Evolution of wheat blast resistance gene Rmg8 accompanied by differentiation of variants recognizing the powdery mildew fungus.

Wheat blast, a devastating disease having spread recently from South America to Asia and Africa, is caused by Pyricularia oryzae (synonym of Magnaporthe oryzae) pathotype Triticum, which first emerged in Brazil in 1985. Rmg8 and Rmg7, genes for resistance to wheat blast found in common wheat and tetraploid wheat, respectively, recognize the same avirulence gene, AVR-Rmg8. Here we show that an ancestral resistance gene, which had obtained an ability to recognize AVR-Rmg8 before the differentiation of Triticum and Aegilops, has expanded its target pathogens. Molecular cloning revealed that Rmg7 was an allele of Pm4, a gene for resistance to wheat powdery mildew on 2AL, whereas Rmg8 was its homoeologue on 2BL ineffective against wheat powdery mildew. Rmg8 variants with the ability to recognize AVR-Rmg8 were distributed not only in Triticum spp. but also in Aegilops speltoides, Aegilops umbellulata and Aegilops comosa. This result suggests that the origin of resistance gene(s) recognizing AVR-Rmg8 dates back to the time before differentiation of A, B, S, U and M genomes, that is, ~5 Myr before the emergence of its current target, the wheat blast fungus. Phylogenetic analyses suggested that, in the evolutionary process thereafter, some of their variants gained the ability to recognize the wheat powdery mildew fungus and evolved into genes controlling dual resistance to wheat powdery mildew and wheat blast.

Highly Efficient and Unsupervised Framework for Moving Object Detection in Satellite Videos.

Moving object detection in satellite videos (SVMOD) is a challenging task due to the extremely dim and small target characteristics. Current learning-based methods extract spatio-temporal information from multi-frame dense representation with labor-intensive manual labels to tackle SVMOD, which needs high annotation costs and contains tremendous computational redundancy due to the severe imbalance between foreground and background regions. In this paper, we propose a highly efficient unsupervised framework for SVMOD. Specifically, we propose a generic unsupervised framework for SVMOD, in which pseudo labels generated by a traditional method can evolve with the training process to promote detection performance. Furthermore, we propose a highly efficient and effective sparse convolutional anchor-free detection network by sampling the dense multi-frame image form into a sparse spatio-temporal point cloud representation and skipping the redundant computation on background regions. Coping these two designs, we can achieve both high efficiency (label and computation efficiency) and effectiveness. Extensive experiments demonstrate that our method can not only process 98.8 frames per second on 1024 ×1024 images but also achieve state-of-the-art performance.

Predicting prostate cancer grade reclassification on active surveillance using a deep Learning-Based grading algorithm.

Deep learning (DL)-based algorithms to determine prostate cancer (PCa) Grade Group (GG) on biopsy slides have not been validated by comparison to clinical outcomes. We used a DL-based algorithm, AIRAProstate, to re-grade initial prostate biopsies in two independent PCa active surveillance (AS) cohorts. In a cohort initially diagnosed with GG1 PCa using only systematic biopsies (n = 138), upgrading of the initial biopsy to ≥GG2 by AIRAProstate was associated with rapid or extreme grade reclassification on AS (odds ratio 3.3, p = .04), whereas upgrading of the initial biopsy by contemporary uropathologist reviews was not associated with this outcome. In a contemporary validation cohort that underwent prostate magnetic resonance imaging before initial biopsy (n = 169), upgrading of the initial biopsy (all contemporary GG1 by uropathologist grading) by AIRAProstate was associated with grade reclassification on AS (hazard ratio 1.7, p = .03). These results demonstrate the utility of a DL-based grading algorithm in PCa risk stratification for AS.

Advancing Real-World Image Dehazing: Perspective, Modules, and Training.

