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.

Revealing underlying regulatory mechanisms of LINC00313 in Osimertinib-resistant LUAD cells by ceRNA network analysis.

BACKGROUND: Osimertinib, a third-generation epidermal growth factor receptor-tyrosine kinase inhibitor (EGFR-TKI), is the preferred treatment for EGFR-mutated lung cancer. However, acquired resistance inevitably develops. While non-coding RNAs have been implicated in lung cancer through various functions, the molecular mechanisms responsible for osimertinib resistance remain incompletely elucidated. METHODS: RNA-sequencing technology was employed to determine differentially expressed lncRNAs (DE-lncRNAs) and mRNAs (DE-mRNAs) between H1975 and H1975OR cell lines. Starbase 2.0 was utilized to predict DE-lncRNA and DE-mRNA interactions, constructing ceRNA networks. Subsequently, functional and pathway enrichment analysis were performed on target DE-mRNAs to identify pathways associated with osimertinib resistance. Key target DE-mRNAs were then selected as potential risk signatures for lung adenocarcinoma (LUAD) prognostic modeling using multivariate Cox regression analyses. The Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR) and immunohistochemistry staining were used for result validation. RESULTS: Functional analysis revealed that the identified DE-mRNAs primarily enriched in EGFR-TKI resistance pathways, especially in the PI3K/Akt signaling pathway, where their concerted actions may lead to osimertinib resistance. Specifically, upregulation of LINC00313 enhanced COL1A1 expression by acting as a miR-218-5p sponge, triggering an upstream response that activates the PI3K/Akt pathway, potentially contributing to osimertinib resistance. Furthermore, the expressions of LINC00313 and COL1A1 were validated by qRT-PCR, and the activation of the PI3K/Akt pathway was confirmed by immunohistochemistry staining. CONCLUSIONS: Our results suggest that the LINC00313/miR-218-5p/COL1A1 axis potentially contributes to osimertinib resistance through the PI3K/Akt signaling pathway, providing novel insights into the molecular mechanisms underlying acquired osimertinib resistance in LUAD. Additionally, our study may aid in the identification of potential therapeutic targets for overcoming resistance to osimertinib.

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.

A transdermal drug delivery system based on dissolving microneedles for boron neutron capture therapy of melanoma.

Boron neutron capture therapy (BNCT) is a promising therapy for malignant tumors that requires selective and high concentrations of 10B accumulation in tumor cells. Despite ongoing developments in novel boron agents and delivery carriers, the progress and clinical application of BNCT is still restricted by the low 10B accumulation and tumor-to-normal tissue (T/N) ratio. Herein, a dissolving microneedle-based transdermal drug delivery system was specifically designed for BNCT in a mouse model of melanoma. By incorporating fructose-BPA (F-BPA) into PVA microneedle tips, this system successfully delivered sufficient F-BPA into the melanoma site after the application of only two patches. Notably, the T/N ratio achieved through the treatment combining PVA/F-BPA MNs with BNCT (PVA/F-BPA MNs-BNCT) surpassed 93.16, signifying a great improvement. Furthermore, this treatment approach effectively inhibited tumor growth and significantly enhanced the survival rate of the mice. In brief, our study introduces a novel, simple, and efficient administration strategy for BNCT, opening new possibilities for the design of nanomedicine for BNCT.

Resourcefulness as a mediator in the relationship between self-perceived burden and depression among the young and middle-aged stroke patients: A cross-sectional study.

Objective: To explore the relationships among self-perceived burden, resourcefulness and depression, and to study further whether resourcefulness mediates the effects of self-perceived burden on depression in young and middle-aged stroke patients. Methods: A cross-sectional survey was conducted with 1050 young and middle-aged stroke patients. We used a general demographic questionnaire, Self-Perceived Burden Scale (SPBS), Resourcefulness Scale© (RS) and Hamilton Depression Scale (HAMD) to assess self-perceived burden, resourcefulness, and depression. Statistical methods included correlation analysis, multiple linear regression, and structural equation model. Results: 1018 valid questionnaires were collected with a response rate of 96.95%. Resourcefulness was inversely correlated with self-perceived burden (r = -0.367, p < 0.01) and depression (r = -0.625, p < 0.01); Self-perceived burden was positively associated with depression (r = 0.698, p < 0.01). Multiple linear regression analyses showed that resourcefulness mediated the effects of self-perceived burden on depression; The structural equation model demonstrated that the resourcefulness mediated the relationship between self-perceived burden and depression. Conclusion: Resourcefulness is a mediator between self-perceived burden and depression. Medical staff adjust the psychological state of stroke patients based on the theory of resourcefulness, thereby improving their problem-solving ability, actively encouraging patients to establish problem-solving strategies, providing disease rehabilitation knowledge and skills, and promoting the improvement of resourcefulness level.

