FNeXter: A Multi-Scale Feature Fusion Network Based on ConvNeXt and Transformer for Retinal OCT Fluid Segmentation.

The accurate segmentation and quantification of retinal fluid in Optical Coherence Tomography (OCT) images are crucial for the diagnosis and treatment of ophthalmic diseases such as age-related macular degeneration. However, the accurate segmentation of retinal fluid is challenging due to significant variations in the size, position, and shape of fluid, as well as their complex, curved boundaries. To address these challenges, we propose a novel multi-scale feature fusion attention network (FNeXter), based on ConvNeXt and Transformer, for OCT fluid segmentation. In FNeXter, we introduce a novel global multi-scale hybrid encoder module that integrates ConvNeXt, Transformer, and region-aware spatial attention. This module can capture long-range dependencies and non-local similarities while also focusing on local features. Moreover, this module possesses the spatial region-aware capabilities, enabling it to adaptively focus on the lesions regions. Additionally, we propose a novel self-adaptive multi-scale feature fusion attention module to enhance the skip connections between the encoder and the decoder. The inclusion of this module elevates the model's capacity to learn global features and multi-scale contextual information effectively. Finally, we conduct comprehensive experiments to evaluate the performance of the proposed FNeXter. Experimental results demonstrate that our proposed approach outperforms other state-of-the-art methods in the task of fluid segmentation.

A coarse-to-fine point completion network with details compensation and structure enhancement.

Point cloud completion, the issue of estimating the complete geometry of objects from partially-scanned point cloud data, becomes a fundamental task in many 3d vision and robotics applications. To address the limitations on inadequate prediction of shape details for traditional methods, a novel coarse-to-fine point completion network (DCSE-PCN) is introduced in this work using the modules of local details compensation and shape structure enhancement for effective geometric learning. The coarse completion stage of our network consists of two branches-a shape structure recovery branch and a local details compensation branch, which can recover the overall shape of the underlying model and the shape details of incomplete point cloud through feature learning and hierarchical feature fusion. The fine completion stage of our network employs the structure enhancement module to reinforce the correlated shape structures of the coarse repaired shape (such as regular arrangement or symmetry), thus obtaining the completed geometric shape with finer-grained details. Extensive experimental results on ShapeNet dataset and ModelNet dataset validate the effectiveness of our completion network, which can recover the shape details of the underlying point cloud whilst maintaining its overall shape. Compared to the existing methods, our coarse-to-fine completion network has shown its competitive performance on both chamfer distance (CD) and earth mover distance (EMD) errors. Such as, the repairing results on the ShapeNet dataset of our completion network are reduced by an average of [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text] in terms of CD error, comparing with PCN, FoldingNet, Atlas, and CRN methods, respectively; and also reduced by an average of [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text] in terms of EMD error, respectively. Meanwhile, the completion results on the ModelNet dataset of our network have an average reduction of [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text] in terms of CD error, comparing to PCN, FoldingNet, Atlas, and CRN methods, respectively; and also an average reduction of [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text] in terms of EMD error, respectively. Our proposed point completion network is also robust to different degrees of data incompleteness and model noise.

The ERAD Pathway Participates in Fungal Growth and Cellulase Secretion in Trichoderma reesei.

Trichoderma reesei is a powerful fungal cell factory for the production of cellulolytic enzymes due to its outstanding protein secretion capacity. Endoplasmic reticulum-associated degradation (ERAD) plays an integral role in protein secretion that responds to secretion pressure and removes misfolded proteins. However, the role of ERAD in fungal growth and endogenous protein secretion, particularly cellulase secretion, remains poorly understood in T. reesei. Here, we investigated the ability of T. reesei to grow under different stresses and to secrete cellulases by disrupting three major genes (hrd1, hrd3 and der1) involved in the critical parts of the ERAD pathway. Under the ER stress induced by high concentrations of DTT, knockout of hrd1, hrd3 and der1 resulted in severely impaired growth, and the mutants Δhrd1 and Δhrd3 exhibited high sensitivity to the cell wall-disturbing agents, CFW and CR. In addition, the absence of either hrd3 or der1 led to the decreased heat tolerance of this fungus. These mutants showed significant differences in the secretion of cellulases compared to the parental strain QM9414. During fermentation, the secretion of endoglucanase in the mutants was essentially consistent with that of the parental strain, while cellobiohydrolase and ß-glucosidase were declined. It was further discovered that the transcription levels of the endoglucanase-encoding genes (eg1 and eg2) and the cellobiohydrolase-encoding gene (cbh1) were not remarkedly changed. However, the ß-glucosidase-encoding gene (bgl1) was significantly downregulated in the ERAD-deficient mutants, which was presumably due to the activation of a proposed feedback mechanism, repression under secretion stress (RESS). Taken together, our results indicate that a defective ERAD pathway negatively affects fungal growth and cellulase secretion, which provides a novel insight into the cellulase secretion mechanism in T. reesei.

