The existing action recognition methods are mainly based on clip-level classifiers such as two-stream CNNs or 3D CNNs, which are trained from the randomly selected clips and applied to densely ...sampled clips during testing. However, this standard setting might be suboptimal for training classifiers and also requires huge computational overhead when deployed in practice. To address these issues, we propose a new framework for action recognition in videos, called Dynamic Sampling Networks (DSN), by designing a dynamic sampling module to improve the discriminative power of learned clip-level classifiers and as well increase the inference efficiency during testing. Specifically, DSN is composed of a sampling module and a classification module, whose objective is to learn a sampling policy to on-the-fly select which clips to keep and train a clip-level classifier to perform action recognition based on these selected clips, respectively. In particular, given an input video, we train an observation network in an associative reinforcement learning setting to maximize the rewards of the selected clips with a correct prediction. We perform extensive experiments to study different aspects of the DSN framework on four action recognition datasets: UCF101, HMDB51, THUMOS14, and ActivityNet v1.3. The experimental results demonstrate that DSN is able to greatly improve the inference efficiency by only using less than half of the clips, which can still obtain a slightly better or comparable recognition accuracy to the state-of-the-art approaches.
Recently, porous hydrophobic/oleophilic materials (PHOMs) have been shown to be the most promising candidates for cleaning up oil spills; however, due to their limited absorption capacity, a large ...quantity of PHOMs would be consumed in oil spill remediation, causing serious economic problems. In addition, the complicated and time‐consuming process of oil recovery from these sorbents is also an obstacle to their practical application. To solve the above problems, we apply external pumping on PHOMs to realize the continuous collection of oil spills in situ from the water surface with high speed and efficiency. Based on this novel design, oil/water separation and oil collection can be simultaneously achieved in the remediation of oil spills, and the oil sorption capacity is no longer limited to the volume and weight of the sorption material. This novel external pumping technique may bring PHOMs a step closer to practical application in oil spill remediation.
Thirsty, thirsty PHOMs: The continuous collection of oil spills in situ from the surface of water can be accomplished through external pumping on porous hydrophobic/oleophilic materials (PHOMs). Based on this novel design, oil/water separation and oil collection can be simultaneously achieved, and the oil sorption capacity is no longer limited to the volume and weight of the sorption material.
Aggravated behaviors of hepatocellular carcinoma (HCC) will occur after inadequate thermal ablation. However, its underlying mechanisms are not fully understood. Here, we assessed whether the ...increased matrix stiffness after thermal ablation could promote the progression of residual HCC. Heat‐treated residual HCC cells were cultured on tailorable 3D gel with different matrix stiffness, simulating the changed physical environment after thermal ablation, and then the mechanical alterations of matrix stiffness on cell phenotypes were explored. Increased stiffness was found to significantly promote the proliferation of the heat‐treated residual HCC cells when the cells were cultured on stiffer versus soft supports, which was associated with stiffness‐dependent regulation of ERK phosphorylation. Heat‐exposed HCC cells cultured on stiffer supports showed enhanced motility. More importantly, vitamin K1 reduced stiffness‐dependent residual HCC cell proliferation by inhibiting ERK phosphorylation and suppressed the in vivo tumor growth, which was further enhanced by combining with sorafenib. Increased matrix stiffness promotes the progression of heat‐treated residual HCC cells, proposing a new mechanism of an altered biomechanical environment after thermal ablation accelerates HCC development. Vitamin K1 plus sorafenib can reverse this protumor effect.
Increased matrix stiffness after thermal ablation accelerates the progression of heat‐exposed residual HCC cells. Vitamin K1 and sorafenib disrupt the stiffness‐induced ERK activation to reverse the pro‐tumor effect as the potential therapeutic drugs. This new finding will help design strategy to treat the local tumor progression and prevent its rapid progression after RFA in the treatment of medium or large HCC.
Along with the development of high-throughput sequencing technologies, both sample size and SNP number are increasing rapidly in genome-wide association studies (GWAS), and the associated computation ...is more challenging than ever. Here, we present a memory-efficient, visualization-enhanced, and parallel-accelerated R package called “rMVP” to address the need for improved GWAS computation. rMVP can 1) effectively process large GWAS data, 2) rapidly evaluate population structure, 3) efficiently estimate variance components by Efficient Mixed-Model Association eXpedited (EMMAX), Factored Spectrally Transformed Linear Mixed Models (FaST-LMM), and Haseman-Elston (HE) regression algorithms, 4) implement parallel-accelerated association tests of markers using general linear model (GLM), mixed linear model (MLM), and fixed and random model circulating probability unification (FarmCPU) methods, 5) compute fast with a globally efficient design in the GWAS processes, and 6) generate various visualizations of GWAS-related information. Accelerated by block matrix multiplication strategy and multiple threads, the association test methods embedded in rMVP are significantly faster than PLINK, GEMMA, and FarmCPU_pkg. rMVP is freely available at https://github.com/xiaolei-lab/rMVP.