SPNE: sample-perturbed network entropy for revealing critical states of complex biological systems.

Complex biological systems do not always develop smoothly but occasionally undergo a sharp transition; i.e. there exists a critical transition or tipping point at which a drastic qualitative shift occurs. Hunting for such a critical transition is important to prevent or delay the occurrence of catastrophic consequences, such as disease deterioration. However, the identification of the critical state for complex biological systems is still a challenging problem when using high-dimensional small sample data, especially where only a certain sample is available, which often leads to the failure of most traditional statistical approaches. In this study, a novel quantitative method, sample-perturbed network entropy (SPNE), is developed based on the sample-perturbed directed network to reveal the critical state of complex biological systems at the single-sample level. Specifically, the SPNE approach effectively quantifies the perturbation effect caused by a specific sample on the directed network in terms of network entropy and thus captures the criticality of biological systems. This model-free method was applied to both bulk and single-cell expression data. Our approach was validated by successfully detecting the early warning signals of the critical states for six real datasets, including four tumor datasets from The Cancer Genome Atlas (TCGA) and two single-cell datasets of cell differentiation. In addition, the functional analyses of signaling biomarkers demonstrated the effectiveness of the analytical and computational results.

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.

Allele-aware chromosome-level genome assembly of Artemisia annua reveals the correlation between ADS expansion and artemisinin yield.

Artemisia annua is the major natural source of artemisinin, an anti-malarial medicine commonly used worldwide. Here, we present chromosome-level haploid maps for two A. annua strains with different artemisinin contents to explore the relationships between genomic organization and artemisinin production. High-fidelity sequencing, optical mapping, and chromatin conformation capture sequencing were used to assemble the heterogeneous and repetitive genome and resolve the haplotypes of A. annua. Approximately 50,000 genes were annotated for each haplotype genome, and a triplication event that occurred approximately 58.12 million years ago was examined for the first time in this species. A total of 3,903,467-5,193,414 variants (SNPs, indels, and structural variants) were identified in the 1.5-Gb genome during pairwise comparison between haplotypes, consistent with the high heterozygosity of this species. Genomic analyses revealed a correlation between artemisinin concents and the copy number of amorpha-4,11-diene synthase genes. This correlation was further confirmed by resequencing of 36 A. annua samples with varied artemisinin contents. Circular consensus sequencing of transcripts facilitated the detection of paralog expression. Collectively, our study provides chromosome-level allele-aware genome assemblies for two A. annua strains and new insights into the biosynthesis of artemisinin and its regulation, which will contribute to conquering malaria worldwide.

Identifying the critical states and dynamic network biomarkers of cancers based on network entropy.

BACKGROUND: There are sudden deterioration phenomena during the progression of many complex diseases, including most cancers; that is, the biological system may go through a critical transition from one stable state (the normal state) to another (the disease state). It is of great importance to predict this critical transition or the so-called pre-disease state so that patients can receive appropriate and timely medical care. In practice, however, this critical transition is usually difficult to identify due to the high nonlinearity and complexity of biological systems. METHODS: In this study, we employed a model-free computational method, local network entropy (LNE), to identify the critical transition/pre-disease states of complex diseases. From a network perspective, this method effectively explores the key associations among biomolecules and captures their dynamic abnormalities. RESULTS: Based on LNE, the pre-disease states of ten cancers were successfully detected. Two types of new prognostic biomarkers, optimistic LNE (O-LNE) and pessimistic LNE (P-LNE) biomarkers, were identified, enabling identification of the pre-disease state and evaluation of prognosis. In addition, LNE helps to find "dark genes" with nondifferential gene expression but differential LNE values. CONCLUSIONS: The proposed method effectively identified the critical transition states of complex diseases at the single-sample level. Our study not only identified the critical transition states of ten cancers but also provides two types of new prognostic biomarkers, O-LNE and P-LNE biomarkers, for further practical application. The method in this study therefore has great potential in personalized disease diagnosis.

