Comparison of Different Machine Learning Algorithms for the Prediction of the Wheat Grain Filling Stage Using RGB Images.

Grain filling is essential for wheat yield formation, but is very susceptible to environmental stresses, such as high temperatures, especially in the context of global climate change. Grain RGB images include rich color, shape, and texture information, which can explicitly reveal the dynamics of grain filling. However, it is still challenging to further quantitatively predict the days after anthesis (DAA) from grain RGB images to monitor grain development. RESULTS: The WheatGrain dataset revealed dynamic changes in color, shape, and texture traits during grain development. To predict the DAA from RGB images of wheat grains, we tested the performance of traditional machine learning, deep learning, and few-shot learning on this dataset. The results showed that Random Forest (RF) had the best accuracy of the traditional machine learning algorithms, but it was far less accurate than all deep learning algorithms. The precision and recall of the deep learning classification model using Vision Transformer (ViT) were the highest, 99.03% and 99.00%, respectively. In addition, few-shot learning could realize fine-grained image recognition for wheat grains, and it had a higher accuracy and recall rate in the case of 5-shot, which were 96.86% and 96.67%, respectively. MATERIALS AND METHODS: In this work, we proposed a complete wheat grain dataset, WheatGrain, which covers thousands of wheat grain images from 6 DAA to 39 DAA, which can characterize the complete dynamics of grain development. At the same time, we built different algorithms to predict the DAA, including traditional machine learning, deep learning, and few-shot learning, in this dataset, and evaluated the performance of all models. CONCLUSIONS: To obtain wheat grain filling dynamics promptly, this study proposed an RGB dataset for the whole growth period of grain development. In addition, detailed comparisons were conducted between traditional machine learning, deep learning, and few-shot learning, which provided the possibility of recognizing the DAA of the grain timely. These results revealed that the ViT could improve the performance of deep learning in predicting the DAA, while few-shot learning could reduce the need for a number of datasets. This work provides a new approach to monitoring wheat grain filling dynamics, and it is beneficial for disaster prevention and improvement of wheat production.

Comparison of bivalirudin versus heparin in adult extracorporeal membrane oxygenation anticoagulant therapy: A retrospective case-control study.

INTRODUCTION: There were controversial opinions on the use of bivalirudin versus heparin for anticoagulant therapy in extracorporeal membrane oxygenation. The aim of our present study is to evaluate the efficacy and safety of bivalirudin versus heparin for the maintenance of systemic anticoagulation during adult veno-venous extracorporeal membrane oxygenation (V-V ECMO). METHODS: Adult patients who received V-V ECMO support in our center between February 2018and February 2022 were retrospectively recruited. We analyzed their ECMO support time, platelet count, coagulation indicators, blood product infusion volume, the incidence of thrombosis and bleeding, probability of successful weaning of ECMO, and in-hospital mortality. RESULTS: A total of 58 patients received V-V ECMO support. Thirty-four patients were finally included according to the exclusion and inclusion criteria, 14 and 20 accepted bivalirudin and heparin for anticoagulant therapy, respectively. The Minimum platelet value (98.50 × 109/L (85.50, 123.75) vs 49.50 × 109/L (31.25, 83.00), p = 0.002) and mean platelet value (149.90 × 109/L (127.40, 164.80) vs 74.55 × 109/L (62.45, 131.60), p = 0.03) and the ratio of successful weaning of ECMO (92.8% vs 60.0%, p = 0.033) in bivalirudin group were significantly higher than those in heparin group. The red blood cell infusion volume (7.00 U (3.00, 13.25) vs 13.75 U (7.25, 22.63), p = 0.039), platelet infusion volume (0.00 mL (0.00, 75.00) vs 300 mL (0.00, 825.00), p = 0.027), and the incidence of major bleeding (0.00% vs 30%, p = 0.024) in bivalirudin group were significantly lower than those in heparin group. CONCLUSIONS: In V-V ECMO-supported adult patients, systemic anticoagulation with bivalirudin has achieved the same anticoagulation targets as heparin with less frequency of major bleeding events and lower requirement for blood products without significantly increased risk of thrombosis. Bivalirudin most likely is a safe and effective anticoagulation method for adult patients supported by V-V ECMO.