Restoring high-quality images from degraded hazy observations is a fundamental and essential task in the field of computer vision. While deep models have achieved significant success with synthetic data, their effectiveness in real-world scenarios remains uncertain. To improve adaptability in real-world environments, we construct an entirely new computational framework by making efforts from three key aspects: imaging perspective, structural modules, and training strategies. To simulate the often-overlooked multiple degradation attributes found in real-world hazy images, we develop a new hazy imaging model that encapsulates multiple degraded factors, assisting in bridging the domain gap between synthetic and real-world image spaces. In contrast to existing approaches that primarily address the inverse imaging process, we design a new dehazing network following the "localization-and-removal" pipeline. The degradation localization module aims to assist in network capture discriminative haze-related feature information, and the degradation removal module focuses on eliminating dependencies between features by learning a weighting matrix of training samples, thereby avoiding spurious correlations of extracted features in existing deep methods. We also define a new Gaussian perceptual contrastive loss to further constrain the network to update in the direction of the natural dehazing. Regarding multiple full/no-reference image quality indicators and subjective visual effects on challenging RTTS, URHI, and Fattal real hazy datasets, the proposed method has superior performance and is better than the current state-of-the-art methods. See more results: https://github.com/fyxnl/KA Net.

Trends in Industry-Sponsored Research Payments to Physician Principal Investigators.

This cohort study investigates trends in total and per-physician industry-sponsored research payments to physician principal investigators from 2015 to 2022.

Enhancing Information Maximization With Distance-Aware Contrastive Learning for Source-Free Cross-Domain Few-Shot Learning.

Existing Cross-Domain Few-Shot Learning (CDFSL) methods require access to source domain data to train a model in the pre-training phase. However, due to increasing concerns about data privacy and the desire to reduce data transmission and training costs, it is necessary to develop a CDFSL solution without accessing source data. For this reason, this paper explores a Source-Free CDFSL (SF-CDFSL) problem, in which CDFSL is addressed through the use of existing pretrained models instead of training a model with source data, avoiding accessing source data. However, due to the lack of source data, we face two key challenges: effectively tackling CDFSL with limited labeled target samples, and the impossibility of addressing domain disparities by aligning source and target domain distributions. This paper proposes an Enhanced Information Maximization with Distance-Aware Contrastive Learning (IM-DCL) method to address these challenges. Firstly, we introduce the transductive mechanism for learning the query set. Secondly, information maximization (IM) is explored to map target samples into both individual certainty and global diversity predictions, helping the source model better fit the target data distribution. However, IM fails to learn the decision boundary of the target task. This motivates us to introduce a novel approach called Distance-Aware Contrastive Learning (DCL), in which we consider the entire feature set as both positive and negative sets, akin to Schrödinger's concept of a dual state. Instead of a rigid separation between positive and negative sets, we employ a weighted distance calculation among features to establish a soft classification of the positive and negative sets for the entire feature set. We explore three types of negative weights to enhance the performance of CDFSL. Furthermore, we address issues related to IM by incorporating contrastive constraints between object features and their corresponding positive and negative sets. Evaluations of the 4 datasets in the BSCD-FSL benchmark indicate that the proposed IM-DCL, without accessing the source domain, demonstrates superiority over existing methods, especially in the distant domain task. Additionally, the ablation study and performance analysis confirmed the ability of IM-DCL to handle SF-CDFSL. The code will be made public at https://github.com/xuhuali-mxj/IM-DCL.

Boosting Convolutional Neural Networks With Middle Spectrum Grouped Convolution.

This article proposes a novel module called middle spectrum grouped convolution (MSGC) for efficient deep convolutional neural networks (DCNNs) with the mechanism of grouped convolution. It explores the broad "middle spectrum" area between channel pruning and conventional grouped convolution. Compared with channel pruning, MSGC can retain most of the information from the input feature maps due to the group mechanism; compared with grouped convolution, MSGC benefits from the learnability, the core of channel pruning, for constructing its group topology, leading to better channel division. The middle spectrum area is unfolded along four dimensions: groupwise, layerwise, samplewise, and attentionwise, making it possible to reveal more powerful and interpretable structures. As a result, the proposed module acts as a booster that can reduce the computational cost of the host backbones for general image recognition with even improved predictive accuracy. For example, in the experiments on the ImageNet dataset for image classification, MSGC can reduce the multiply-accumulates (MACs) of ResNet-18 and ResNet-50 by half but still increase the Top-1 accuracy by more than 1% . With a 35% reduction of MACs, MSGC can also increase the Top-1 accuracy of the MobileNetV2 backbone. Results on the MS COCO dataset for object detection show similar observations. Our code and trained models are available at https://github.com/hellozhuo/msgc.