Analysis of the lncRNA-miRNA-mRNA Network Reveals a Potential Regulatory Mechanism of EGFR-TKI Resistance in NSCLC.

Epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) are widely used for patients with EGFR-mutated lung cancer. Despite its initial therapeutic efficacy, most patients eventually develop drug resistance, which leads to a poor prognosis in lung cancer patients. Previous investigations have proved that non-coding RNAs including long non-coding RNAs (lncRNAs), circular RNAs (circRNAs), and microRNAs (miRNAs) contribute to drug resistance by various biological functions, whereas how they regulate EGFR-TKI resistance remains unclear. In this study, we examined gene expression using the microarray technology on gefitinib-resistant NSCLC cells to obtain differentially expressed (DE) lncRNAs and mRNAs. A total of 45 DE-lncRNAs associated with overall survival and 1799 target DE-mRNAs were employed to construct a core lncRNA-miRNA-mRNA network to illustrate underlying molecular mechanisms of how EGFR-TKI resistance occurs in NSCLC. We found that target DE-mRNAs were mainly enriched in pathways involved in EGFR-TKI resistance, especially the target DE-mRNAs regulated by LINC01128 were significantly enriched in the PI3K/Akt signaling pathway, where the synergy of these target DE-mRNAs may play a key role in EGFR-TKI resistance. In addition, downregulated LINC01128, acting as a specific miRNA sponge, decreases PTEN via sponging miR-25-3p. Furthermore, signaling reactions caused by the downregulation of PTEN would activate the PI3K/Akt signaling pathway, which may lead to EGFR-TKI resistance. In addition, a survival analysis indicated the low expression of LINC01128, and PTEN is closely related to poor prognosis in lung adenocarcinoma (LUAD). Therefore, the LINC01128/miR-25-3p/PTEN axis may promote EGFR-TKI resistance via the PI3K/Akt signaling pathway, which provides new insights into the underlying molecular mechanisms of drug resistance to EGFR-TKIs in NSCLC. In addition, our study sheds light on developing novel therapeutic approaches to overcome EGFR-TKI resistance in NSCLC.

Intracellular marriage of bicarbonate and Mn ions as "immune ion reactors" to regulate redox homeostasis and enhanced antitumor immune responses.

BACKGROUND: Different from Fe ions in Fenton reaction, Mn ions can function both as catalyst for chemodynamic therapy and immune adjuvant for antitumor immune responses. In Mn-mediated Fenton-like reaction, bicarbonate ([Formula: see text]), as the most important component to amplify therapeutic effects, must be present, however, intracellular [Formula: see text] is strictly limited because of the tight control by live cells. RESULTS: Herein, Stimuli-responsive manganese carbonate-indocyanine green complexes (MnCO3-ICG) were designed for intracellular marriage of bicarbonate and Mn ions as "immune ion reactors" to regulate intracellular redox homeostasis and antitumor immune responses. Under the tumor acidic environment, the biodegradable complex can release "ion reactors" of Mn2+ and [Formula: see text], and ICG in the cytoplasm. The suddenly increased [Formula: see text] in situ inside the cells regulate intracellular pH, and accelerate the generation of hydroxyl radicals for the oxidative stress damage of tumors cells because [Formula: see text] play a critical role to catalyze Mn-mediated Fenton-like reaction. Investigations in vitro and in vivo prove that the both CDT and phototherapy combined with Mn2+-enhanced immunotherapy effectively suppress tumor growth and realize complete tumor elimination. CONCLUSIONS: The combination therapy strategy with the help of novel immune adjuvants would produce an enhanced immune response, and be used for the treatment of deep tumors in situ.

Carbonized paramagnetic complexes of Mn (II) as contrast agents for precise magnetic resonance imaging of sub-millimeter-sized orthotopic tumors.