Simultaneous Prediction of Wheat Yield and Grain Protein Content Using Multitask Deep Learning from Time-Series Proximal Sensing.

Wheat yield and grain protein content (GPC) are two main optimization targets for breeding and cultivation. Remote sensing provides nondestructive and early predictions of yield and GPC, respectively. However, whether it is possible to simultaneously predict yield and GPC in one model and the accuracy and influencing factors are still unclear. In this study, we made a systematic comparison of different deep learning models in terms of data fusion, time-series feature extraction, and multitask learning. The results showed that time-series data fusion significantly improved yield and GPC prediction accuracy with R 2 values of 0.817 and 0.809. Multitask learning achieved simultaneous prediction of yield and GPC with comparable accuracy to the single-task model. We further proposed a two-to-two model that combines data fusion (two kinds of data sources for input) and multitask learning (two outputs) and compared different feature extraction layers, including RNN (recurrent neural network), LSTM (long short-term memory), CNN (convolutional neural network), and attention module. The two-to-two model with the attention module achieved the best prediction accuracy for yield (R 2 = 0.833) and GPC (R 2 = 0.846). The temporal distribution of feature importance was visualized based on the attention feature values. Although the temporal patterns of structural traits and spectral traits were inconsistent, the overall importance of both structural traits and spectral traits at the postanthesis stage was more important than that at the preanthesis stage. This study provides new insights into the simultaneous prediction of yield and GPC using deep learning from time-series proximal sensing, which may contribute to the accurate and efficient predictions of agricultural production.

Exploring the Multifunctional Roles of Odontoglossum Ringspot Virus P126 in Facilitating Cymbidium Mosaic Virus Cell-to-Cell Movement during Mixed Infection.

Synergistic interactions among viruses, hosts and/or transmission vectors during mixed infection can alter viral titers, symptom severity or host range. Viral suppressors of RNA silencing (VSRs) are considered one of such factors contributing to synergistic responses. Odontoglossum ringspot virus (ORSV) and cymbidium mosaic virus (CymMV), which are two of the most significant orchid viruses, exhibit synergistic symptom intensification in Phalaenopsis orchids with unilaterally enhanced CymMV movement by ORSV. In order to reveal the underlying mechanisms, we generated infectious cDNA clones of ORSV and CymMV isolated from Phalaenopsis that exerted similar unilateral synergism in both Phalaenopsis orchid and Nicotiana benthamiana. Moreover, we show that the ORSV replicase P126 is a VSR. Mutagenesis analysis revealed that mutation of the methionine in the carboxyl terminus of ORSV P126 abolished ORSV replication even though some P126 mutants preserved VSR activity, indicating that the VSR function of P126 alone is not sufficient for viral replication. Thus, P126 functions in both ORSV replication and as a VSR. Furthermore, P126 expression enhanced cell-to-cell movement and viral titers of CymMV in infected Phalaenopsis flowers and N. benthamiana leaves. Taking together, both the VSR and protein function of P126 might be prerequisites for unilaterally enhancing CymMV cell-to-cell movement by ORSV.

An Integrated Method for Tracking and Monitoring Stomata Dynamics from Microscope Videos.