Radiomics Signature Based on Support Vector Machines for the Prediction of Pathological Complete Response to Neoadjuvant Chemoradiotherapy in Locally Advanced Rectal Cancer.

The objective of this study was to evaluate the discriminative capabilities of radiomics signatures derived from three distinct machine learning algorithms and to identify a robust radiomics signature capable of predicting pathological complete response (pCR) after neoadjuvant chemoradiotherapy in patients diagnosed with locally advanced rectal cancer (LARC). In a retrospective study, 211 LARC patients were consecutively enrolled and divided into a training cohort (n = 148) and a validation cohort (n = 63). From pretreatment contrast-enhanced planning CT images, a total of 851 radiomics features were extracted. Feature selection and radiomics score (Radscore) construction were performed using three different machine learning methods: least absolute shrinkage and selection operator (LASSO), random forest (RF) and support vector machine (SVM). The SVM-derived Radscore demonstrated a strong correlation with the pCR status, yielding area under the receiver operating characteristic curves (AUCs) of 0.880 and 0.830 in the training and validation cohorts, respectively, outperforming the RF and LASSO methods. Based on this, a nomogram was developed by combining the SVM-based Radscore with clinical indicators to predict pCR after neoadjuvant chemoradiotherapy. The nomogram exhibited superior predictive power, achieving AUCs of 0.910 and 0.866 in the training and validation cohorts, respectively. Calibration curves and decision curve analyses confirmed its appropriateness. The SVM-based Radscore demonstrated promising performance in predicting pCR for LARC patients. The machine learning-driven nomogram, which integrates the Radscore and clinical indicators, represents a valuable tool for predicting pCR in LARC patients.

Protective Effects and Mechanism of Polysaccharides from Edible Medicinal Plants in Alcoholic Liver Injury: A Review.

Alcohol use accounts for a large variety of diseases, among which alcoholic liver injury (ALI) poses a serious threat to human health. In order to overcome the limitations of chemotherapeutic agents, some natural constituents, especially polysaccharides from edible medicinal plants (PEMPs), have been applied for the prevention and treatment of ALI. In this review, the protective effects of PEMPs on acute, subacute, subchronic, and chronic ALI are summarized. The pathogenesis of alcoholic liver injury is analyzed. The structure-activity relationship (SAR) and safety of PEMPs are discussed. In addition, the mechanism underlying the hepatoprotective activity of polysaccharides from edible medicinal plants is explored. PEMPs with hepatoprotective activities mainly belong to the families Orchidaceae, Solanaceae, and Liliaceae. The possible mechanisms of PEMPs include activating enzymes related to alcohol metabolism, attenuating damage from oxidative stress, regulating cytokines, inhibiting the apoptosis of hepatocytes, improving mitochondrial function, and regulating the gut microbiota. Strategies for further research into the practical application of PEMPs for ALI are proposed. Future studies on the mechanism of action of PEMPs will need to focus more on the utilization of multi-omics approaches, such as proteomics, epigenomics, and lipidomics.

Lightweight Pixel Difference Networks for Efficient Visual Representation Learning.

Recently, there have been tremendous efforts in developing lightweight Deep Neural Networks (DNNs) with satisfactory accuracy, which can enable the ubiquitous deployment of DNNs in edge devices. The core challenge of developing compact and efficient DNNs lies in how to balance the competing goals of achieving high accuracy and high efficiency. In this paper we propose two novel types of convolutions, dubbed Pixel Difference Convolution (PDC) and Binary PDC (Bi-PDC) which enjoy the following benefits: capturing higher-order local differential information, computationally efficient, and able to be integrated with existing DNNs. With PDC and Bi-PDC, we further present two lightweight deep networks named Pixel Difference Networks (PiDiNet) and Binary PiDiNet (Bi-PiDiNet) respectively to learn highly efficient yet more accurate representations for visual tasks including edge detection and object recognition. Extensive experiments on popular datasets (BSDS500, ImageNet, LFW, YTF, etc.) show that PiDiNet and Bi-PiDiNet achieve the best accuracy-efficiency trade-off. For edge detection, PiDiNet is the first network that can be trained without ImageNet, and can achieve the human-level performance on BSDS500 at 100 FPS and with 1 M parameters. For object recognition, among existing Binary DNNs, Bi-PiDiNet achieves the best accuracy and a nearly 2× reduction of computational cost on ResNet18.