Background
Laparoscopic hepatectomy (LH) has been proposed as a safe and feasible treatment option for liver diseases. However, the short- and long-term outcomes of LH versus open hepatectomy (OH) ...for hepatocellular carcinoma (HCC) have not been adequately assessed. Thus, as another means of surgical therapy for hepatocellular carcinoma (HCC), we assessed the feasibility of performing LH as the standard procedure for disease in the left lateral lobe and peripheral right segments for HCC in selected patients.
Methods
Literature search included PubMed, Embase, Science Citation Index, SpringerLink, and secondary sources, from inception to March 2012, with no restrictions on languages or regions. The fixed-effects and random-effects models were used to measure the pooled estimates. The test of heterogeneity was performed by the Q statistic. Subgroup and sensitivity analyses were performed to explore heterogeneity between studies and to assess the effects of study quality.
Results
A total of 1238 patients (LH 485, OH 753) from 15 studies were included. The pooled odds ratios for postoperative morbidity and incidence of negative surgical margin in LH were found to be 0.37 (95 % confidence interval CI 0.27–0.52;
P
< 0.01) and 1.63 (95 % CI 0.82–3.22;
P
= 0.16), respectively, compared with OH. Blood loss was significantly decreased in the LH (weighted mean difference −224.63; 95 % CI −384.87 to −64.39;
P
= 0.006). No significant difference was observed between the both groups for long-term outcomes of overall survival and recurrence-free survival.
Conclusions
In patients with solitary left lateral lobe/right peripheral subcapsular tumors treated with minor resection, this meta-analysis demonstrated that compared to OH, LH may have short-term advantages in terms of blood loss and postoperative morbidity for HCC. Both procedures have similar long-term outcomes. It may be time to consider changing the standard procedures for treatment of HCC in the left lateral lobe and peripheral subcapsular right segments in selected patients.
Synaptic pruning during adolescence is important for appropriate neurodevelopment and synaptic plasticity. Aberrant synaptic pruning may underlie a variety of brain disorders such as schizophrenia, ...autism and anxiety. Dopamine D2 receptor (Drd2) is associated with several neuropsychiatric diseases and is the target of some antipsychotic drugs. Here we generate self-reporting Drd2 heterozygous (SR-Drd2
) rats to simultaneously visualize Drd2-positive neurons and downregulate Drd2 expression. Time course studies on the developing anterior cingulate cortex (ACC) from control and SR-Drd2
rats reveal important roles of Drd2 in regulating synaptic pruning rather than synapse formation. Drd2 also regulates LTD, a form of synaptic plasticity which includes some similar cellular/biochemical processes as synaptic pruning. We further demonstrate that Drd2 regulates synaptic pruning via cell-autonomous mechanisms involving activation of mTOR signaling. Deficits of Drd2-mediated synaptic pruning in the ACC during adolescence lead to hyper-glutamatergic function and anxiety-like behaviors in adulthood. Taken together, our results demonstrate important roles of Drd2 in cortical synaptic pruning.
Display omitted
•Optimal conditions for VFAs production from CM via AD was evaluated at 37 °C.•Highest net VFAs yield of 0.53 g-VFA/g-VS was achieved with heat shocking.•Acetate, propionate and ...butyrate were the dominant VFAs with ratio of 7:1.5:1.5.•Through a 16-day digestion period, maximum amount of VFAs can be achieved.•Alkaline pH was suggested as an effective promoter for VFAs production from CM.
Although the high nitrogen content of chicken manure (CM) poses major challenges for methane production through anaerobic digestion, on the bright side, it has a great potential for production of value-added intermediate products, such as volatile fatty acids (VFAs). However, in order to enhance VFAs yield, methane formation should be substantially suppressed. In the current research, individual and multiple effects of initial pH, heat-shock pretreatment, chemical methanogens inhibitor and the inoculum to substrate ratio (ISR) on optimization VFAs fermentation from CM were evaluated via batch assays. In this regard, the highest net VFAs yield, 0.53 g-VFA/g-VS, was achieved at conditions with heat-shocked inoculum and CM at ISR 1:6 and pH uncontrolled. Acetate dominated the VFAs mixture, accounting for up to 75% of total. Increased inoculum content enhanced the bioconversion efficiency to 78% at ISR 1:3. The study results suggest that alkalinity is a key promoter of VFAs production from CM.
Illustrations of the methanogenic pathways indicated by carbon isotope (δ13C) analysis under mesophilic and thermophilic conditions.
Display omitted
•The Methanogenesis pathway was quantitatively ...determined by 2–13C isotope.•SAO-HM account for 41% and 50% of methane formation at 37 °C and 55 °C.•The thermophilic digester had 15% lower methane yield rates.•Stable processes were achieved at free ammonia of 31 and 145 mg/L, respectively.