Novel YOF-Based Theranostic Agents with a Cascade Effect for NIR-II Fluorescence Imaging and Synergistic Starvation/Photodynamic Therapy of Orthotopic Gliomas.

Accurate diagnosis and highly effective treatment of glioblastoma are still challenges in clinic. Near-infrared (NIR) light triggered fluorescence imaging and photodynamic therapy (PDT) showed the potential for theranostics of glioblastoma, but the presence of blood-brain barrier (BBB) and hypoxia limited treatment effect. Herein, the novel theranostic nanoagents with YOF:Nd3+ as core, MnO2 as shell, and further loading photosensitizer (indocyanine green, ICG) and glucose oxidase (GOx) were successfully constructed, and further modified with lactoferrin to endow them with BBB penetration and target abilities (YOF:Nd3+@MnO2-ICG-GOx-LF, YMIGL). The YOF:Nd3+ core with good fluorescence performances makes YMIGL act as promising probes for fluorescence imaging in the second biowindow (NIR-II FL). The combination of GOx and MnO2 shell significantly increased the O2 generation from the cascade reactions and consumed glucose, improving the treatment effect of PDT and achieving starvation treatment (ST). These theranostic nanoagents exhibit a highly efficient inhibition effect on orthotopic gliomas by cascade reactions, which improved PDT and ST.

Population Risk Improvement with Model Compression: An Information-Theoretic Approach.

It has been reported in many recent works on deep model compression that the population risk of a compressed model can be even better than that of the original model. In this paper, an information-theoretic explanation for this population risk improvement phenomenon is provided by jointly studying the decrease in the generalization error and the increase in the empirical risk that results from model compression. It is first shown that model compression reduces an information-theoretic bound on the generalization error, which suggests that model compression can be interpreted as a regularization technique to avoid overfitting. The increase in empirical risk caused by model compression is then characterized using rate distortion theory. These results imply that the overall population risk could be improved by model compression if the decrease in generalization error exceeds the increase in empirical risk. A linear regression example is presented to demonstrate that such a decrease in population risk due to model compression is indeed possible. Our theoretical results further suggest a way to improve a widely used model compression algorithm, i.e., Hessian-weighted K-means clustering, by regularizing the distance between the clustering centers. Experiments with neural networks are provided to validate our theoretical assertions.

Clarifying the high on/off ratio mechanism of nanowire UV photodetector by characterizing surface barrier height.

The response of semiconductor nanowire UV sensors represented by ZnO nanowire UV sensor is usually explained by the adsorption and desorption of oxygen molecules, but with the great increase of these sensors' on/off ratio in recent years, this explanation is inadequate and the inner mechanism for the large on/off ratio urgently needs to be explored. Here, the distribution of carrier concentration in a ZnO nanowire is found to be determined as a function of the radius of the nanowire, using a calibrated surface photovoltage method and space charge model. A critical radius is indicated which determines the carrier concentration and photoresponse behavior of the nanowire. When the radius is below this critical value, the carrier concentration in the dark decreases dramatically compared with that of the nanowire under UV light illumination. Specifically, a decrease of carrier concentration by 4-5 orders of magnitude occurs when the radius is below 50 nm, which causes the on/off ratio to vary by the same orders of magnitude. When the radius is above the critical value, the influence of radius on carrier concentration is nonsignificant and the on/off ratio is below 100. Finally, we found that the high on/off ratio of the ZnO nanowire should be ascribed to the complete depletion of the nanowire led by the interplay of radius and surface band bending rather than the change in width of the depletion layer as most papers have suggested.