Patchy stomata are a common and characteristic phenomenon in plants. Understanding and studying the regulation mechanism of patchy stomata are of great significance to further supplement and improve the stomatal theory. Currently, the common methods for stomatal behavior observation are based on static images, which makes it difficult to reflect dynamic changes of stomata. With the rapid development of portable microscopes and computer vision algorithms, it brings new chances for stomatal movement observation. In this study, a stomatal behavior observation system (SBOS) was proposed for real-time observation and automatic analysis of each single stoma in wheat leaf using object tracking and semantic segmentation methods. The SBOS includes two modules: the real-time observation module and the automatic analysis module. The real-time observation module can shoot videos of stomatal dynamic changes. In the automatic analysis module, object tracking locates every single stoma accurately to obtain stomatal pictures arranged in time-series; semantic segmentation can precisely quantify the stomatal opening area (SOA), with a mean pixel accuracy (MPA) of 0.8305 and a mean intersection over union (MIoU) of 0.5590 in the testing set. Moreover, we designed a graphical user interface (GUI) so that researchers could use this automatic analysis module smoothly. To verify the performance of the SBOS, the dynamic changes of stomata were observed and analyzed under chilling. Finally, we analyzed the correlation between gas exchange and SOA under drought stress, and the correlation coefficients between mean SOA and net photosynthetic rate (Pn), intercellular CO2 concentration (Ci), stomatal conductance (Gs), and transpiration rate (Tr) are 0.93, 0.96, 0.96, and 0.97.

Focal Adhesion Kinase Inhibitor Inhibits the Oxidative Damage Induced by Central Venous Catheter via Abolishing Focal Adhesion Kinase-Protein Kinase B Pathway Activation.

The vascular injury induced by central venous catheter (CVC) indwelling is the basis for the occurrence and development of CVC-related complications, such as phlebitis, venous thrombosis, and catheter-related infections. Focal adhesion kinase (FAK) and FAK-protein kinase B (AKT) signaling pathway are of great significance in tissue repair after trauma. Here, we investigated the role and mechanism of the FAK inhibitor (1,2,4,5-phenyltetramine tetrahydrochloride (Y15)) in oxidative damage caused by CVC. EA.hy926 cells were divided into the control group (normal control), CVCs+scratches group (the intercepted CVC segments coculturing with scratched EA.hy926 cells), and CVCs+scratches+Y15 group (Y15 was added to the cell culture supernatant with CVCs + scratches at a final concentration of 50 µmol·L-1). New Zealand rabbits were randomly divided into the control group (normal control), CVC group (CVC was inserted through the rabbit's right jugular vein to the junction of the right atrium and superior vena cava), and CVC+Y15 group (CVC was immersed in a 50 µmol·L-1 Y15 solutions before insertion). The levels of markers and proteins related to oxidative damage in cells, cell culture supernatant, serum, and external jugular vein were measured by commercial kits and western blot, respectively. We found that Y15 treatment significantly decreased ROS and MDA levels and increased cell viability, NO, and SOD levels in a time-dependent manner in rabbit serum and cell culture supernatant. In addition, Y15 effectively reduced the CVC-induced pathological changes of damaged vascular tissues. Y15 also downregulated the levels of p-FAK Tyr 397 and p-Akt Ser 473 in damaged external jugular vein and EA.hy926 cells. These findings suggest that Y15 alleviated CVC-induced oxidative damage to blood vessels by suppressing focal FAK-Akt pathway activation.

miR-155 Regulates claudin1 Expression in Humans With Intestinal Mucosa Dysfunction After Brain Injury.

Patients with craniocerebral trauma often have intestinal mucosal dysfunction, and the claudin1 protein plays an important role in intestinal mucosal function. Our previous work has shown that the expression of microRNA-155 (miR-155) in the peripheral blood of patients with craniocerebral trauma is decreased. Animal experiments also suggest that the expression of miR-155 is increased in the intestinal mucosa of mice with brain injury and the expression of claudin1 is decreased. We recruited 56 samples (35 patients with traumatic brain injury [TBI] and 21 patients without history of head trauma) to detect the expression of miR-155 on claudin1 regulation by quantitative polymerase chain reaction, reverse transcriptase polymerase chain reaction, and so on. We also used the receiver operating characteristic curve (ROC) to further evaluate the diagnostic value of the 2 biomarkers. From the results, we found that the expression level of miR-155 and claudin1 in the case group was lower than that in the control group. Human miR-155 (Hsa-miR-155) may positively regulate intestinal mucosal function by inhibiting the expression of claudin1, leading to intestinal mucosal barrier dysfunction. Combining the ROC curve data, the results further prove that miR-155 and claudin1 might be the new clinical diagnostic markers and treatment targets for the intestinal mucosal barrier dysfunction after TBI.