The impact of temperature on the anaerobic digestion of chicken manure was investigated by studying the process performance and pathway for continuously-fed digesters under mesophilic and thermophilic conditions. The mesophilic digester obtained a 15% higher methane yield compared with the thermophilic digester. Mesophilic and thermophilic digester had free ammonia of 31 and 145 mg/L, respectively. The stable carbon isotope analysis indicated that 41% and 50% of acetate was converted to methane through the syntrophic acetate oxidation and hydrogenotrophic methanogenesis (SAO-HM) pathway under mesophilic and thermophilic conditions, respectively. The genus Pseudomonas represented 10% and 16% under mesophilic and thermophilic conditions, respectively. A high abundance of the methanogens genus Methanoculleus (94% of total methanogens) in mesophilic and the genus Methanothermobacter (96%) in thermophilic digesters indicated they were the main hydrogenotrophic partners in SAO. The present study therefore illustrated that methanogenic pathway shifting, induced by free ammonia, closely correlated to the process performance.
•Proppant degradation declines productivity in coal seam gas extraction.•Proppant embedment is the most plausible degradation to occur in coal reservoirs.•Proppant embedment depends on the coal ...elastic modulus.•Elastic modulus depends on geological factors and on the coal rank.•Elastic modulus changes with interaction between water, pore fluid and methane.
Coal seam gas has attracted global interest due to its economic feasibility, high gas storage capacity, and its capacity to cater to the world energy crisis in the long run. One of the most effective ways of extracting coal seam gas is through the hydraulic fracturing process. For the success of the hydraulic fracturing process, the proppant performance plays a key role. Poor proppant performance is a common observation in the field, and studies on shale and siltstone reservoirs show that the primary damage mechanisms associated with hydraulic fracturing is proppant crushing, proppant diagenesis, and proppant embedment. However, coal material due to its softer rock surface is more prone to proppant embedment than the rest of the degradation mechanisms. Primary factors affecting the proppant behaviour in coal seam reservoirs are: 1) the implementation of the hydraulic fracturing process – this includes the proppant concentration, size, type of fracturing fluid used, and the proppant type selected. The selection of proppants is critical in the coal seam gas extraction process, where the softer coal rock surface restricts the use of high strength proppants due to its vulnerability of embedding into the surface. 2) coal properties – elastic modulus of coal is the most important factor deciding the proppant behaviour in the extraction process. The elastic modulus is affected by mainly geological factors (i.e. confining pressure, temperature, tectonic deformation etc.) and the maturity of coal. Moreover, we highlight the changes in elastic modulus of coal that are occurred due to the coal-water and coal-methane interaction, in which the effect of moisture on coal properties is significantly higher than methane interaction, for the reason that the coal matrix tends to chemically bind well with the water molecules. However, due to the heterogeneous nature of coal and the changing of reservoir conditions, we infer that influential factors must be re-evaluated for targeted seams, and thereby develop a suitable proppant implementation process.
Nitrate (NO3−) is a key component of secondary inorganic aerosols and PM2.5. However, the contributions of nitrogen oxides (NOx) emission sources to NO3− in PM2.5 remain poorly constrained. This ...study measured nitrogen (N) isotopes of NO3− (hereafter as δ15N-NO3-) in PM2.5 collected at Beijing in 2014. We observed that δ15N-NO3- values in PM2.5 (−2.3‰ − 19.7‰; 7.3 ± 5.4‰ annually) were significantly higher in winter (11.9 ± 4.4‰) than in summer (2.2 ± 2.5‰). The δ15N differences between source NOx and NO3− in PM2.5 (hereafter as Δ values) were estimated by a computation module as 7.8 ± 2.2‰ − 10.4 ± 1.6‰ (8.8 ± 2.4‰). Using the Δ values and δ15N values of NOx from major fossil (coal combustion, vehicle exhausts) and non-fossil sources (biomass burning, microbial N cycle), contributions of major NOx sources to NO3− in PM2.5 were further estimated by the SIAR model. We found that seasonal variations of δ15N-NO3- values in PM2.5 of Beijing were mainly caused by those of NOx contributions from coal combustion (38 ± 10% in winter, 20 ± 9% in summer). Annually, NOx from coal combustion, vehicle exhausts, biomass burning, and microbial N cycle contributed 28 ± 12%, 29 ± 17%, 27 ± 15%, and 16 ± 7% to NO3− in PM2.5, respectively, showing actually comparable contributions between non-fossil NOx (43 ± 16%) and fossil NOx (57 ± 21%). These results are useful for planning the reduction of NOx emissions in city environments and for elucidating relationships between regional NOx emissions and atmospheric NO3− pollution or deposition.
Display omitted
•δ15N-NO3- values of PM2.5 in Beijing were higher in winter than in summer.•Seasonal δ15N-NO3- variations of PM2.5 were mainly caused by those of NOx emission from coal combustion.•Fossil and non-fossil NOx sources contributed equally to NO3− of PM2.5 in Beijing.
Isotope partitioning of major nitrate sources in PM2.5
PDF
