Explore the vast range of titles available.

Automatic medical image segmentation plays a critical role in scientific research and medical care. Existing high-performance deep learning methods typically rely on large training datasets with high-quality manual annotations, which are difficult to obtain in many clinical applications. Here, we introduce Annotation-effIcient Deep lEarning (AIDE), an open-source framework to handle imperfect training datasets. Methodological analyses and empirical evaluations are conducted, and we demonstrate that AIDE surpasses conventional fully-supervised models by presenting better performance on open datasets possessing scarce or noisy annotations. We further test AIDE in a real-life case study for breast tumor segmentation. Three datasets containing 11,852 breast images from three medical centers are employed, and AIDE, utilizing 10% training annotations, consistently produces segmentation maps comparable to those generated by fully-supervised counterparts or provided by independent radiologists. The 10-fold enhanced efficiency in utilizing expert labels has the potential to promote a wide range of biomedical applications. Existing high-performance deep learning methods typically rely on large training datasets with high-quality manual annotations, which are difficult to obtain in many clinical applications. Here, the authors introduce an open-source framework to handle imperfect training datasets.

Electrical recordings in humans and monkeys show attentional enhancement of evoked responses and gamma synchrony in ventral stream cortical areas. Does this synchrony result from intrinsic activity in visual cortex or from inputs from other structures? Using paired recordings in the frontal eye field (FEF) and area V4, we found that attention to a stimulus in their joint receptive field leads to enhanced oscillatory coupling between the two areas, particularly at gamma frequencies. This coupling appeared to be initiated by FEF and was time-shifted by about 8 to 13 milliseconds across a range of frequencies. Considering the expected conduction and synaptic delays between the areas, this time-shifted coupling at gamma frequencies may optimize the postsynaptic impact of spikes from one area upon the other, improving cross-area communication with attention.

In this study, we chose a carnivorous fish, turbot (Scophthalmus maximus L.), to examine its nutrient-sensing and metabolic responses after ingestion of diets with fishmeal (FM), or 45 % of FM replaced by soyabean meal (34·6 % dry diet) balanced with or without essential amino acids (EAA) to match the amino acid profile of FM diet for 30 d. After a 1-month feeding trial, fish growth, feed efficiency and nutrient retention were markedly reduced by soyabean meal-incorporated (SMI) diets. Compared with the FM diet, SMI led to a reduction of postprandial influx of free amino acids, hypoactivated target of rapamycin signalling and a hyperactivated amino acid response pathway after refeeding, a status associated with reduced protein synthesis, impaired postprandial glycolysis and lipogenesis. These differential effects were not ameliorated by matching an EAA profile of soyabean meal to that of the FM diet through dietary amino acid supplementation. Therefore, this study demonstrated that the FM diet and SMI diets led to distinct nutrient-sensing responses, which in turn modulated metabolism and determined the utilisation efficiency of diets. Our results provide a new molecular explanation for the role of nutrient sensing in the inferior performance of aquafeeds in which FM is replaced by soyabean meal.

Supercapacitors have high power densities, high efficiencies, and long cycling lifetimes; however, to enable their wider use, their energy densities must be significantly improved. The design and synthesis of improved carbon materials with better capacitance, rate performance, and cycling stability has emerged as the main theme of supercapacitor research. Herein, we report a facile synthetic method to prepare nitrogen-rich carbon particles based on a continuous aerosol- spraying process. The method yields particles that have high surface areas, a uniform microporous structure, and are highly N-doped, resulting in a synergism that enables the construction of supercapacitors with high energy and power density for use in both aqueous and commercial organic electrolytes. Furthermore, we have used density functional theory calculations to show that the improved performance is due to the enhanced wettability and ion adsorption interactions at the carbon/electrolyte interface that result from nitrogen doping. These findings provide new insights into the role of heteroatom doping in the capacitance enhancement of carbon materials; in addition, our method offers an efficient route for large-scale production of doped carbon.

Prostate cancer is the most common malignancy in men in developed countries. Overexpression of enhancer of zeste homolog 2 (EZH2), the major histone H3 lysine 27 methyltransferase, has been connected to prostate cancer malignancy. However, its downstream genes and pathways have not been well established. Here, we show tumor suppressor Hepatocyte Nuclear Factor 1β (HNF1B) as a direct downstream target of EZH2. EZH2 binds HNF1B locus and suppresses HNF1B expression in prostate cancer cell lines, which is further supported by the reverse correlation between EZH2 and HNF1B expression in clinical samples. Consistently, restored HNF1B expression significantly suppresses EZH2-mediated overgrowth and EMT processes, including migration and invasion of prostate cancer cell lines. Mechanistically, we find that HNF1B primarily binds the promoters of thousands of target genes, and differentially regulates the expression of 876 genes. We also identify RBBP7/RbAP46 as a HNF1B interacting protein which is required for HNF1B-mediated repression of SLUG expression and EMT process. Importantly, we find that higher HNF1B expression strongly predicts better prognosis of prostate cancer, alone or together with lower EZH2 expression. Taken together, we have established a previously underappreciated axis of EZH2-HNF1B-SLUG in prostate cancer, and also provide evidence supporting HNF1B as a potential prognosis marker for metastatic prostate cancer.

