Dynamic immunoediting by macrophages in homologous recombination deficiency-stratified pancreatic ductal adenocarcinoma.

Pancreatic ductal adenocarcinoma (PDAC) is a lethal disease, notably resistant to existing therapies. Current research indicates that PDAC patients deficient in homologous recombination (HR) benefit from platinum-based treatments and poly-ADP-ribose polymerase inhibitors (PARPi). However, the effectiveness of PARPi in HR-deficient (HRD) PDAC is suboptimal, and significant challenges remain in fully understanding the distinct characteristics and implications of HRD-associated PDAC. We analyzed 16 PDAC patient-derived tissues, categorized by their homologous recombination deficiency (HRD) scores, and performed high-plex immunofluorescence analysis to define 20 cell phenotypes, thereby generating an in-situ PDAC tumor-immune landscape. Spatial phenotypic-transcriptomic profiling guided by regions-of-interest (ROIs) identified a crucial regulatory mechanism through localized tumor-adjacent macrophages, potentially in an HRD-dependent manner. Cellular neighborhood (CN) analysis further demonstrated the existence of macrophage-associated high-ordered cellular functional units in spatial contexts. Using our multi-omics spatial profiling strategy, we uncovered a dynamic macrophage-mediated regulatory axis linking HRD status with SIGLEC10 and CD52. These findings demonstrate the potential of targeting CD52 in combination with PARPi as a therapeutic intervention for PDAC.

Pixel Distillation: Cost-flexible Distillation across Image Sizes and Heterogeneous Networks.

Previous knowledge distillation (KD) methods mostly focus on compressing network architectures, which is not thorough enough in deployment as some costs like transmission bandwidth and imaging equipment are related to the image size. Therefore, we propose Pixel Distillation that extends knowledge distillation into the input level while simultaneously breaking architecture constraints. Such a scheme can achieve flexible cost control for deployment, as it allows the system to adjust both network architecture and image quality according to the overall requirement of resources. Specifically, we first propose an input spatial representation distillation (ISRD) mechanism to transfer spatial knowledge from large images to student's input module, which can facilitate stable knowledge transfer between CNN and ViT. Then, a Teacher-Assistant-Student (TAS) framework is further established to disentangle pixel distillation into the model compression stage and input compression stage, which significantly reduces the overall complexity of pixel distillation and the difficulty of distilling intermediate knowledge. Finally, we adapt pixel distillation to object detection via an aligned feature for preservation (AFP) strategy for TAS, which aligns output dimensions of detectors at each stage by manipulating features and anchors of the assistant. Comprehensive experiments on image classification and object detection demonstrate the effectiveness of our method.

Walnut-Derived Peptides Alleviate Learning and Memory Impairments in a Mice Model via Inhibition of Microglia Phagocytose Synapses.

Microglia phagocytose synapses have an important effect on the pathogenesis of neurological disorders. Here, we investigated the neuroprotective effects of the walnut-derived peptide, TWLPLPR(TW-7), against LPS-induced cognitive deficits in mice and explored the underlying C1q-mediated microglia phagocytose synapses mechanisms in LPS-treated HT22 cells. The MWM showed that TW-7 improved the learning and memory capacity of the LPS-injured mice. Both transmission electron microscopy and immunofluorescence analysis illustrated that synaptic density and morphology were increased while associated with the decreased colocalized synapses with C1q. Immunohistochemistry and immunofluorescence demonstrated that TW-7 effectively reduced the microglia phagocytosis of synapses. Subsequently, overexpression of C1q gene plasmid was used to verify the contribution of the TW-7 via the classical complement pathway-regulated mitochondrial function-mediated microglia phagocytose synapses in LPS-treated HT22 cells. These data suggested that TW-7 improved the learning and memory capability of LPS-induced cognitively impaired mice through a mechanism associated with the classical complement pathway-mediated microglia phagocytose synapse.

A frozen suspension external calibration strategy for elemental quantitative imaging of fresh plant soft tissues by LA-ICP-MS with cryogenic ablation cell.