Dual-Function Smart Windows Using Polymer Stabilized Cholesteric Liquid Crystal Driven with Interdigitated Electrodes.

In this study, we present an electrically switchable window that can dynamically transmit both visible light and infrared (IR) light. The window is based on polymer stabilized cholesteric liquid crystals (PSCLCs), which are placed between a top plate electrode substrate and a bottom interdigitated electrode substrate. By applying a vertical alternating current electric field between the top and bottom substrates, the transmittance of the entire visible light can be adjusted. The cholesteric liquid crystals (CLC) texture will switch to a scattering focal conic state. The corresponding transmittance decreases from 90% to less than 15% in the whole visible region. The reflection bandwidth in the IR region can be tuned by applying an in-plane interdigital direct current (DC) electric field. The non-uniform distribution of the in-plane electric field will lead to helix pitch distortion of the CLC, resulting in a broadband reflection. The IR reflection bandwidth can be dynamically adjusted from 158 to 478 nm. The electric field strength can be varied to regulate both the transmittance in the visible range and the IR reflection bandwidth. After removing the electric field, both features can be restored to their initial states. This appealing feature of the window enables on-demand indoor light and heat management, making it a promising addition to the current smart windows available. This technology has considerable potential for practical applications in green buildings and automobiles.

Identification of immune and Toll-like receptor signaling pathway related feature lncRNAs to construct diagnostic nomograms for acute ischemic stroke.

We aimed to identify the immune and Toll-like receptor (TLR) signaling pathway related feature lncRNAs to construct the diagnostic nomograms for acute ischemic stroke (AIS). Two AIS-associated expression profiles GSE16561 and GSE22255 were downloaded from NCBI Gene Expression Omnibus, the former was the training set and the latter was the validation set. The differential expression genes (DEGs) and lncRNAs (DElncRNAs) related to TLR signaling pathway were identified between AIS and control groups. The single sample gene set enrichment analysis (ssGSEA) was applied to evaluate the immune infiltration. The immune and TLR signaling pathway related DElncRNAs were determined. Three optimization algorithms were utilized to select the immune and TLR signaling pathway related feature lncRNAs to construct the diagnostic nomograms of AIS. Based on the lncRNA signature, a ceRNA network was constructed. 37 DEGs and 28 DElncRNAs related to TLR signaling pathway were identified in GSE16561. 16 immune cell types exhibited significant differences in distribution between AIS and control groups. 28 immune and TLR signaling pathway related DElncRNAs were determined. 8 immune and TLR signaling pathway related feature lncRNAs were selected. The diagnostic nomograms of AIS performed well in both datasets. A ceRNA network was constructed consisting of 7 immune and TLR signaling pathway related feature lncRNAs as well as 19 AIS related miRNAs and 21 TLR signaling pathway related genes. LINC00173, LINC01089, LINC02210, MIR600HG, SNHG14, TP73-AS1, LINC00680 and CASC2 may be the potential biomarkers of AIS diagnosis, and TLR signaling pathway may be a promising immune related therapeutic target for AIS.

Environmental impact of single-use and reusable flexible cystoscopes.

OBJECTIVES: To compare the carbon footprint and environmental impact of single-use and reusable flexible cystoscopes. MATERIALS AND METHODS: We analysed the expected clinical lifecycle of single-use (Ambu aScope™ 4 Cysto) and reusable (Olympus CYF-V2) flexible cystoscopes, from manufacture to disposal. Performance data on cumulative procedures between repairs and before decommissioning were derived from a high-volume multispecialty practice. We estimated carbon expenditures per-case using published data on endoscope manufacturing, energy consumption during transportation and reprocessing, and solid waste disposal. RESULTS: A fleet of 16 reusable cystoscopes in service for up to 135 months averaged 207 cases between repairs and 3920 cases per lifecycle. Based on a manufacturing carbon footprint of 11.49 kg CO2 /kg device for reusable flexible endoscopes and 8.54 kg CO2 /kg device for single-use endoscopes, the per-case manufacturing cost was 1.37 kg CO2 for single-use devices and 0.0017 kg CO2 for reusable devices. The solid mass of single-use and reusable devices was 0.16 and 0.57 kg, respectively. For reusable devices, the energy consumption of reusable device reprocessing using an automated endoscope reprocessor was 0.20 kg CO2 , and per-case costs of device repackaging and repair were 0.005 and 0.02 kg CO2 , respectively. The total estimated per-case carbon footprint of single-use and reusable devices was 2.40 and 0.53 kg CO2 , respectively, favouring reusable devices. CONCLUSION: In this lifecycle analysis, the environmental impact of reusable flexible cystoscopes is markedly less than single-use cystoscopes. The primary contributor to the per-case carbon cost of reusable devices is energy consumption of reprocessing.