According to the polarization curves, we can see that all of the reactors showed a similar open circuit potential (OCP) (Fig. 2b). According to the EIS analysis, the estimated total resistance of MFC-160 mT, MFC-80 mT, MFC-20 mT, and MFC-0 mT was around 651.2 Ω, 435.2 Ω, 676.9 Ω, and 907.9 Ω, respectively. The magnetic MFCs had a much lower diffusion resistance: around 585.9 Ω for MFC-160 mT, 365.1 Ω for MFC-80 mT, and 608.2 Ω for MFC-20 mT, compared with non-magnetic MFC-0 mT (835.3 Ω). [...]the SMF decreased the internal resistance of the MFCs with magnets as anodes mainly by reducing their diffusion resistance. Additionally, different bacteria may have different magnetic field tolerances [37]. [...]a higher SMF intensity may contribute to higher magnetohydrodynamic effects, increased activity of substrate-degradation bacteria, and inhibit the activity of exoelectrogens.

This study was designed to examine the cellular and systemic nutrient sensing mechanisms as well as the intermediary metabolism responses in turbot (Scophthalmus maximus L.) fed with fishmeal diet (FM diet), 45% of FM replaced by meat and bone meal diet (MBM diet) or MBM diet supplemented with essential amino acids to match the amino acid profile of FM diet (MBM+AA diet). During the one month feeding trial, feed intake was not affected by the different diets. However, MBM diet caused significant reduction of specific growth rate and nutrient retentions. Compared with the FM diet, MBM diet down-regulated target of rapamycin (TOR) and insulin-like growth factor (IGFs) signaling pathways, whereas up-regulated the amino acid response (AAR) signaling pathway. Moreover, MBM diet significantly decreased glucose and lipid anabolism, while increased muscle protein degradation and lipid catabolism in liver. MBM+AA diet had no effects on improvement of MBM diet deficiencies. Compared with fasted, re-feeding markedly activated the TOR signaling pathway, IGF signaling pathway and glucose, lipid metabolism, while significantly depressed the protein degradation signaling pathway. These results thus provided a comprehensive display of molecular responses and a better explanation of deficiencies generated after fishmeal replacement by other protein sources.

Spiking Neural Networks (SNNs), inspired by brain science, offer low energy consumption and high biological plausibility with their event-driven nature. However, the current SNNs are still suffering from insufficient performance. Recognizing the brain's adeptness at information processing for various scenarios with complex neuronal connections within and across regions, as well as specialized neuronal architectures for specific functions, we propose a Spiking Global-Local-Fusion Transformer (SGLFormer), that significantly improves the performance of SNNs. This novel architecture enables efficient information processing on both global and local scales, by integrating transformer and convolution structures in SNNs. In addition, we uncover the problem of inaccurate gradient backpropagation caused by Maxpooling in SNNs and address it by developing a new Maxpooling module. Furthermore, we adopt spatio-temporal block (STB) in the classification head instead of global average pooling, facilitating the aggregation of spatial and temporal features. SGLFormer demonstrates its superior performance on static datasets such as CIFAR10/CIFAR100, and ImageNet, as well as dynamic vision sensor (DVS) datasets including CIFAR10-DVS and DVS128-Gesture. Notably, on ImageNet, SGLFormer achieves a top-1 accuracy of 83.73% with 64 M parameters, outperforming the current SOTA directly trained SNNs by a margin of 6.66%. With its high performance, SGLFormer can support more computer vision tasks in the future. The codes for this study can be found in https://github.com/ZhangHanN1/SGLFormer.

Spiking neural networks (SNNs) offer a promising energy-efficient alternative to artificial neural networks (ANNs), in virtue of their high biological plausibility, rich spatial-temporal dynamics, and event-driven computation. The direct training algorithms based on the surrogate gradient method provide sufficient flexibility to design novel SNN architectures and explore the spatial-temporal dynamics of SNNs. According to previous studies, the performance of models is highly dependent on their sizes. Recently, direct training deep SNNs have achieved great progress on both neuromorphic datasets and large-scale static datasets. Notably, transformer-based SNNs show comparable performance with their ANN counterparts. In this paper, we provide a new perspective to summarize the theories and methods for training deep SNNs with high performance in a systematic and comprehensive way, including theory fundamentals, spiking neuron models, advanced SNN models and residual architectures, software frameworks and neuromorphic hardware, applications, and future trends.

This paper proposes a texture-enhanced network (TENet) for intertidal sediment and habitat classification using multiband multipolarization synthetic aperture radar (SAR) images. The architecture introduces the texture enhancement module (TEM) into the UNet framework to explicitly learn global texture information from SAR images. The study sites are chosen from the northern part of the intertidal zones in the German Wadden Sea. Results show that the presented TENet model is able to detail the intertidal surface types, including land, seagrass, bivalves, bright sands/beach, water, sediments, and thin coverage of vegetation or bivalves. To further assess its performance, we quantitatively compared our results from the TENet model with different instance segmentation models for the same areas of interest. The TENet model gives finer classification accuracies and shows great potential in providing more precise locations.