Search

Endoscopic surgery is currently an important treatment method in the field of spinal surgery and avoiding damage to the spinal nerves through video guidance is a key challenge. This paper presents the first real-time segmentation method for spinal nerves in endoscopic surgery, which provides crucial navigational information for surgeons. A finely annotated segmentation dataset of approximately 10,000 consec-utive frames recorded during surgery is constructed for the first time for this field, addressing the problem of semantic segmentation. Based on this dataset, we propose FUnet (Frame-Unet), which achieves state-of-the-art performance by utilizing inter-frame information and self-attention mechanisms. We also conduct extended exper-iments on a similar polyp endoscopy video dataset and show that the model has good generalization ability with advantageous performance. The dataset and code of this work are presented at: https://github.com/zzzzzzpc/FUnet ., Comment: Accepted by MICCAI 2023

Recently, Dense Retrieval (DR) has become a promising solution to document retrieval, where document representations are used to perform effective and efficient semantic search. However, DR remains challenging on long documents, due to the quadratic complexity of its Transformer-based encoder and the finite capacity of a low-dimension embedding. Current DR models use suboptimal strategies such as truncating or splitting-and-pooling to long documents leading to poor utilization of whole document information. In this work, to tackle this problem, we propose Segment representation learning for long documents Dense Retrieval (SeDR). In SeDR, Segment-Interaction Transformer is proposed to encode long documents into document-aware and segment-sensitive representations, while it holds the complexity of splitting-and-pooling and outperforms other segment-interaction patterns on DR. Since GPU memory requirements for long document encoding causes insufficient negatives for DR training, Late-Cache Negative is further proposed to provide additional cache negatives for optimizing representation learning. Experiments on MS MARCO and TREC-DL datasets show that SeDR achieves superior performance among DR models, and confirm the effectiveness of SeDR on long document retrieval.

Distant metastases occur when non-small cell lung cancer (NSCLC) is at late stages. Bone metastasis is one of the most frequent metastases of NSCLC and leads to poor prognosis. It has been reported that high expression of BMP2 in NSCLC correlates with poor survival, but whether BMP2 contributes to NSCLC bone metastasis remains largely unknown. The activation of BMP signalling is found in metastatic bone tumours of mice Lewis lung carcinoma and predicts poor survival in human NSCLC. BMP2 signalling activation can enhance bone metastasis of Lewis lung carcinoma. Moreover, BMP2 secreted by stroma fibroblasts can promote the migration and invasion of NSCLC cells. Besides, in combination with pre-osteoblast and LLCs, BMP2 could enhance the differentiation of macrophages into osteoclasts to play roles in the osteolytic mechanism of NSCLC bone metastasis. Interestingly, NSCLC cells can also enrich BMP2 to pre-osteoblasts to function in the osteoblastic mechanism. Our results firstly demonstrate the detailed mechanisms about what roles BMP2 signalling play in enhancing NSCLC bone metastases. These findings provide a new potential therapy choice for preventing bone metastases of NSCLC via the inhibition of BMP2 signalling.

Epilepsy is a neurological disorder with recurrent seizures of complexity and randomness. Until now, the mechanism of epileptic randomness has not been fully elucidated. Inspired by the recent finding that astrocyte GTPase-activating protein (G-protein)-coupled receptors could be involved in stochastic epileptic seizures, we proposed a neuron-astrocyte network model, incorporating the noise of the astrocytic second messager, inositol triphosphate (IP3) which is modulated by the G-protein)-coupled receptor activation. Based on this model, we have statistically analysed the transitions of epileptic seizures by performing tens of simulation trials. Our simulation results show that the increase of the IP3 noise intensity induces the depolarization-block epileptic seizures together with an increase in neuronal firing frequency. Meanwhile, a bistable state of neuronal firing emerges under certain noise intensity, during which the neuronal firing pattern switches between regular sparse spiking and epileptic seizure states. This random presence of epileptic seizures is absent when the noise intensity continues to increase, accompanying with an increase in the epileptic depolarization block duration. The simulation results also shed light on the fact that calcium signals in astrocytes play significant roles in the pattern formations of the epileptic seizure. Our results provide a potential pathway for understanding the epileptic randomness., Comment: 26 pages, 8 figures

Background: An unexpected dengue outbreak occurred in the Hunan Province in 2018. This is the first dengue outbreak in this area of inland China resulting 172 infected. Methods: To verify the causative agent of this outbreak and investigate gene characterization, the structural protein C/prM/E genes of viruses isolated from local residents were sequenced followed by mutation, phylogenetic analysis. The recombination, selection pressure, potential secondary structure and three-dimensional structure analysis were also performed. Results: Phylogenetic analysis revealed that all epidemic strains were classified as the cosmopolitan DENV-2 genotype, closest to the Zhejiang strain (MH010629, 2017) and then Malaysia strain (KJ806803, 2013). Compared with the DENV-2SS, 151 base substitutions were found in 89 sequences of isolates, resulting in 20 nonsynonymous mutations, of which 17 mutations existed among all samples (two in capsid protein, six in prM/M, and nine in envelope proteins). Moreover, amino acid substitutions at 602 th (E322:Q→H) and 670 th (E390: N→S) may result in heightened virulence of the epidemic strains. One new DNA-binding site and five new protein binding sites were observed. Two polynucleotide-binding sites and seven protein binding sites were lost compared with DENV-2SS. Meanwhile, five changes were found in helix regions. The helical transmembrane and disordered regions have minor changes. Protein tertiary structure prediction revealed the 429 th amino acid of E proteins was switch from histamine (positively charged) to asparagines (neutral) in 89 isolate strains. No recombination events or positive selection pressure sites were detected. To our knowledge, this study is the first gene analysis of epidemic strain in the first dengue outbreak in Hunan Province, inland China. Conclusions: The causative agent is likely to come from Zhejiang Province, a neighbouring Province where dengue fever broke out in 2017. This study may help understand the intrinsic geographical relatedness of DENV-2 and contributes further to research on pathogenicity and vaccine development.