Intestinal barrier dysfunction following traumatic brain injury.

Traumatic brain injury (TBI) can cause non-neurological injuries to other organs such as the intestine. Newer studies have shown that paracellular hyperpermeability is the basis of intestinal barrier dysfunction following TBI. Ischemia-reperfusion injury, inflammatory response, abnormal release of neurotransmitters and hormones, and malnutrition contribute to TBI-induced intestinal barrier dysfunction. Several interventions that may protect intestinal barrier function and promote the recovery of TBI have been proposed, but relevant studies are still limited. This review is to clarify the established mechanisms of intestinal barrier dysfunction following TBI and to describe the possible strategies to reduce or prevent intestinal barrier dysfunction.

Enteral nutrition feeding in Chinese intensive care units: a cross-sectional study involving 116 hospitals.

BACKGROUND: There is a lack of large-scale epidemiological data on the clinical practice of enteral nutrition (EN) feeding in China. This study aimed to provide such data on Chinese hospitals and to investigate factors associated with EN delivery. METHODS: This cross-sectional study was launched in 118 intensive care units (ICUs) of 116 mainland hospitals and conducted on April 26, 2017. At 00:00 on April 26, all patients in these ICUs were included. Demographic and clinical variables of patients on April 25 were obtained. The dates of hospitalization, ICU admission and nutrition initiation were reviewed. The outcome status 28 days after the day of investigation was obtained. RESULTS: A total of 1953 patients were included for analysis, including 1483 survivors and 312 nonsurvivors. The median study day was day 7 (IQR 2-19 days) after ICU entry. The proportions of subjects starting EN within 24, 48 and 72 h after ICU entry was 24.8% (84/352), 32.7% (150/459) and 40.0% (200/541), respectively. The proportion of subjects receiving > 80% estimated energy target within 24, 48, 72 h and 7 days after ICU entry was 10.5% (37/352), 10.9% (50/459), 11.8% (64/541) and 17.8% (162/910), respectively. Using acute gastrointestinal injury (AGI) 1 as the reference in a Cox model, patients with AGI 2-3 were associated with reduced likelihood of EN initiation (HR 0.46, 95% CI 0.353-0.599; p < 0.001). AGI 4 was significantly associated with lower hazard of EN administration (HR 0.056; 95% CI 0.008-0.398; p = 0.004). In a linear regression model, greater Sequential Organ Failure Assessment scores (coefficient - 0.002, 95% CI - 0.008 to - 0.001; p = 0.024) and male gender (coefficient - 0.144, 95% CI - 0.203 to - 0.085; p < 0.001) were found to be associated with lower EN proportion. As compared with AGI 1, AGI 2-3 was associated with lower EN proportion (coefficient - 0.206, 95% CI - 0.273 to - 0.139; p < 0.001). CONCLUSIONS: The study showed that EN delivery was suboptimal in Chinese ICUs. More attention should be paid to EN use in the early days after ICU admission.

MicroRNA­34a induces apoptosis in PC12 cells by reducing B­cell lymphoma 2 and sirtuin­1 expression.

MicroRNA­34a (miR­34a) is a direct target of p53 and was reported to induce cell cycle arrest, apoptosis and senescence. Inhibition of the NAD­dependent deacetylase sirtuin­1 (SIRT1) by miR­34a leads to an increase in acetylated p53, which promotes cell apoptosis. B­cell lymphoma 2 (Bcl­2) is also involved in apoptosis, and was originally characterized with respect to its role in controlling outer mitochondrial membrane integrity. The effect of miR­34a in PC12 cells has not yet been reported. In the present study, it was hypothesized that Bcl­2 and SIRT1 may be critical downstream targets of miR­34a that participate in apoptosis induction. miR­34a mimics and inhibitors were transfected into PC12 cells, and the apoptosis and proliferation rates were compared between groups. It was demonstrated that induction of miR­34a promotes apoptosis and senescence, inhibits proliferation, and leads to marked alterations in SIRT1, Bcl­12 and acetyl (ac)­p53 expression. These data indicate that miR­34a may be important in neuropathy.

