GluN2B-containing NMDA receptor attenuated neuronal apoptosis in the mouse model of HIBD through inhibiting endoplasmic reticulum stress-activated PERK/eIF2α signaling pathway.

Introduction: Neonatal hypoxic-ischemic brain damage (HIBD) refers to brain damage in newborns caused by hypoxia and reduced or even stopped cerebral blood flow during the perinatal period. Currently, there are no targeted treatments for neonatal ischemic hypoxic brain damage, primarily due to the incomplete understanding of its pathophysiological mechanisms. Especially, the role of NMDA receptors is less studied in HIBD. Therefore, this study explored the molecular mechanism of endogenous protection mediated by GluN2B-NMDAR in HIBD. Method: Hypoxic ischemia was induced in mice aged 9-11 days. The brain damage was examined by Nissl staining and HE staining, while neuronal apoptosis was examined by Hoechst staining and TTC staining. And cognitive deficiency of mice was examined by various behavior tests including Barnes Maze, Three Chamber Social Interaction Test and Elevated Plus Maze. The activation of ER stress signaling pathways were evaluated by Western blot. Results: We found that after HIBD induction, the activation of GluN2B-NMDAR attenuated neuronal apoptosis and brain damage. Meanwhile, the ER stress PERK/eIF2α signaling pathway was activated in a time-dependent manner after HIBE. Furthermore, after selective inhibiting GluN2B-NMDAR in HIBD mice with ifenprodil, the PERK/eIF2α signaling pathway remains continuously activated, leading to neuronal apoptosis, morphological brain damage. and aggravating deficits in spatial memory, cognition, and social abilities in adult mice. Discussion: The results of this study indicate that, unlike its role in adult brain damage, GluN2B in early development plays a neuroprotective role in HIBD by inhibiting excessive activation of the PERK/eIF2α signaling pathway. This study provides theoretical support for the clinical development of targeted drugs or treatment methods for HIBD.

Nanoscale myelinogenesis image in developing brain via super-resolution nanoscopy by near-infrared emissive curcumin-BODIPY derivatives.

Understanding the intricate nanoscale architecture of neuronal myelin during central nervous system development is of utmost importance. However, current visualization methods heavily rely on electron microscopy or indirect fluorescent method, lacking direct and real-time imaging capabilities. Here, we introduce a breakthrough near-infrared emissive curcumin-BODIPY derivative (MyL-1) that enables direct visualization of myelin structure in brain tissues. The remarkable compatibility of MyL-1 with stimulated emission depletion nanoscopy allows for unprecedented super-resolution imaging of myelin ultrastructure. Through this innovative approach, we comprehensively characterize the nanoscale myelinogenesis in three dimensions over the course of brain development, spanning from infancy to adulthood in mouse models. Moreover, we investigate the correlation between myelin substances and Myelin Basic Protein (MBP), shedding light on the essential role of MBP in facilitating myelinogenesis during vertebral development. This novel material, MyL-1, opens up new avenues for studying and understanding the intricate process of myelinogenesis in a direct and non-invasive manner, paving the way for further advancements in the field of nanoscale neuroimaging.

Metal-Phenolic Nanocapsules with Photothermal Antibacterial and Ros Scavenging Ability for Diabetic Wound Healing.

The presence of bacteria in diabetic wounds not only leads to the formation of biofilms but also triggers oxidative stress and inflammatory responses, which hinder the wound-healing process. Therefore, it is imperative to formulate a comprehensive strategy that can proficiently eliminate bacteria and enhance the wound microenvironment. Herein, this work develops multifunctional metal-phenolic nanozymes (TA-Fe/Cu nanocapsules), wherein the one-pot coordination of tannic acid (TA)and Fe3+/Cu2+ using a self-sacrificial template afforded hollow nanoparticles (NPs) with exceptional photothermal and reactive oxygen species scavenging capabilities. After photothermal disruption of the biofilms, TA-Fe/Cu NPs autonomously capture bacteria through hydrogen bonding interactions with peptidoglycans (the bacterial cell wall component), ultimately bolstering the bactericidal efficacy. Furthermore, these NPs exhibit peroxidase-like enzymatic activity, efficiently eliminating surplus hydrogen peroxide in the vicinity of the wound and mitigating inflammatory responses. As the wound transitions into the remodeling phase, the presence of Cu2+ stimulates vascular migration and regeneration, expediting the wound-healing process. This study innovatively devises a minimalist approach to synthesize multifunctional metal-phenolic nanozymes integrating potent photothermal antibacterial activity, bacterial capture, anti-inflammatory, and angiogenesis properties, showcasing their great potential for diabetic wound treatment.