A simple and reliable external calibration strategy of LA-ICP-MS for fresh plant soft tissues was developed. The prepared plant suspension was frozen by the designed cryogenic ablation cell and used as external standard for quantitative elemental imaging analysis of fresh plant tissues. The controllable water content of the prepared external standards provides a similar matrix with fresh soft tissues, and a homogeneous elemental distribution could be ensured due to the fine grinding particle sizes. More interestingly, the presence of water increased the signal intensity produced by the suspension by a factor of 1.6 (Pb) to 66.6 (La) compared to that of the pressed cake. The excellent dispersing property and advantage of long-term use were achieved owing to the employment of 0.1% PAANa as suspending agent. A series of plant reference materials were analyzed, and the relative errors of most elements were less than 10 %, indicating that there is a reliable accuracy of the proposed method. The limits of detection (LODs) ranged from 0.1 ng·g-1 (La) to 1279 ng·g-1 (S). This method was used for elemental imaging analysis in rice leaves under arsenic stress, and the results were consistent with previous studies, which mean that the proposed method could provide technical support for researchers in the fields of agriculture and environment.

Deep Generative Adversarial Reinforcement Learning for Semi-Supervised Segmentation of Low-Contrast and Small Objects in Medical Images.

Deep reinforcement learning (DRL) has demonstrated impressive performance in medical image segmentation, particularly for low-contrast and small medical objects. However, current DRL-based segmentation methods face limitations due to the optimization of error propagation in two separate stages and the need for a significant amount of labeled data. In this paper, we propose a novel deep generative adversarial reinforcement learning (DGARL) approach that, for the first time, enables end-to-end semi-supervised medical image segmentation in the DRL domain. DGARL ingeniously establishes a pipeline that integrates DRL and generative adversarial networks (GANs) to optimize both detection and segmentation tasks holistically while mutually enhancing each other. Specifically, DGARL introduces two innovative components to facilitate this integration in semi-supervised settings. First, a task-joint GAN with two discriminators links the detection results to the GAN's segmentation performance evaluation, allowing simultaneous joint evaluation and feedback. This ensures that DRL and GAN can be directly optimized based on each other's results. Second, a bidirectional exploration DRL integrates backward exploration and forward exploration to ensure the DRL agent explores the correct direction when forward exploration is disabled due to lack of explicit rewards. This mitigates the issue of unlabeled data being unable to provide rewards and rendering DRL unexplorable. Comprehensive experiments on three generalization datasets, comprising a total of 640 patients, demonstrate that our novel DGARL achieves 85.02% Dice and improves at least 1.91% for brain tumors, achieves 73.18% Dice and improves at least 4.28% for liver tumors, and achieves 70.85% Dice and improves at least 2.73% for pancreas compared to the ten most recent advanced methods, our results attest to the superiority of DGARL. Code is available at GitHub.

Uncertainty Modeling for Gaze Estimation.

Gaze estimation is an important fundamental task in computer vision and medical research. Existing works have explored various effective paradigms and modules for precisely predicting eye gazes. However, the uncertainty for gaze estimation, e.g., input uncertainty and annotation uncertainty, have been neglected in previous research. Existing models use a deterministic function to estimate the gaze, which cannot reflect the actual situation in gaze estimation. To address this issue, we propose a probabilistic framework for gaze estimation by modeling the input uncertainty and annotation uncertainty. We first utilize probabilistic embeddings to model the input uncertainty, representing the input image as a Gaussian distribution in the embedding space. Based on the input uncertainty modeling, we give an instance-wise uncertainty estimation to measure the confidence of prediction results, which is critical in practical applications. Then, we propose a new label distribution learning method, probabilistic annotations, to model the annotation uncertainty, representing the raw hard labels as Gaussian distributions. In addition, we develop an Embedding Distribution Smoothing (EDS) module and a hard example mining method to improve the consistency between embedding distribution and label distribution. We conduct extensive experiments, demonstrating that the proposed approach achieves significant improvements over baseline and state-of-the-art methods on two widely used benchmark datasets, GazeCapture and MPIIFaceGaze, as well as our collected dataset using mobile devices.

Capsule Networks With Residual Pose Routing.