RobustFusion: Robust Volumetric Performance Reconstruction Under Human-Object Interactions From Monocular RGBD Stream.

High-quality 4D reconstruction of human performance with complex interactions to various objects is essential in real-world scenarios, which enables numerous immersive VR/AR applications. However, recent advances still fail to provide reliable performance reconstruction, suffering from challenging interaction patterns and severe occlusions, especially for the monocular setting. To fill this gap, in this paper, we propose RobustFusion, a robust volumetric performance reconstruction system for human-object interaction scenarios using only a single RGBD sensor, which combines various data-driven visual and interaction cues to handle the complex interaction patterns and severe occlusions. We propose a semantic-aware scene decoupling scheme to model the occlusions explicitly, with a segmentation refinement and robust object tracking to prevent disentanglement uncertainty and maintain temporal consistency. We further introduce a robust performance capture scheme with the aid of various data-driven cues, which not only enables re-initialization ability, but also models the complex human-object interaction patterns in a data-driven manner. To this end, we introduce a spatial relation prior to prevent implausible intersections, as well as data-driven interaction cues to maintain natural motions, especially for those regions under severe human-object occlusions. We also adopt an adaptive fusion scheme for temporally coherent human-object reconstruction with occlusion analysis and human parsing cue. Extensive experiments demonstrate the effectiveness of our approach to achieve high-quality 4D human performance reconstruction under complex human-object interactions whilst still maintaining the lightweight monocular setting.

Active Surveillance versus Immediate Intervention for Small Renal Masses: A Cost-Effectiveness and Clinical Decision Analysis.

PURPOSE: Active surveillance (AS) with the possibility of delayed intervention (DI) is emerging as a safe alternative to immediate intervention for many patients with small renal masses (SRMs). However, limited comparative data exist to inform the most appropriate management strategy for SRMs. MATERIALS AND METHODS: Decision analytic Markov modeling was performed to estimate the health outcomes and costs of 4 management strategies for 65-year-old patients with an incidental SRM: AS (with possible DI), immediate partial nephrectomy, radical nephrectomy, and thermal ablation. Mortality, direct medical costs, quality-adjusted life-years, and incremental cost-effectiveness ratios were evaluated over 10 years. RESULTS: The 10-year all-cause mortality was 22.6% for AS, 21.9% for immediate partial nephrectomy, 22.4% for immediate radical nephrectomy, and 23.7% for immediate thermal ablation. At a willingness-to-pay threshold of $100,000/quality-adjusted life-year, AS was the most cost-effective management strategy. The results were robust in univariate, multivariate, and probabilistic sensitivity analyses. Clinical decision analysis demonstrated that the tumor's metastatic potential, patient age, individual preferences, and health status were important factors influencing the optimal management strategy. Notably, if the annual probability of metastatic progression from AS was sufficiently low (under 0.35%-0.45% for most ages at baseline), consistent with the typical metastatic potential of SRMs <2 cm, AS would achieve higher health utilities than the other strategies. CONCLUSIONS: Compared to immediate intervention, AS with timely DI offers a safe and cost-effective approach to managing patients with SRMs. For patients harboring tumors of very low metastatic potential, AS may lead to better patient outcomes than immediate intervention.

Melatonin inhibits angiotensin II-induced atrial fibrillation through preventing degradation of Ang II Type I Receptor-Associated Protein (ATRAP).