Gasotransmitter nitric oxide imaging in Alzheimer's disease and glioblastoma with diamino-cyclic-metalloiridium phosphorescence probes.

Nitric Oxide (NO), a significant gasotransmitter in biological systems, plays a crucial role in neurological diseases and cancer. Currently, there is a lack of effective methods for rapidly and sensitively identifying NO and elucidating its relationship with neurological diseases. Novel diamino-cyclic-metalloiridium phosphorescence probes, Ir-CDA and Ir-BDA, have been designed to visualize the gasotransmitter NO in Alzheimer's disease (AD) and glioblastoma (GBM). Ir-CDA and Ir-BDA utilize iridium (III) as the central ion and incorporate a diamino group as a ligand. The interaction between the diamino structure and NO leads to the formation of a three-nitrogen five-membered ring structure, which opens up phosphorescence. The two probes can selectively bind to NO and offer low detection limits. Additionally, Ir-BDA/Ir-CDA can image NO in brain cancer cell models, neuroinflammatory models, and AD cell models. Furthermore, the NO content in fresh brain sections from AD mice was considerably higher than that in wild-type (WT) mice. Consequently, it is plausible that NO is generated in significant quantities around cells hosting larger Aß deposits, gradually diffusing throughout the entire brain region. Furthermore, we posit that this phenomenon is a key factor contributing to the higher brain NO content in AD mice compared to that in WT mice. This discovery offers novel insights into the diagnosis and treatment of AD.

Sensory nerves, but not sympathetic nerves, promote reparative dentine formation after dentine injury via CGRP-mediated angiogenesis: An in vivo study.

AIM: Dental pulp is richly innervated by nerve fibres, which are mainly involved in the sensation of pain. Aside from pain sensation, little is known regarding the role of dental innervation in reparative dentine formation. We herein generated a mouse model of experimental dentine injury to examine nerve sprouting within the odontoblast and subodontoblastic layers and investigated the potential effects of this innervation in reparative dentinogenesis. METHODOLOGY: Mouse tooth cavity model (bur preparation + etching) was established, and then nerve sprouting, angiogenesis and reparative dentinogenesis were determined by histological and immunofluorescent staining at 1, 3, 7, 14 and 28 days postoperatively. We also established the mouse-denervated molar models to determine the role of sensory and sympathetic nerves in reparative dentinogenesis, respectively. Finally, we applied calcitonin gene-related peptide (CGRP) receptor antagonist to analyse the changes in angiogenesis and reparative dentinogenesis. RESULTS: Sequential histological results from dentine-exposed teeth revealed a significant increase in innervation directly beneath the injured area on the first day after dentine exposure, followed by vascularisation and reparative dentine production at 3 and 7 days, respectively. Intriguingly, abundant type H vessels (CD31+ Endomucin+ ) were present in the innervated area, and their formation precedes the onset of reparative dentine formation. Additionally, we found that sensory denervation led to blunted angiogenesis and impaired dentinogenesis, while sympathetic denervation did not affect dentinogenesis. Moreover, a marked increase in the density of CGRP+ nerve fibres was seen on day 3, which was reduced but remained elevated over the baseline level on day 14, whereas the density of substance P-positive nerve fibres did not change significantly. CGRP receptor antagonist-treated mice showed similar results as those with sensory denervation, including impairments in type H angiogenesis, which confirms the importance of CGRP in the formation of type H vessels. CONCLUSIONS: Dental pulp sensory nerves act as an essential upstream mediator to promote angiogenesis, including the formation of type H vessels, and reparative dentinogenesis. CGRP signalling governs the nerve-vessel-reparative dentine network, which is mostly produced by newly dense sensory nerve fibres within the dental pulp.