Paramagnetic complexes containing gadolinium ions have been widely used for magnetic resonance imaging (MRI) in clinic. However, these paramagnetic complexes pose some safety concerns. There is still a demand for the development of stable MRI contrast agents that exhibit higher sensitivity and superior functionality to existing contrast agents. Here, we develop carbonized paramagnetic complexes of manganese (II) (Mn@CCs) to encapsulate Mn2+ in sealed carbonized shells with superhigh r1 relaxivity. Compared to the most common clinical contrast agent Magnevist, investigations in vivo demonstrate that the Mn@CCs cross the intact blood-brain barrier of normal health mice with minor metal deposition; preferentially target the glioma tissues distribute homogeneously with high penetration in an intracranial mouse model; delineate clear tumor margins in MRIs of ultrasmall single-nodule brain tumors, and multi-nodular liver tumors. The sensitivity, accuracy and low toxicity offer by Mn@CCs provides new opportunities for early molecular diagnostics and imaging-guided biomedical applications.

Oxygen-Independent Sulfate Radical for Stimuli-Responsive Tumor Nanotherapy.

Variant modalities are quested and merged into the tumor nanotherapy by leveraging the excitation from external or intratumoral incentives. However, the ubiquitous hypoxia and the insufficient content of hydrogen peroxide (H2 O2 ) in tumor microenvironments inevitably hinder the effective production of reactive oxygen species (ROS). To radically extricate from the shackles, peroxymonosulfate (PMS: HSO5- )-loaded hollow mesoporous copper sulfide (CuS) nanoparticles (NPs) are prepared as the distinct ROS donors for sulfate radical (•SO4- )-mediated and stimuli-responsive tumor nanotherapy in an oxygen-independent manner. In this therapeutic modality, the second near-infrared laser irradiation, together with the released copper ions as well as the heat produced by CuS after illumination, work together to activate PMS thus triply ensuring the copious production of •SO4- . Different from conventional ROS, the emergence of •SO4- , possessing a longer half-life and more rapid reaction, is independent of the oxygen (O2 ) and H2 O2 content within the tumor. In addition, this engineered nanosystem also exerts the function of photoacoustic imaging and skin restoration on the corresponding animal models. This study reveals the enormous potential of sulfate radical in oncotherapy and broadens pave for exploring the application of multifunctional and stimuli-responsive nanosystems in biomedicine.

Ion drugs for precise orthotopic tumor management by in situ the generation of toxic ion and drug pools.

Background: Asymmetric intracellular and extracellular ionic gradients are critical to the survivability of mammalian cells. Given the importance of manganese (Mn2+), calcium (Ca2+), and bicarbonate (HCO3-) ions, any alteration of the ion-content balance could induce a series of cellular responses. HCO3- plays an indispensable role for Mn-mediated Fenton-like reaction, but this is difficult to achieve because bicarbonates are tightly regulated by live cells, and are limited in anticancer efficacy. Methods: A responsive and biodegradable biomineral, Mn-doped calcium carbonate integrated with dexamethasone phosphate (DEX) (Mn:CaCO3-DEX), was reported to enable synergistic amplification of tumor oxidative stress, reduce inflammation, and induce Ca-overload cell apoptosis by elevating the intracellular and extracellular ionic gradients. Results: Under the acidic environment in tumor region, the ions (Mn2+, CO32-, Ca2+) were released by the degradation of Mn:CaCO3-DEX and then escalated oxidative stresses by triggering a HCO3--indispensable Mn-based Fenton-like reaction and breaking Ca2+ ion homeostasis to cause oxidative stress in cells and calcification. The released anti-inflammatory and antitumor drug, DEX, could alleviate the inflammatory environment. The investigations in vitro and in vivo demonstrated that the synergistic oncotherapy could effectively inhibit the growth of subcutaneous tumors and orthotopic liver tumors. Notably, normal cells showed greater tolerance of the synergistic influences. Conclusion: As an ion drug, Mn:CaCO3-DEX is an excellent potential diagnostic agent for precise orthotopic tumor management by the generation in situ of toxic ion and drug pools in the environment of tumor region, with synergistic effects of enhanced chemodynamic therapy, calcification, and anti-inflammation effects.

Biprediction-Based Video Quality Enhancement via Learning.