Capsule networks (CapsNets) have been known difficult to develop a deeper architecture, which is desirable for high performance in the deep learning era, due to the complex capsule routing algorithms. In this article, we present a simple yet effective capsule routing algorithm, which is presented by a residual pose routing. Specifically, the higher-layer capsule pose is achieved by an identity mapping on the adjacently lower-layer capsule pose. Such simple residual pose routing has two advantages: 1) reducing the routing computation complexity and 2) avoiding gradient vanishing due to its residual learning framework. On top of that, we explicitly reformulate the capsule layers by building a residual pose block. Stacking multiple such blocks results in a deep residual CapsNets (ResCaps) with a ResNet-like architecture. Results on MNIST, AffNIST, SmallNORB, and CIFAR-10/100 show the effectiveness of ResCaps for image classification. Furthermore, we successfully extend our residual pose routing to large-scale real-world applications, including 3-D object reconstruction and classification, and 2-D saliency dense prediction. The source code has been released on https://github.com/liuyi1989/ResCaps.

Position-based anchor optimization for point supervised dense nuclei detection.

Nuclei detection is one of the most fundamental and challenging problems in histopathological image analysis, which can localize nuclei to provide effective computer-aided cancer diagnosis, treatment decision, and prognosis. The fully-supervised nuclei detector requires a large number of nuclei annotations on high-resolution digital images, which is time-consuming and needs human annotators with professional knowledge. In recent years, weakly-supervised learning has attracted significant attention in reducing the labeling burden. However, detecting dense nuclei of complex crowded distribution and diverse appearances remains a challenge. To solve this problem, we propose a novel point-supervised dense nuclei detection framework that introduces position-based anchor optimization to complete morphology-based pseudo-label supervision. Specifically, we first generate cellular-level pseudo labels (CPL) for the detection head via a morphology-based mechanism, which can help to build a baseline point-supervised detection network. Then, considering the crowded distribution of the dense nuclei, we propose a mechanism called Position-based Anchor-quality Estimation (PAE), which utilizes the positional deviation between an anchor and its corresponding point label to suppress low-quality detections far from each nucleus. Finally, to better handle the diverse appearances of nuclei, an Adaptive Anchor Selector (AAS) operation is proposed to automatically select positive and negative anchors according to morphological and positional statistical characteristics of nuclei. We conduct comprehensive experiments on two widely used benchmarks, MO and Lizard, using ResNet50 and PVTv2 as backbones. The results demonstrate that the proposed approach has superior capacity compared with other state-of-the-art methods. In particularly, in dense nuclei scenarios, our method can achieve 95.1% performance of the fully-supervised approach. The code is available at https://github.com/NucleiDet/DenseNucleiDet.

Weakly Supervised Semantic Segmentation via Alternate Self-Dual Teaching.

Weakly supervised semantic segmentation (WSSS) is a challenging yet important research field in vision community. In WSSS, the key problem is to generate high-quality pseudo segmentation masks (PSMs). Existing approaches mainly depend on the discriminative object part to generate PSMs, which would inevitably miss object parts or involve surrounding image background, as the learning process is unaware of the full object structure. In fact, both the discriminative object part and the full object structure are critical for deriving of high-quality PSMs. To fully explore these two information cues, we build a novel end-to-end learning framework, alternate self-dual teaching (ASDT), based on a dual-teacher single-student network architecture. The information interaction among different network branches is formulated in the form of knowledge distillation (KD). Unlike the conventional KD, the knowledge of the two teacher models would inevitably be noisy under weak supervision. Inspired by the Pulse Width (PW) modulation, we introduce a PW wave-like selection signal to alleviate the influence of the imperfect knowledge from either teacher model on the KD process. Comprehensive experiments on the PASCAL VOC 2012 and COCO-Stuff 10K demonstrate the effectiveness of the proposed ASDT framework, and new state-of-the-art results are achieved.

Complement in Human Brain Health: Potential of Dietary Food in Relation to Neurodegenerative Diseases.

The complement pathway is a major component of the innate immune system, which is critical for recognizing and clearing pathogens that rapidly react to defend the body against external pathogens. Many components of this pathway are expressed throughout the brain and play a beneficial role in synaptic pruning in the developing central nervous system (CNS). However, excessive complement-mediated synaptic pruning in the aging or injured brain may play a contributing role in a wide range of neurodegenerative diseases. Complement Component 1q (C1q), an initiating recognition molecule of the classical complement pathway, can interact with a variety of ligands and perform a range of functions in physiological and pathophysiological conditions of the CNS. This review considers the function and immunomodulatory mechanisms of C1q; the emerging role of C1q on synaptic pruning in developing, aging, or pathological CNS; the relevance of C1q; the complement pathway to neurodegenerative diseases; and, finally, it summarizes the foods with beneficial effects in neurodegenerative diseases via C1q and complement pathway and highlights the need for further research to clarify these roles. This paper aims to provide references for the subsequent study of food functions related to C1q, complement, neurodegenerative diseases, and human health.