Quaternion Cross-Modality Spatial Learning for Multi-Modal Medical Image Segmentation.

Recently, the Deep Neural Networks (DNNs) have had a large impact on imaging process including medical image segmentation, and the real-valued convolution of DNN has been extensively utilized in multi-modal medical image segmentation to accurately segment lesions via learning data information. However, the weighted summation operation in such convolution limits the ability to maintain spatial dependence that is crucial for identifying different lesion distributions. In this paper, we propose a novel Quaternion Cross-modality Spatial Learning (Q-CSL) which explores the spatial information while considering the linkage between multi-modal images. Specifically, we introduce to quaternion to represent data and coordinates that contain spatial information. Additionally, we propose Quaternion Spatial-association Convolution to learn the spatial information. Subsequently, the proposed De-level Quaternion Cross-modality Fusion (De-QCF) module excavates inner space features and fuses cross-modality spatial dependency. Our experimental results demonstrate that our approach compared to the competitive methods perform well with only 0.01061 M parameters and 9.95G FLOPs.

Hybrid Reinforced Medical Report Generation With M-Linear Attention and Repetition Penalty.

To reduce doctors' workload, deep-learning-based automatic medical report generation has recently attracted more and more research efforts, where deep convolutional neural networks (CNNs) are employed to encode the input images, and recurrent neural networks (RNNs) are used to decode the visual features into medical reports automatically. However, these state-of-the-art methods mainly suffer from three shortcomings: 1) incomprehensive optimization; 2) low-order and unidimensional attention; and 3) repeated generation. In this article, we propose a hybrid reinforced medical report generation method with m-linear attention and repetition penalty mechanism (HReMRG-MR) to overcome these problems. Specifically, a hybrid reward with different weights is employed to remedy the limitations of single-metric-based rewards, and a local optimal weight search algorithm is proposed to significantly reduce the complexity of searching the weights of the rewards from exponential to linear. Furthermore, we use m-linear attention modules to learn multidimensional high-order feature interactions and to achieve multimodal reasoning, while a new repetition penalty is proposed to apply penalties to repeated terms adaptively during the model's training process. Extensive experimental studies on two public benchmark datasets show that HReMRG-MR greatly outperforms the state-of-the-art baselines in terms of all metrics. The effectiveness and necessity of all components in HReMRG-MR are also proved by ablation studies. Additional experiments are further conducted and the results demonstrate that our proposed local optimal weight search algorithm can significantly reduce the search time while maintaining superior medical report generation performances.

Implanted, Wireless, Self-Powered Photodynamic Therapeutic Tablet Synergizes with Ferroptosis Inducer for Effective Cancer Treatment.

The effective and targeted treatment of resistant cancer cells presents a significant challenge. Targeting cell ferroptosis has shown remarkable efficacy against apoptosis-resistant tumors due to their elevated iron metabolism and oxidative stress levels. However, various obstacles have limited its effectiveness. To overcome these challenges and enhance ferroptosis in cancer cells, we have developed a self-powered photodynamic therapeutic tablet that integrates a ferroptosis inducer (FIN), imidazole ketone erastin (IKE). FINs augment the sensitivity of photodynamic therapy (PDT) by increasing oxidative stress and lipid peroxidation. Furthermore, they utilize the Fenton reaction to supplement oxygen, generating a greater amount of reactive oxygen species (ROS) during PDT. Additionally, PDT facilitates the release of iron ions from the labile iron pool (LIP), accelerating lipid peroxidation and inducing ferroptosis. In vitro and in vivo experiments have demonstrated a more than 85% tumor inhibition rate. This synergistic treatment approach not only addresses the limitations of inadequate penetration and tumor hypoxia associated with PDT but also reduces the required medication dosage. Its high efficiency and specificity towards targeted cells minimize adverse effects, presenting a novel approach to combat clinical resistance in cancer treatment.

