characteristics of original and G-PCC compressed point clouds, we attribute the G-PCC distortion to two key factors: point vanishing and point displacement. Visible impairments on a point cloud are usually dominated by an individual factor or superimposed by both factors, which are determined by the density of the original point cloud. To this end, we employ two different models for coordinate reconstruction, termed Coordinate Expansion and Coordinate Refinement, to attack the point vanishing and displacement, respectively. In addition, 4-byte auxiliary density information is signaled in the bitstream to assist the selection of Coordinate Expansion, Coordinate Refinement, or their combination. Before being fed into the coordinate reconstruction module, the G-PCC compressed point cloud is first processed by a Feature Analysis Module for multiscale information fusion, in which kNN-based Transformer is leveraged at each scale to adaptively characterize neighborhood geometric dynamics for effective restoration. Following the common test conditions recommended in the MPEG standardization committee, GRNet significantly improves the G-PCC anchor and remarkably outperforms state-of-the-art methods on a great variety of point clouds (e.g., solid, dense, and sparse samples) both quantitatively and qualitatively. Meanwhile, GRNet runs fairly fast and uses a smaller-size model when compared with existing learning-based approaches, making it attractive to industry practitioners.

Unified Architecture Adaptation for Compressed Domain Semantic Inference.

Advances in both lossy image compression and semantic content understanding have been greatly fueled by deep learning techniques, yet these two tasks have been developed separately for the past decades. In this work, we address the problem of directly executing semantic inference from quantized latent features in the deep compressed domain without pixel reconstruction. Although different methods have been proposed for this problem setting, they either are restrictive to a specific architecture, or are sub-optimal in terms of compressed domain task accuracy. In contrast, we propose a lightweight, plug-and-play solution which is generally compliant with popular learned image coders and deep vision models, making it attractive to vast applications. Our method adapts prevalent pixel domain neural models that are deployed for various vision tasks to directly accept quantized latent features (other than pixels). We further suggest training the compressed domain model by transferring knowledge from its corresponding pixel domain counterpart. Experiments show that our method is compliant with popular learned image coders and vision task models. Under fair comparison, our approach outperforms a baseline method by a) more than 3% top-1 accuracy for compressed domain classification, and b) more than 7% mIoU for compressed domain semantic segmentation, at various data rates.

Somatic-to-primordial germ cell-like transformation is critical in tumor initiation of mouse breast tumor 4T1 cells.

It has been proposed that tumorigenicity was an intrinsic feature of embryonic/germ cell developmental axis as well as embryonic/germ cell-related genes play a crucial role in tumorigenicity. Our previous studies indicated that primordial germ cell (PGC)-like potential could be reactivated in tumorigenesis. In this study, 4T1, 168FARN and 67NR cells which originated from the same mouse breast cancer were studied and the results indicated that the acquisition of embryonic/germ cell-like state is essential for tumorigenicity. We further demonstrated that somatic to PGC-like transformation (SPLT) was activated in 4T1 cells and that inhibition of PGC-like cell formation by depleting pluripotency and/or PGC specification-related genes markedly repressed SPLT and the tumorigenicity. Collectively, our findings reveal that tumorigenicity is linked to the acquisition of PGC-like state through SPLT in 4T1 cells, providing new insight into deeper understanding the biological nature of tumors and novel therapeutical strategies for cancer targeting.

Stereodivergent Access to [6.7]-Fused N-Heterocycles Bearing 1,3-Nonadjacent Stereogenic Centers by Pd-Catalyzed [4 + 2] Annulations.

We describe a highly efficient stereodivergent [4 + 2] annulation reaction of vinyl benzoxazinaones and seven-membered cyclic N-sulfonyl aldimines for the synthesis of a wide array of N-heterocycles with 1,3-nonadjacent stereogenic centers via palladium catalysis. The polarity of solvents was found to play a key role in the switch of diastereoselectivity. Furthermore, good enantioselectivities of these reactions were achieved by the employment of commercially available Wingphos as the chiral ligand.

Single-cell transcriptomic analyses reveal cellular and molecular patterns of rubber tree response to early powdery mildew infection.

As a perennial woody plant, the rubber tree (Hevea brasiliensis) must adapt to various environmental challenges through gene expression in multiple cell types. It is still unclear how genes in this species are expressed at the cellular level and the precise mechanisms by which cells respond transcriptionally to environmental stimuli, especially in the case of pathogen infection. Here, we characterized the transcriptomes in Hevea leaves during early powdery mildew infection using single-cell RNA sequencing. We identified 10 cell types and constructed the first single-cell atlas of Hevea leaves. Distinct gene expression patterns of the cell clusters were observed under powdery mildew infection, which was especially significant in the epidermal cells. Most of the genes involved in host-pathogen interactions in epidermal cells exhibited a pattern of dramatically increased expression with increasing pseudotime. Interestingly, we found that the HbCNL2 gene, encoding a nucleotide-binding leucine-rich repeat protein, positively modulated the defence of rubber leaves against powdery mildew. Overexpression of the HbCNL2 gene triggered a typical cell death phenotype in tobacco leaves and a higher level of reactive oxygen species in the protoplasts of Hevea leaves. The HbCNL2 protein was located in the cytomembrane and nucleus, and its leucine-rich repeat domain interacted with the histidine kinase-like ATPase domain of the molecular chaperone HbHSP90 in the nucleus. Collectively, our results provide the first observation of the cellular and molecular responses of Hevea leaves to biotrophic pathogen infection and can guide the identification of disease-resistance genes in this important tree species.

Sparse Tensor-Based Multiscale Representation for Point Cloud Geometry Compression.

This study develops a unified Point Cloud Geometry (PCG) compression method through the processing of multiscale sparse tensor-based voxelized PCG. We call this compression method SparsePCGC. The proposed SparsePCGC is a low complexity solution because it only performs the convolutions on sparsely-distributed Most-Probable Positively-Occupied Voxels (MP-POV). The multiscale representation also allows us to compress scale-wise MP-POVs by exploiting cross-scale and same-scale correlations extensively and flexibly. The overall compression efficiency highly depends on the accuracy of estimated occupancy probability for each MP-POV. Thus, we first design the Sparse Convolution-based Neural Network (SparseCNN) which stacks sparse convolutions and voxel sampling to best characterize and embed spatial correlations. We then develop the SparseCNN-based Occupancy Probability Approximation (SOPA) model to estimate the occupancy probability either in a single-stage manner only using the cross-scale correlation, or in a multi-stage manner by exploiting stage-wise correlation among same-scale neighbors. Besides, we also suggest the SparseCNN based Local Neighborhood Embedding (SLNE) to aggregate local variations as spatial priors in feature attribute to improve the SOPA. Our unified approach not only shows state-of-the-art performance in both lossless and lossy compression modes across a variety of datasets including the dense object PCGs (8iVFB, Owlii, MUVB) and sparse LiDAR PCGs (KITTI, Ford) when compared with standardized MPEG G-PCC and other prevalent learning-based schemes, but also has low complexity which is attractive to practical applications.