MEFA-Net: A mask enhanced feature aggregation network for polyp segmentation.

Accurate polyp segmentation is crucial for early diagnosis and treatment of colorectal cancer. This is a challenging task for three main reasons: (i) the problem of model overfitting and weak generalization due to the multi-center distribution of data; (ii) the problem of interclass ambiguity caused by motion blur and overexposure to endoscopic light; and (iii) the problem of intraclass inconsistency caused by the variety of morphologies and sizes of the same type of polyps. To address these challenges, we propose a new high-precision polyp segmentation framework, MEFA-Net, which consists of three modules, including the plug-and-play Mask Enhancement Module (MEG), Separable Path Attention Enhancement Module (SPAE), and Dynamic Global Attention Pool Module (DGAP).

No-reference stereoscopic image quality assessment based on binocular collaboration.

Stereoscopic images typically consist of left and right views along with depth information. Assessing the quality of stereoscopic/3D images (SIQA) is often more complex than that of 2D images due to scene disparities between the left and right views and the intricate process of fusion in binocular vision. To address the problem of quality prediction bias of multi-distortion images, we investigated the visual physiology and the processing of visual information by the primary visual cortex of the human brain and proposed a no-reference stereoscopic image quality evaluation method.

Revealing the Dark Side of Non-Local Attention in Single Image Super-Resolution.

Single Image Super-Resolution (SISR) aims to reconstruct a high-resolution image from its corresponding low-resolution input. A common technique to enhance the reconstruction quality is Non-Local Attention (NLA), which leverages self-similar texture patterns in images. However, we have made a novel finding that challenges the prevailing wisdom.

Nonintrusive and Effective Volume Reconstruction Model of Swimming Sturgeon Based on RGB-D Sensor.

The sturgeon is an important commercial aquaculture species in China. The measurement of sturgeon mass plays a remarkable role in aquaculture management. Furthermore, the measurement of sturgeon mass serves as a key phenotype, offering crucial information for enhancing growth traits through genetic improvement.

The development and nutritional quality of affected by monochromatic or mixed light provided by light-emitting diode.

Edible fungi has certain photo-sensitivity during the mushroom emergence stage, but there has been few relevant studies on the responses of to different light quality. were planted in growth chambers with different light qualities that were, respectively, white light (CK), monochromatic red light (R), monochromatic blue light (B), mixed red and blue light (RB), and the mixture of far-red and blue light (FrB). The photo-sensitivity of was investigated by analyzing the growth characteristics, nutritional quality, extracellular enzymes as well as the light photoreceptor genes in mushroom exposed to different light treatments.

Attributed Multi-Order Graph Convolutional Network for Heterogeneous Graphs.

Heterogeneous graph neural networks play a crucial role in discovering discriminative node embeddings and relations from multi-relational networks. One of the key challenges in heterogeneous graph learning lies in designing learnable meta-paths, which significantly impact the quality of learned embeddings. In this paper, we propose an Attributed Multi-Order Graph Convolutional Network (AMOGCN), which automatically explores meta-paths that involve multi-hop neighbors by aggregating multi-order adjacency matrices.

Effects of irrigation scheduling on the yield and irrigation water productivity of cucumber in coconut coir culture.

Optimum irrigation scheduling is important for ensuring high yield and water productivity in substrate-cultivated vegetables and is determined based on information such as substrate water content, meteorological parameters, and crop growth. The aim of this study was to determine a precise irrigation schedule for coconut coir culture in a solar greenhouse by comparing the irrigation, evapotranspiration (ET), substrate water content (VWC), as well as the crop growth indices and yield of cucumber, and irrigation water productivity (IWP) under three irrigation schedules: the soil moisture sensor-based method (T-VWC), the accumulated radiation combined with soil moisture sensor-based method (Rn-VWC), and the crop evapotranspiration estimated method using the hourly PM-ETo equation with an improved calculation of Kc (T-ETc). The results showed that the daily irrigation and evapotranspiration amount were the highest under T-VWC treatment, while the lowest under T-ETc treatment.

High-Similarity-Pass Attention for Single Image Super-Resolution.

Recent developments in the field of non-local attention (NLA) have led to a renewed interest in self-similarity-based single image super-resolution (SISR). Researchers usually use the NLA to explore non-local self-similarity (NSS) in SISR and achieve satisfactory reconstruction results. However, a surprising phenomenon that the reconstruction performance of the standard NLA is similar to that of the NLA with randomly selected regions prompted us to revisit NLA.

Antibody-based binding domain fused to TCRγ chain facilitates T cell cytotoxicity for potent anti-tumor response.