Angiotensin II (Ang II) induced Atrial fibrillation (AF) often accompanied with reduced ATRAP which is a negative modulator of Ang II type 1 receptor (AT1R). Melatonin can protect against AF, but the underlying molecular mechanism remains poorly understood. In this study, Ang II was used to induce AF, and AF inducibility and duration were documented telemetrically. Ang II-infused mice had a higher AF incidence, which was associated with atrial fibrosis, inflammation, and oxidative stress. Melatonin partially inhibited these effects, and enforced expression of siRNA-ATRAP in atria counteracted the beneficial role of melatonin. Specifically, melatonin inhibited expression of Ang II-induced proteasome and immunoproteasome subunits ß2, ß2i, ß5, and ß5i as well as their corresponding trypsin-like and chymotrypsin-like activities and blocked ATRAP degradation. In turn, this inhibited AT1R-mediated NF-κB signaling, transforming growth factor (TGF)-ß1/Smad signaling in the atria, and thereby affected atrial remodeling and AF. Melatonin receptor inhibition by the chemical inhibitor luzindole partially inhibited the inhibitory effects of melatonin on proteasome activity and also Ang II-induced pathological changes in the atria. Overall, our study demonstrates that melatonin protects against Ang II-induced AF by inhibiting proteasome activity and stabilizing ATRAP expression, and these effects are partially dependent on melatonin receptor activation.

Peptide identification of hepatocyte growth-promoting factor and its function in cytoprotection and promotion of liver cell proliferation through the JAK2/STAT3/c-MYC pathway.

Hepatocyte growth-promoting factor (pHGF) has a significant effect in promoting liver cell proliferation and restoring liver function. In this study, 815 short peptides of pHGF were identified by liquid chromatography-tandem mass spectrometry (LC-MS/MS), of which 574 short peptides were assigned to 152 proteins related to hemoglobin subunits and some catalytic enzymes, indicating that pHGF might participate in the oxidation-reduction process by regulating reactive oxygen species (ROS) production. Proteomic analysis was used to identify the differentially expressed proteins (DEPs) in SMMC-7721 and L-02 cells after pHGF treatment, which suggested that pHGF had a significant impact on the JAK-STAT signaling pathway and the cell cycle of liver cells. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) and Western blot analysis revealed the mechanisms through which pHGF might activate the JAK2/STAT3/c-MYC pathway to up-regulate the expression of CDK4/6, thereby accelerating the G1/S transition to promote liver cell proliferation. These findings, for the first time, indicate the potential role of pHGF against the early or middle stages of acute, sub-acute, and chronic severe hepatitis. pHGF was also found to restore the reduced SOD1 and SOD2 protein levels that result from H2O2 exposure and significantly increase the HO-1 protein levels in L-02 cells, thus improving the viability of L-02 cells that have been damaged by H2O2 by reducing the ROS and lipid peroxidation levels.

Self-reported quality of life as a predictor of mortality in renal cell carcinoma.

BACKGROUND: This study evaluated the utility of self-reported quality of life (QOL) metrics in predicting mortality among all-comers with renal cell carcinoma (RCC) and externally tested the findings in a registry of patients with small renal masses. METHODS: The Surveillance, Epidemiology, and End Results-Medicare Health Outcomes Survey (SEER-MHOS) captured QOL metrics composed of mental component summary (MCS) and physical component summary (PCS) scores. Regression models assessed associations of MCS and PCS with all-cause, RCC-specific, and non-RCC-specific mortality. Harrell's concordance statistic (the C-index) and the Akaike information criterion (AIC) determined predictive accuracy and parsimony, respectively. Findings were tested in the prospective Delayed Intervention and Surveillance for Small Renal Masses (DISSRM) registry. RESULTS: In SEER-MHOS, 1494 patients had a median age of 73.4 years and a median follow-up time of 5.6 years. Each additional MCS and PCS point reduced the hazard of all-cause mortality by 1.3% (95% CI, 0.981-0.993; P < .001) and 2.3% (95% CI, 0.971-0.984; P < .001), respectively. Models with QOL metrics demonstrated higher predictive accuracy (C-index, 72.3% vs 70.1%) and parsimony (AIC, 9376.5 vs 9454.5) than models without QOL metrics. QOL metrics exerted a greater effect on non-RCC-specific mortality than RCC-specific mortality. External testing in the DISSRM registry confirmed these findings with similar results for all-cause mortality. CONCLUSIONS: Models with self-reported QOL metrics predicted all-cause mortality in patients with RCC with higher accuracy and parsimony than those without QOL metrics. Physical health was a stronger predictor of mortality than mental health. The findings support the incorporation of QOL metrics into prognostic models and patient counseling for RCC.