O2-Generation-Enhanced Responsive Starvation/Photothermal Synergistic Tumor Therapy Based on the AuNRs@MnO2@SiO2 Nanocarrier and Thermosensitive Biomimetic Camouflaging.

Cancer starvation/photothermal combined tumor therapy (CST/PTT) has attracted great interest attributed to their mutual compensation and synergistically enhanced effect. However, the very low O2 supply in the tumor microenvironment (TME) greatly limits the CST efficiency of glucose oxidase (GOx). Additionally, the easy degradation in blood circulation and significant off-target effects are big challenges for clinical applications of the GOx-based CST. In this study, a drug delivery system (DDS) with specific tumor-targeted GOx delivery, near-infrared (NIR) light and TME responsive O2 generation, NIR-responsive glucose consumption, high GOx loading, and efficient NIR photothermia was developed. Positively charged AuNRs@MnO2@SiO2 nanoparticles (named AMS+ NPs) were synthesized. GOx was covalently loaded with a high loading ratio of 36.0%. Finally, a thermosensitive biomimetic hybrid membrane composed of a thermosensitive lipid (TSL) membrane, red blood cell membrane (RBCM), and 4T1 cancer cell membrane (CCM) was coated on the NPs through a double-layer strategy. The AMS+-G@TSL@[RBC-CC-TSL]M NPs consumed 32.7 times glucose at 50 °C as that at 37 °C and generated 4.9 times O2 upon NIR laser irradiation. The thermosensitive biomimetic NPs showed an efficient targeting capability to the homotypic 4T1 cancer cells/tumors accompanied by good biocompatibility, macrophage evading capability, high cancer cell cytotoxicity, and excellent antitumor efficacy. The tumor growth inhibition ratio with NIR laser irradiation reached 92.8%. The AMS+-GOx@TSL@[RBC-CC-TSL]M NPs provide a smart, efficient, safe, PTT/CST combined DDS for highly efficient tumor therapy.

ETU-Net: efficient Transformer and convolutional U-style connected attention segmentation network applied to endoscopic image of epistaxis.

Epistaxis is a typical presentation in the otolaryngology and emergency department. When compressive therapy fails, directive nasal cautery is necessary, which strongly recommended operating under the nasal endoscope if it is possible. Limited by the operator's clinical experience, complications such as recurrence, nasal ulcer, and septum perforation may occur due to insufficient or excessive cautery. At present, deep learning technology is widely used in the medical field because of its accurate and efficient recognition ability, but it is still blank in the research of epistaxis. In this work, we first gathered and retrieved the Nasal Bleeding dataset, which was annotated and confirmed by many clinical specialists, filling a void in this sector. Second, we created ETU-Net, a deep learning model that smartly integrated the excellent performance of attention convolution with Transformer, overcoming the traditional model's difficulties in capturing contextual feature information and insufficient sequence modeling skills in picture segmentation. On the Nasal Bleeding dataset, our proposed model outperforms all others models that we tested. The segmentation recognition index, Intersection over Union, and F1-Score were 94.57 and 97.15%. Ultimately, we summarized effective ways of combining artificial intelligence with medical treatment and tested it on multiple general datasets to prove its feasibility. The results show that our method has good domain adaptability and has a cutting-edge reference for future medical technology development.

Room Temperature Spin-to-Charge Conversion in Amorphous Topological Insulating Gd-Alloyed BixSe1-x/CoFeB Bilayers.