Chimeric antigen receptor T-cell (CAR-T) therapy has demonstrated potent clinical efficacy in the treatment of hematopoietic malignancies. However, the application of CAR-T in solid tumors has been limited due in part to the expression of inhibitory molecules in the tumor microenvironment, leading to T-cell exhaustion. To overcome this limitation, we have developed a synthetic T-cell receptor (TCR) that targets programmed death-ligand 1 (PD-L1), a molecule that is widely expressed in various solid tumors and plays a pivotal role in T-cell exhaustion.

AGNN: Alternating Graph-Regularized Neural Networks to Alleviate Over-Smoothing.

Graph convolutional network (GCN) with the powerful capacity to explore graph-structural data has gained noticeable success in recent years. Nonetheless, most of the existing GCN-based models suffer from the notorious over-smoothing issue, owing to which shallow networks are extensively adopted. This may be problematic for complex graph datasets because a deeper GCN should be beneficial to propagating information across remote neighbors.

Responses of butter leaf lettuce to mixed red and blue light with extended light/dark cycle period.

To investigate the effects of extended light/dark (L/D) cycle period (relative to the diurnal L/D cycle) on lettuce and explore potential advantages of abnormal L/D cycles, butter leaf lettuce were grown in a plant factory with artificial light (PFAL) and exposed to mixed red (R) and blue (B) LED light with different L/D cycles that were respectively 16 h light/8 h dark (L16/D8, as control), L24/D12, L48/D24, L96/D48 and L120/D60. The results showed that, all the abnormal L/D cycles increased shoot dry weight (DW) of lettuce (by 34-83%) compared with the control, and lettuce DW increased with the L/D cycle period prolonged. The contents of soluble sugar and crude fiber in lettuce showed an overall upward trend with the length of L/D cycle extended, and the highest vitamin C content as well as low nitrate content were both detected in lettuce treated with L120/D60.

100+ FPS detector of personal protective equipment for worker safety: A deep learning approach for green edge computing.

In industrial production, personal protective equipment (PPE) protects workers from accidental injuries. However, wearing PPE is not strictly enforced among workers due to all kinds of reasons. To enhance the monitoring of workers and thus avoid safety accidents, it is essential to design an automatic detection method for PPE.

Red and blue wavelengths affect the morphology, energy use efficiency and nutritional content of lettuce (Lactuca sativa L.).

Since red (R) and blue (B) LED light has different quantum efficiency and photoelectric conversion efficiency, mixed RB with different proportions of R and B results in varied energy consumption. In order to improve the energy use efficiency of the closed-type plant production systems, the effects of R and B proportions on the electric use efficiency (EUE), light use efficiency (LUE) as well as the quality of butter leaf lettuce were evaluated in this study. Lettuce seedlings were cultivated in a plant factory with artificial lighting (PFAL) and subjected to eleven combinations of R and B (100%R, 90%R, 80%R, 70%R, 60%R, 50%R, 40%R, 30%R, 20%R, 10%R, 0%R; the rest of the photons in each treatment were B) with the same total photosynthetic photon flux density (PPFD) and photoperiod (200 ± 3 μmol·m·s, 16 h) for 35 days.

HDR-GAN: HDR Image Reconstruction From Multi-Exposed LDR Images With Large Motions.

Synthesizing high dynamic range (HDR) images from multiple low-dynamic range (LDR) exposures in dynamic scenes is challenging. There are two major problems caused by the large motions of foreground objects. One is the severe misalignment among the LDR images.

Drug-drug interaction prediction with Wasserstein Adversarial Autoencoder-based knowledge graph embeddings.

An interaction between pharmacological agents can trigger unexpected adverse events. Capturing richer and more comprehensive information about drug-drug interactions (DDIs) is one of the key tasks in public health and drug development. Recently, several knowledge graph (KG) embedding approaches have received increasing attention in the DDI domain due to their capability of projecting drugs and interactions into a low-dimensional feature space for predicting links and classifying triplets.

Boundary-aware RGBD Salient Object Detection with Cross-modal Feature Sampling.

Mobile devices usually mount a depth sensor to resolve ill-posed problems, like salient object detection on cluttered background. The main barrier of exploring RGBD data is to handle the information from two different modalities. To cope with this problem, in this paper, we propose a boundary-aware cross-modal fusion network for RGBD salient object detection.

Novel decoy cellular vaccine strategy utilizing transgenic antigen-expressing cells as immune presenter and adjuvant in vaccine prototype against SARS-CoV-2 virus.