Simvastatin attenuates TNF­α­induced apoptosis in endothelial progenitor cells via the upregulation of SIRT1.

Endothelial progenitor cells (EPCs) originate from the bone marrow and can be classified as either early or late EPCs. The focus of this study was on late EPCs, as they play an important role in angiogenesis and vascular proliferation. Evidence suggests that inflammatory and oxidative changes can increase EPC apoptosis. Of note, tumor necrosis factor-α (TNF-α) is a contributing risk factor to the development of atherosclerosis and plays a key role as both an inflammatory mediator and an inducer of apoptosis in endothelial cells. Additionally, a member of the sirtuin family, silent information regulator type-1 (SIRT1), promotes cell survival by repressing p53- and non-p53-dependent apoptosis in response to DNA damage and oxidative stress. Statins have also been shown to play a key role in the prevention of endothelial apoptosis and senescence via their lipid-lowering and anti-inflammatory actions. However, there is little evidence that statins themselves attenuate EPC apoptosis induced by TNF-α. The aim of this study was to demonstrate the effectiveness of one of the most commonly used statins, simvastatin, on decreasing TNF-α-induced apoptosis in EPCs. The results indicated that SIRT1 protein expression was decreased by TNF-α in a time- and dose-dependent manner and that while TNF-α caused a marked increase in the percentage of apoptotic EPCs, application of simvastatin decreased this percentage. A high concentration of simvastatin promoted the expression of SIRT1 and increased the proliferation of EPCs. In conclusion, findings of this study showed that simvastatin is crucial in counteracting the TNF-α-induced apoptosis of EPCs and that this protection may involve the actions of SIRT1.

[The study of AutoFlow role in volume controlled ventilation].

OBJECTIVE: To evaluate the efficacy and safety of automatic variable flow rate (AutoFlow) for volume control ventilation through monitoring the number of ventilator alarm. METHODS: Forty-eight adult patients receiving the Drager Evita 4 ventilator with an expectation of more than 2 days duration were divided into two groups by randomly digital methods, each n=24. The patients in control group were received routinely mode with synchronized intermittent mandatory ventilation (SIMV), and the others in observation group were given SIMV and assist with AutoFlow. The midazolam and fentanyl was given to retain the Ramsay score 2-3 by continuous micro-pump. The ventilator alarm, blood gas analysis and respiratory function were recorded. RESULTS: There were no significant differences in respiratory rate (RR), tidal volume (VT), positive end-expiratory pressure (PEEP), fraction of inspired oxygen (FiO2), pH, arterial partial pressure of carbon dioxide (PaCO2), arterial partial pressure of oxygen (PaO2), oxygenation index (PaO2/FiO2), as well as sedative dose and time between two groups within 5 days of mechanical ventilation. Duration of mechanical ventilation in all patients was 164 days (3756 hours), and 78 days (1812 hours) in control group, 86 days (1944 hours) in observation group. The duration of mechanical ventilation in observation group was longer than that in control group [3 (1-15) days vs. 2 (1-28) days, P>0.05]. A total of 23 843 alarms were recorded, approximately 6 times/h, and 17 386 alarms in control group, averagely 9.6 times/h, 6457 alarms in the observation group, averagely 3.3 times/h. The number of ventilator alarm in observation group was less than that in control group (P<0.01). The number of airway pressure alarm in observation group was less than that in the control group [122 (8-1068) vs. 565 (13-1898), P<0.01]. There was no significant difference in sequential organ failure assessment (SOFA) score within 5 days between the two sets of mechanical ventilation. In the observation group ventilator-associated pneumonia (VAP) was occurred in 4 cases, and no pneumothorax happened, while in the control group there were 8 cases and 2 cases respectively. The mortality rate in intensive care unit (ICU) in observation group was lower than that in control group, but there was no statistical difference (25.0% vs. 37.5%, P>0.05). CONCLUSIONS: AutoFlow is confirmed be safe for volume control ventilation mode, and could significantly reduce the alarm of ventilator.