Disordered topological insulator (TI) films have gained intense interest by benefiting from both the TI's exotic transport properties and the advantage of mass production by sputtering. Here, we report on the clear evidence of spin-charge conversion (SCC) in amorphous Gd-alloyed BixSe1-x (BSG)/CoFeB bilayers fabricated by sputtering, which could be related to the amorphous TI surface states. Two methods have been employed to study SCC in BSG (tBSG = 6-16 nm)/CoFeB(5 nm) bilayers with different BSG thicknesses. First, spin pumping is used to generate a spin current in CoFeB and detect SCC by the inverse Edelstein effect (IEE). The maximum SCC efficiency (SCE) is measured to be as large as 0.035 nm (IEE length λIEE) in a 6 nm thick BSG sample, which shows a strong decay when tBSG increases due to the increase of BSG surface roughness. The second method is THz time-domain spectroscopy, which reveals a small tBSG dependence of SCE, validating the occurrence of a pure interface state-related SCC. Furthermore, our angle-resolved photoemission spectroscopy data show dispersive two-dimensional surface states that cross the bulk gap until the Fermi level, strengthening the possibility of SCC due to the amorphous TI states. Our studies provide a new experimental direction toward the search for topological systems in amorphous solids.

Revealing the amyloid ß-protein with zinc finger protein of micronucleus during Alzheimer's disease progress by a quaternary ammonium terpyridine probe.

Micronucleus (MN) is regarded as an abnormal structure in eukaryotic cells which can be used as a biomarker for genetic instability. However, direct observation of MN in living cells is rarely achieved due to the lack of probes that are capable of distinguishing nuclear- and MN-DNA. Herein, a water-soluble terpyridine organic small molecule (ABT) was designed and employed to recognize Zinc-finger protein (ZF) for imaging intracellular MN. The in vitro experiments suggested ABT has a high affinity towards ZF. Further live cell staining showed that ABT could selectively target MN in HeLa and NSC34 cells when combined with ZF. Importantly, we use ABT to uncover the correlation between neurotoxic amyloid ß-protein (Aß) and MN during Alzheimer's disease (AD) progression. Thus, this study provides profound insight into the relationship between Aß and genomic disorders, offering a deeper understanding for the diagnosis and treatment of AD.

An Efficient Method for Monitoring Birds Based on Object Detection and Multi-Object Tracking Networks.

To protect birds, it is crucial to identify their species and determine their population across different regions. However, currently, bird monitoring methods mainly rely on manual techniques, such as point counts conducted by researchers and ornithologists in the field. This method can sometimes be inefficient, prone to errors, and have limitations, which may not always be conducive to bird conservation efforts. In this paper, we propose an efficient method for wetland bird monitoring based on object detection and multi-object tracking networks. First, we construct a manually annotated dataset for bird species detection, annotating the entire body and head of each bird separately, comprising 3737 bird images. We also built a new dataset containing 11,139 complete, individual bird images for the multi-object tracking task. Second, we perform comparative experiments using a state-of-the-art batch of object detection networks, and the results demonstrated that the YOLOv7 network, trained with a dataset labeling the entire body of the bird, was the most effective method. To enhance YOLOv7 performance, we added three GAM modules on the head side of the YOLOv7 to minimize information diffusion and amplify global interaction representations and utilized Alpha-IoU loss to achieve more accurate bounding box regression. The experimental results revealed that the improved method offers greater accuracy, with mAP@0.5 improving to 0.951 and mAP@0.5:0.95 improving to 0.815. Then, we send the detection information to DeepSORT for bird tracking and classification counting. Finally, we use the area counting method to count according to the species of birds to obtain information about flock distribution. The method described in this paper effectively addresses the monitoring challenges in bird conservation.

Geniposidic acid attenuates DSS-induced colitis through inhibiting inflammation and regulating gut microbiota.