Self-Supervised Interactive Embedding for One-Shot Organ Segmentation.

One-shot organ segmentation (OS2) aims at segmenting the desired organ regions from the input medical imaging data with only one pre-annotated example as the reference. By using the minimal annotation data to facilitate organ segmentation, OS2 receives great attention in the medical image analysis community due to its weak requirement on human annotation. In OS2, one core issue is to explore the mutual information between the support (reference slice) and the query (test slice). Existing methods rely heavily on the similarity between slices, and additional slice allocation mechanisms need to be designed to reduce the impact of the similarity between slices on the segmentation performance. To address this issue, we build a novel support-query interactive embedding (SQIE) module, which is equipped with the channel-wise co-attention, spatial-wise co-attention, and spatial bias transformation blocks to identify "what to look", "where to look", and "how to look" in the input test slice. By combining the three mechanisms, we can mine the interactive information of the intersection area and the disputed area between slices, and establish the feature connection between the target in slices with low similarity. We also propose a self-supervised contrastive learning framework, which transforms knowledge from the physical position to the embedding space to facilitate the self-supervised interactive embedding of the query and support slices. Comprehensive experiments on two large benchmarks demonstrate the superior capacity of the proposed approach when compared with the current alternatives and baseline models.

Facile Preparation of Dense Polysulfone UF Membranes with Enhanced Salt Rejection by Post-Heating.

Polysulfone (PSf) membranes typically have a negligible rejection of salts due to the intrinsic larger pore size and wide pore size distribution. In this work, a facile and scalable heat treatment was proposed to increase the salt rejection. The influence of heat treatment on the structure and performance of PSf membranes was systematically investigated. The average pore size decreased from 9.94 ± 5.5 nm for pristine membranes to 1.18 ± 0.19 nm with the increase in temperature to 50 °C, while the corresponding porosity decreased from 2.07% to 0.13%. Meanwhile, the thickness of the sponge structure decreased from 20.20 to 11.5 µm as the heat treatment temperature increased to 50 °C. The MWCO of PSf decreased from 290,000 Da to 120 Da, whereas the membrane pore size decreased from 5.5 to 0.19 nm. Correspondingly, the water flux decreased from 1545 to 27.24 L·m-2·h-1, while the rejection ratio increased from 3.1% to 74.0% for Na2SO4, from 1.3% to 48.2% for MgSO4, and from 0.6% to 23.8% for NaCl. Meanwhile, mechanism analysis indicated that the water evaporation in the membranes resulted in the shrinkage of the membrane pores and decrease in the average pore size, thus improving the separation performance. In addition, the desalting performance of the heat-treated membranes for real actual industrial wastewater was improved. This provides a facile and scalable route for PSf membrane applications for enhanced desalination.

Direct multi-elemental analysis of cerebrospinal fluid samples by LA-ICP-MS employing an aerosol local extraction cryogenic ablation cell.

A novel method for direct high-throughput analysis of multi-elements in cerebrospinal fluid (CSF) samples by laser ablation inductively coupled plasma mass spectrometry with an aerosol local extraction cryogenic ablation cell (ALEC-LA-ICP-MS) was developed. Microliter-level CSF samples were frozen by a designed cryogenic ablation cell and directly analyzed by laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) without time-consuming pretreatment. Compared with the precision obtained at room temperature (20℃), that obtained at low temperature (- 25℃) was significantly improved; the RSDs were reduced from 8.3% (Zn) to 32.6% (Mn) to 2.2% (Pb) to 6.5% (Mn) with six times parallel determination. To meet the analytical requirement of the micro-volume CSF samples, the laminar flow aerosol local extraction strategy was adopted to improve the transmission efficiency of aerosols, and the signal intensity was increased by four times compared with the standard commercial ablation cell. The standard solution with 0.4% bovine serum albumin (BSA) matrix was used as matrix-match external standard, and Rh was added into the samples as internal standard. The limits of detection (LODs) ranged from 0.17 µg·L-1 (Mn) to 8.67 µg·L-1 (Mg). Standard addition recovery experiments and the determination of CRM serum L-1 and L-2 were carried out to validate the accuracy of the method; all results indicated there were excellent accuracy and precision in the proposed method. The matrix-scanning function in the GeoLas software combined with the microwell plate realizes the high-throughput automatic analysis. Twenty-four CSF samples from different patients were determined; the results showed that there might be a correlation between the metal elements in CSF and the diseases, which means that the proposed method has potential in the diagnosis of neurological diseases.