Convolutional neural networks (CNNs)-based video quality enhancement generally employs optical flow for pixelwise motion estimation and compensation, followed by utilizing motion-compensated frames and jointly exploring the spatiotemporal correlation across frames to facilitate the enhancement. This method, called the optical-flow-based method (OPT), usually achieves high accuracy at the expense of high computational complexity. In this article, we develop a new framework, referred to as biprediction-based multiframe video enhancement (PMVE), to achieve a one-pass enhancement procedure. PMVE designs two networks, that is, the prediction network (Pred-net) and the frame-fusion network (FF-net), to implement the two steps of synthesization and fusion, respectively. Specifically, the Pred-net leverages frame pairs to synthesize the so-called virtual frames (VFs) for those low-quality frames (LFs) through biprediction. Afterward, the slowly fused FF-net takes the VFs as the input to extract the correlation across the VFs and the related LFs, to obtain an enhanced version of those LFs. Such a framework allows PMVE to leverage the cross-correlation between successive frames for enhancement, hence capable of achieving high accuracy performance. Meanwhile, PMVE effectively avoids the explicit operations of motion estimation and compensation, hence greatly reducing the complexity compared to OPT. The experimental results demonstrate that the peak signal-to-noise ratio (PSNR) performance of PMVE is fully on par with that of OPT while its computational complexity is only 1% of OPT. Compared with other state-of-the-art methods in the literature, PMVE is also confirmed to achieve superior performance in both objective quality and visual quality at a reasonable complexity level. For instance, PMVE can surpass its best counterpart method by up to 0.42 dB in PSNR.

miR-193a as a potential mediator of WT-1/synaptopodin in the renoprotective effect of Losartan on diabetic kidney.

Diabetic nephropathy (DN) is the most common complication of diabetic patients, and has become a global healthcare problem. In this study, we used diabetic mice to evaluate the effect of Losartan on DN, in which the experimental animals were divided into three groups: non-diabetic mice (db/m group), untreated-diabetic mice (db/db group), and Losartan-treated diabetic mice (db/db-losartan). Next, immunohistochemistry and immunofluorescence were used to detect Wilms tumor protein 1 (WT-1) and synaptopodin expression, respectively. Protein levels of WT-1, synaptopodin, claudin1, and Pax-2 were assessed by Western blotting and real-time PCR. The miR-193a mRNA levels were quantitated by real-time PCR. The results showed that albuminuria was increased in diabetic mice compared with control animals and was significantly ameliorated by treatment with Losartan. In addition, Losartan significantly upregulated the immunopositive cell numbers of WT-1, the expression of WT-1 and synaptopodin in renal tissue. By contrast, expression of claudin1 and Pax-2 in renal tissue were decreased in db/db-losartan group. Besides, expression of miR-193a was decreased significantly in db/db-losartan group compared with the untreated diabetic group. Thus, Losartan has renoprotective effects on the control of tissue damage possibly by inhibiting the expression of miR-193a, thereby promoting the repair of podocyte injury in mice with DN.

Neural Reference Synthesis for Inter Frame Coding.

This work proposes the neural reference synthesis (NRS) to generate high-fidelity reference block for motion estimation and motion compensation (MEMC) in inter frame coding. The NRS is comprised of two submodules: one for reconstruction enhancement and the other for reference generation. Although numerous methods have been developed in the past for these two submodules using either handcrafted rules or deep convolutional neural network (CNN) models, they basically deal with them separately, resulting in limited coding gains. By contrast, the NRS proposes to optimize them collaboratively. It first develops two CNN-based models, namely EnhNet and GenNet. The EnhNet only uses spatial correlations within the current frame for reconstruction enhancement and the GenNet is then augmented by further aggregating temporal correlations across multiple frames for reference synthesis. However, a direct concatenation of EnhNet and GenNet without considering the complex temporal reference dependency across inter frames would implicitly induce iterative CNN processing and cause the data overfitting problem, leading to visually-disturbing artifacts and oversmoothed pixels. To tackle this problem, the NRS applies a new training strategy to coordinate the EnhNet and GenNet for more robust and generalizable models, and also devises a lightweight multi-level R-D (rate-distortion) selection policy for the encoder to adaptively choose reference blocks generated from the proposed NRS model or conventional coding process. Our NRS not only offers state-of-the-art coding gains, e.g., >10% BD-Rate (Bjøntegaard Delta Rate) reduction against the High Efficiency Video Coding (HEVC) anchor for a variety of common test video sequences encoded at a wide bit range in both low-delay and random access settings, but also greatly reduces the complexity relative to existing learning-based methods by utilizing more lightweight DNNs. All models are made publicly accessible at https://github.com/IVC-Projects/NRS for reproducible research.