Geniposidic acid (GPA) is a bioactive compound isolated from Gardenia jasminoides Ellis (Rubiaceae) that has long been used to treat arthritis, jaundice, and hypertension. However, the therapeutic effects of GPA against colitis remain underexplored. This study aimed to investigate the effect of GPA on the remission of colitis and the underlying mechanisms. A DSS-induced colitis mouse model was used to evaluate the influence of GPA on the modulation of gut microbiota and intestinal epithelial barrier function. Our results indicated that GPA improved DSS-induced mouse colitis, including loss of body weight, disease activity index (DAI), colon length, and colonic pathological damage. DSS-induced destruction of the intestinal barrier was also significantly repaired by GPA treatment. In addition, the relative levels of pro-inflammatory cytokines, such as IL-1ß and TNF-α, were markedly alleviated by GPA. Furthermore, western blot analysis revealed that GPA downregulated the protein expression of the nuclear transcription factor NF-κB. Finally, we first demonstrated that GPA could alleviate gut microbiota dysbiosis in mice with colitis by bacterial 16S rRNA sequencing. In conclusion, our study demonstrates the therapeutic and protective effects of GPA on IBD and provides novel insights into the prevention of colitis by targeting gut microbiota metabolism using natural products.

Irritation of Dental Sensory Nerves Promotes the Occurrence of Pulp Calcification.

INTRODUCTION: Pulp calcification (PC) often appears in strong association with nerve fiber bundles, which indicates the important role of dental nerves in the formation of PC. Additionally, given that sensory nerves and calcitonin gene-related peptide (CGRP) secreted from sensory nerve fibers are involved in physiological and pathological bone formation, we aimed to determine whether chronic irritation of sensory nerves can promote the occurrence of PC. METHODS: A sensory nerve irritation rat model was established via ligation of the inferior alveolar nerve (IAN), and face grooming behavior was analyzed as a measure of pain sensation. Two months post-surgery, PC was determined by imaging and histologic analyses. RESULTS: Rats in the IAN-chronic constriction injury (IAN-CCI) group showed spontaneous pain-associated behavior after the operations and pain tolerance on the 60th postoperative day. The imaging and histological analysis showed more calcified particles in the IAN-innervated first and second molars after day 60 of the dental sensory nerve irritation. These calcified masses had a dentin-like structure that contained sparse, irregularly oriented tubules. Compared to the control and sham groups, the odontoblasts located in the periphery of radicular pulp aligned along a thicker layer of predentin; which expressed more nestin with longer and stouter processes in the IAN-CCI group. Additionally, more CGRP-positive nerves were observed in the IAN-CCI group. CONCLUSIONS: Irritation of sensory nerves promotes PC formation, and the increased density of CGRP-immunolabeled fibers probably contributes to this process. This highlights the significance of dental sensory nerves in the formation of PC.

PAC-Net: Multi-pathway FPN with position attention guided connections and vertex distance IoU for 3D medical image detection.

Automatic medical image detection aims to utilize artificial intelligence techniques to detect lesions in medical images accurately and efficiently, which is one of the most important tasks in computer-aided diagnosis (CAD) systems, and can be embedded into portable imaging devices for intelligent Point of Care (PoC) Diagnostics. The Feature Pyramid Networks (FPN) based models are widely used deep-learning-based solutions for automatic medical image detection. However, FPN-based medical lesion detection models have two shortcomings: the object position offset problem and the degradation problem of IoU-based loss. Therefore, in this work, we propose a novel FPN-based backbone model, i.e., Multi-Pathway Feature Pyramid Networks with Position Attention Guided Connections and Vertex Distance IoU (abbreviated as PAC-Net), to replace vanilla FPN for more accurate lesion detection, where two innovative improvements, a position attention guided connection (PAC) module and Vertex Distance IoU Vertex Distance Intersection over Union loss, are proposed to address the above-mentioned shortcomings of vanilla FPN, respectively. Extensive experiments are conducted on a public medical image detection dataset, i.e., Deeplesion, and the results showed that i) PAC-Net outperforms all state-of-the-art FPN-based depth models in both evaluation metrics of lesion detection on the DeepLesion dataset, ii) the proposed PAC module and VDIoU loss are both effective and important for PAC-Net to achieve a superior performance in automatic medical image detection tasks, and iii) the proposed VDIoU loss converges more quickly than the existing IoU-based losses, making PAC-Net an accurate and also highly efficient 3D medical image detection model.