Separating Noisy Samples From Tail Classes for Long-Tailed Image Classification With Label Noise.

Most existing methods that cope with noisy labels usually assume that the classwise data distributions are well balanced. They are difficult to deal with the practical scenarios where training samples have imbalanced distributions, since they are not able to differentiate noisy samples from tail classes' clean samples. This article makes an early effort to tackle the image classification task in which the provided labels are noisy and have a long-tailed distribution. To deal with this problem, we propose a new learning paradigm which can screen out noisy samples by matching between inferences on weak and strong data augmentations. A leave-noise-out regularization (LNOR) is further introduced to eliminate the effect of the recognized noisy samples. Besides, we propose a prediction penalty based on the online classwise confidence levels to avoid the bias toward easy classes which are dominated by head classes. Extensive experiments on five datasets including CIFAR-10, CIFAR-100, MNIST, FashionMNIST, and Clothing1M demonstrate that the proposed method outperforms the existing algorithms for learning with long-tailed distribution and label noise.

A review on the identification of transgenic oilseeds and oils.

Transgenic technology can increase the quantity and quality of vegetable oils worldwide. However, people are skeptical about the safety of transgenic oil-bearing crops and the oils they produce. In order to protect consumers' rights and avoid transgenic oils being adulterated or labeled as nontransgenic oils, the transgenic detection technology of oilseeds and oils needs careful consideration. This paper first summarized the current research status of transgenic technologies implemented at oil-bearing crops. Then, an inspection process was proposed to detect a large number of samples to be the subject rapidly, and various inspection strategies for transgenic oilseeds and oils were summarized according to the process sequence. The detection indicators included oil content, fatty acid, triglyceride, tocopherol, and nucleic acid. The detection technologies involved chromatography, spectroscopy, nuclear magnetic resonance, and polymerase chain reaction. It is hoped that this article can provide crucial technical reference and support for staff engaging in the supervision of transgenic food and for researchers developing fast and efficient monitoring methods in the future.

Mutant KRAS Mediates circARFGEF2 Biogenesis to Promote Lymphatic Metastasis of Pancreatic Ductal Adenocarcinoma.

Circular RNAs (circRNA) contribute to cancer stemness, proliferation, and metastasis. The biogenesis of circRNAs can be impacted by the genetic landscape of tumors. Herein, we identified a novel circRNA, circARFGEF2 (hsa_circ_0060665), which was upregulated in KRASG12D pancreatic ductal adenocarcinoma (PDAC) and positively associated with KRASG12D PDAC lymph node (LN) metastasis. CircARFGEF2 overexpression significantly facilitated KRASG12D PDAC LN metastasis in vitro and in vivo. Mechanistically, circARFGEF2 biogenesis in KRASG12D PDAC was significantly activated by the alternative splicing factor QKI-5, which recruited U2AF35 to facilitate spliceosome assembly. QKI-5 bound the QKI binding motifs and neighboring reverse complement sequence in intron 3 and 6 of ARFGEF2 pre-mRNA to facilitate circARFGEF2 biogenesis. CircARFGEF2 sponged miR-1205 and promoted the activation of JAK2, which phosphorylated STAT3 to trigger KRASG12D PDAC lymphangiogenesis and LN metastasis. Importantly, circARFGEF2 silencing significantly inhibited LN metastasis in the KrasG12D/+Trp53R172H/+Pdx-1-Cre (KPC) mouse PDAC model. These findings provide insight into the mechanism and metastasis-promoting function of mutant KRAS-mediated circRNA biogenesis. SIGNIFICANCE: Increased splicing-mediated biogenesis of circARFGEF2 in KRAS-mutant pancreatic ductal adenocarcinoma activates JAK2-STAT3 signaling and triggers lymph node metastasis, suggesting circARFGEF2 could be a therapeutic target to inhibit pancreatic cancer progression.