A novel approach modifying cells to express viral markers to elicit protective immunity responses (decoy cellular vaccination) in the prevention of COVID-19 disease is currently being explored. Our approach entails utilizing SARS-CoV-2 Spike antigen-expressing, non-replicating cells as carriers and presenters of immunogenic antigens, so called "I-cells". By using irradiated cells as presenting vehicles of SARS-CoV-2 viral antigens(s) in a cellular context, these presented viral proteins can be recognized by the host immune system, thus, an efficient protective immune response might be elicited.

Two Projection Neural Networks With Reduced Model Complexity for Nonlinear Programming.

Recent reports show that projection neural networks with a low-dimensional state space can enhance computation speed obviously. This paper proposes two projection neural networks with reduced model dimension and complexity (RDPNNs) for solving nonlinear programming (NP) problems. Compared with existing projection neural networks for solving NP, the proposed two RDPNNs have a low-dimensional state space and low model complexity.

Sugar accumulation and growth of lettuce exposed to different lighting modes of red and blue LED light.

The present study evaluated the growth response and sugar accumulation of lettuce exposed to different lighting modes of red and blue LED light based on the same daily light integral (7.49 μmol·m). Six lighting treatments were performed, that were monochromatic red light (R), monochromatic blue light (B), simultaneous red and blue light as the control (RB, R:B = 1:1), mixed modes of R, B and RB (R/RB/B, 4 h R to 4 h RB and then 4 h B), and alternating red and blue light with alternating intervals of 4 h and 1 h respectively recorded as R/B(4 h) and R/B(1 h).

Soil moisture and nitrate-nitrogen dynamics and economic yield in the greenhouse cultivation of tomato and cucumber under negative pressure irrigation in the North China Plain.

A field experiment was carried out for two years to investigate the benefits of negative pressure water supply on surface soil water content, nitrate-nitrogen (NO-N) distribution in the soil profile, economic yield and water and fertilizer use efficiency of tomato and cucumber under greenhouse cultivation in the North China Plain. The experiment included two irrigation treatments: drip irrigation with nutrient solution (DIN) and negative pressure irrigation with nutrient solution (NIN). The results showed that the NIN treatment had a relatively stable soil moisture (about 87% of field capacity), and the fluctuation of soil water content in the 0-20 cm soil layer was 20.

Spatial-Temporal Data Collection with Compressive Sensing in Mobile Sensor Networks.

Compressive sensing (CS) provides an energy-efficient paradigm for data gathering in wireless sensor networks (WSNs). However, the existing work on spatial-temporal data gathering using compressive sensing only considers either multi-hop relaying based or multiple random walks based approaches. In this paper, we exploit the mobility pattern for spatial-temporal data collection and propose a novel mobile data gathering scheme by employing the Metropolis-Hastings algorithm with delayed acceptance, an improved random walk algorithm for a mobile collector to collect data from a sensing field.

Comparative Transcriptomic Analysis of Vernalization- and Cytokinin-Induced Floral Transition in Dendrobium nobile.

Vernalization is required for floral initiation in Dendrobium. Interestingly, those beneficial effects can also be achieved by exogenous cytokinin application in greenhouses. Thus, an as yet unknown crosstalk/interaction may exist between vernalization and cytokinin signaling pathways.

A Game Theory Algorithm for Intra-Cluster Data Aggregation in a Vehicular Ad Hoc Network.

Vehicular ad hoc networks (VANETs) have an important role in urban management and planning. The effective integration of vehicle information in VANETs is critical to traffic analysis, large-scale vehicle route planning and intelligent transportation scheduling. However, given the limitations in the precision of the output information of a single sensor and the difficulty of information sharing among various sensors in a highly dynamic VANET, effectively performing data aggregation in VANETs remains a challenge.

[Analysis of inorganic elements in hydroponic Taraxacum mongolicum grown under different spectrum combinations by ICP-AES].

Dandelion (Taraxacum mongolicum) was hydroponically cultured in a completely enclosed plant factory, in which fluorescence and LED emitting spectra of different bands were used as the sole light source for plant growth. Effects of spectral component on the growth of dandelion were studied and the contents of ten inorganic elements such as K, P, Ca, Mg, Na, Fe, Mn, Zn, Cu and B in dandelion were analyzed by ICP-AES technology. The results showed that: (1) Under the condition of similar photosynthetic active radiation (PAR), single R or combined spectrums of FLRB were beneficial for biomass accumulation, while single B was the contrary; (2) Macroelements content ratio in Taraxacum mongolicum grown under FLwas K:Ca:P:Mg : Na=79.