Styrene (ST) is a common volatile petroleum-based diluent in epoxy-based vinyl ester (VE) resin system. In recent years, a number of renewable molecules, such as eugenol, have been methacrylated and products behaved lower volatility property than ST. However, these renewable resin systems exhibited decreased thermal performance, especially glass transition temperature (Tg). Therefore, these resin systems cannot be used in hash environment. In our work, we used isosorbide to react with methacrylic anhydride and the synthesized monomer was mixed with VE resin to produce new bio-based resin system-MIVE. Volatility, viscosity and thermomechanical properties of MIVE and VE/ST resin were examined. The results indicated that MIVE had excellent low volatility property and Tg of MIVE can achieve 210 oC. Furthermore, we discussed the approach to improve heat resistant property of resin system. Acrylates with vinyl groups, including hydroxyethyl methacrylate (HEMA), pentaerythritol triacrylate (PETA) and dipentaerythritol pentaacrylate (DPHA), was grafted to MIVE respectively. Dynamic thermomechanical analysis (DMA), heat distortion temperature (HDT) and thermogravimetric (TGA) tests were used to measure thermal performance of grafted resin systems. The results showed Tg of grafted MIVE ranged from 210 to 240 oC, HDT ranged from 160 to 230 oC and initial decomposition temperature (IDT) (5% weight loss) were over 400 oC. Overall, studies showed that grafting acrylates with vinyl groups to VE was an effective way to improve the heat resistant properties of bio-based resin systems.

A novel robust, anti-icing, self-cleaning superhydrophobic coating was prepared via a simple, environmentally, friendly and efficient strategy which bond nanoparticles SiO2 to the polyurethane surface under UV-curing. By controlling the content of nano-silica, a superhydrophobic coating with sliding angle (SA) of 2° and water contact angle (WCA) of 165° was constructed. The coating can be widely applied to various substrates with excellent anti-fouling properties and acid/alkali resistance. Meanwhile, the surface of coating is robust, can maintain superhydrophobic after 750 cycles of friction. In addition, the coating has a good anti-icing property which maintains excellent superhydrophobic after being placed at −10 °C for 100 h. Hence, the coating could be widely used on the surface of transmission lines and wind turbines.

Medical ultrasound imaging technology has a large scale practical applications in medical diagnostics. In medical ultrasonic synthetic aperture imaging, an array of sensors receive ultrasound echoes and process the echoes to gain low-resolution images which are further processed to obtain high-resolution images. Currently the most common processing algorithm to obtain low-resolution images is delay-and-sum method. This paper mainly focuses on implementing synthetic aperture algorithm in high-performance embedded platform and evaluating its performance. In the estimation of the ultrasonic synthetic aperture algorithm, Field II simulator was used to generate the needed digital ultrasound transducer data. The high-performance embedded computing platform with a graphics processing unit was used to build the synthetic aperture solution and gained an 85 × speedup as compared to its single-core embedded processor implementation. Furthermore, the embedded implementation framework we have built can be easily used to build other high-definition medical ultrasound imaging algorithms.

This paper presents a quick and highly efficient method to obtain soil moisture content by remote sensing. This method can obtain the moisture content information of a large area in short time. In order to improve the problem of the low accuracy of the model for monitoring soil moisture when using remote sensing, this paper’s method is based upon spot data of the soil moisture in the second branch irrigation canal of Baojixia. The data was obtained from October, 2007 to March, 2010. After determining the normalization for bands TM5, TM6 and TM7 of Landsat TM, a normalized soil moisture index was obtained, namely NDSMI= (TM5 -TM7+TM6)/ (TM5 +TM7+TM6). Then a monitoring model for remote sensing and soil moisture was created by using difference values between the normalized difference soil moisture index and a reference normalized soil moisture index. The research took a quantitative inversion of soil moisture based on an established model and remote sensing image as a sample on June 28, 2010 of second branch irrigation canal of Baojixia and Famen irrigation area. The results show that inversion accuracy reaches over 80%, and the best depth of soil layer is 0~40 cm layer.

The effects of mycophenolic acid (MPA) on high glucose-induced expression of transforming growth factor-beta (TGF-beta) and connective tissue growth factor (CTGF) in mesangial cells (MC) were investigated. Rat MC were cultured in the presence of different concentrations of MPA (1.0 and 10.0 micromol/L) or MPA plus high glucose for 72 h. The expression of TGF-beta and CTGF was detected by Western blot. The results showed that high glucose could induce the expression of TGF-beta and CTGF in MC, but MPA could inhibit this effects. MPA did not influence the expression of TGF-beta and CTGF in normal glucose. It was concluded that MPA might prevent the progression of diabetic nephropathy by inhibiting the expression of TGF-beta and CTGF in MC.