Reliable Mutual Distillation for Medical Image Segmentation Under Imperfect Annotations.

Convolutional neural networks (CNNs) have made enormous progress in medical image segmentation. The learning of CNNs is dependent on a large amount of training data with fine annotations. The workload of data labeling can be significantly relieved via collecting imperfect annotations which only match the underlying ground truths coarsely. However, label noises which are systematically introduced by the annotation protocols, severely hinders the learning of CNN-based segmentation models. Hence, we devise a novel collaborative learning framework in which two segmentation models cooperate to combat label noises in coarse annotations. First, the complementary knowledge of two models is explored by making one model clean training data for the other model. Secondly, to further alleviate the negative impact of label noises and make sufficient usage of the training data, the specific reliable knowledge of each model is distilled into the other model with augmentation-based consistency constraints. A reliability-aware sample selection strategy is incorporated for guaranteeing the quality of the distilled knowledge. Moreover, we employ joint data and model augmentations to expand the usage of reliable knowledge. Extensive experiments on two benchmarks showcase the superiority of our proposed method against existing methods under annotations with different noise levels. For example, our approach can improve existing methods by nearly 3% DSC on the lung lesion segmentation dataset LIDC-IDRI under annotations with 80% noise ratio. Code is available at: https://github.com/Amber-Believe/ReliableMutualDistillation.

Construction of a TiO2/BiOCl heterojunction for enhanced solar photocatalytic oxidation of nitric oxide.

TiO2/BiOCl heterojunction photocatalysts with different molar ratios (Ti : Bi) were synthesized by a simple solvothermal method. Various spectroscopic techniques including X-ray diffraction (XRD), scanning electron microscopy (SEM), high-resolution transmission electron microscopy (HRTEM), nitrogen adsorption-desorption, X-ray photoelectron spectroscopy (XPS), photoluminescence (PL) and UV-Vis diffuse reflectance spectroscopy (UV-vis DRS) were used to characterize the prepared photocatalysts. The photocatalytic activity of the catalysts was investigated by removing low concentrations of nitrogen oxides. The characterization results show that the TiO2/BiOCl composite photocatalyst exhibits superior visible light response performance than pure BiOCl and TiO2. The optimized TiO2/BiOCl heterojunction with a Ti : Bi molar ratio of 4 : 1 has the best photocatalytic performance. The removal rate of nitrogen oxides of the composite photocatalyst can reach 75%, which is 2.34 times higher than that of pure BiOCl. The observed photocatalytic degradation activity of nitrogen oxides outperforms current state-of-the-art functional photocatalysts. The TiO2/BiOCl composite photocatalyst has a larger specific surface area, stronger visible light absorption and higher charge separation efficiency compared to other control samples, which contribute to the enhanced photocatalytic activity. The experimental results indicate that the combination of TiO2 with BiOCl is a promising technique to design visible light-responsive photocatalysts.

CLRNet: Component-Level Refinement Network for Deep Face Parsing.

Face parsing aims to assign pixel-wise semantic labels to different facial components (e.g., hair, brows, and lips) in given face images. However, directly predicting pixel-level labels for each facial component over the whole face image would obtain limited accuracy, especially for tiny facial components. To address this problem, some recent works propose to first crop tiny patches from the whole face image and then predict masks for each facial component. However, such cropping-and-segmenting strategy consists of two independent stages, which cannot be jointly optimized. Besides, as one valuable piece of information for parsing the highly structured facial components, context cues are not elaborately explored by the existing works. To address these issues, we propose a component-level refinement network (CLRNet) for precisely segmenting out each facial component. Specifically, we introduce an attention mechanism to bridge the two independent stages together and form an end-to-end trainable pipeline for face parsing. Furthermore, we incorporate the global context information into the refining process for each cropped facial component patch, providing informative cues for accurate parsing. Extensive experiments are carried out on two benchmark datasets, LFW-PL and HELEN. The results demonstrate the superiority of the proposed CLRNet over other state-of-the-art methods, especially for tiny facial components.