Painless and accurate medical image analysis using deep reinforcement learning with task-oriented homogenized automatic pre-processing.

Pre-processing is widely applied in medical image analysis to remove the interference information. However, the existing pre-processing solutions mainly encounter two problems: (i) it is heavily relied on the assistance of clinical experts, making it hard for intelligent CAD systems to deploy quickly; (ii) due to the personnel and information barriers, it is difficult for medical institutions to conduct the same pre-processing operations, making a deep model that performs well on a specific medical institution difficult to achieve similar performances on the same task in other medical institutions. To overcome these problems, we propose a deep-reinforcement-learning-based task-oriented homogenized automatic pre-processing (DRL-HAPre) framework to overcome these two problems. This framework utilizes deep reinforcement learning techniques to learn a policy network to automatically and adaptively select the optimal pre-processing operations for the input medical images according to different analysis tasks, thus helping the intelligent CAD system to achieve a rapid deployment (i.e., painless) and maintain a satisfactory performance (i.e., accurate) among different medical institutes. To verify the effectiveness and advantages of the proposed DRL-HAPre framework, we further develop a homogenized automatic pre-processing model based on the DRL-HAPre framework to realize the automatic pre-processing of key region selection (called HAPre-KRS) in the pneumonia image classification task. Extensive experimental studies are conducted on three pediatric pneumonia classification datasets with different image qualities, and the results show that: (i) There does exist a hard-to-reproduce problem in clinical practices and the fact that having different medical image qualities in different medical institutes is an important reason for the existing of hard-to-reproduce problem, so it is compelling to propose homogenized automatic pre-processing method. (ii) The proposed HAPre-KRS model and DRL-HAPre framework greatly outperform three kinds of state-of-the-art baselines (i.e., pre-processing, attention and pneumonia baseline), and the lower the medical image quality, the greater the improvements of using our HAPre-KRS model and DRL-HAPre framework. (iii) With the help of homogenized pre-processing, HAPre-KRS (and DRL-HAPre framework) can greatly avoid performance degradation in real-world cross-source applications (i.e., thus overcoming the hard-to-reproduce problem).

Rethinking Maximum Mean Discrepancy for Visual Domain Adaptation.

Existing domain adaptation approaches often try to reduce distribution difference between source and target domains and respect domain-specific discriminative structures by some distribution [e.g., maximum mean discrepancy (MMD)] and discriminative distances (e.g., intra-class and inter-class distances). However, they usually consider these losses together and trade off their relative importance by estimating parameters empirically. It is still under insufficient exploration so far to deeply study their relationships to each other so that we cannot manipulate them correctly and the model's performance degrades. To this end, this article theoretically proves two essential facts: 1) minimizing MMD equals to jointly minimizing their data variance with some implicit weights but, respectively, maximizing the source and target intra-class distances so that feature discriminability degrades and 2) the relationship between intra-class and inter-class distances is as one falls and another rises. Based on this, we propose a novel discriminative MMD with two parallel strategies to correctly restrain the degradation of feature discriminability or the expansion of intra-class distance; specifically: 1) we directly impose a tradeoff parameter on the intra-class distance that is implicit in the MMD according to 1) and 2) we reformulate the inter-class distance with special weights that are analogical to those implicit ones in the MMD and maximizing it can also lead to the intra-class distance falling according to 2). Notably, we do not consider the two strategies in one model due to 2). The experiments on several benchmark datasets not only prove the validity of our revealed theoretical results but also demonstrate that the proposed approach could perform better than some compared state-of-art methods substantially. Our preliminary MATLAB code will be available at https://github.com/WWLoveTransfer/.

A Multiscale Lightweight and Efficient Model Based on YOLOv7: Applied to Citrus Orchard.

With the gradual increase in the annual production of citrus, the efficiency of human labor has become the bottleneck limiting production. To achieve an unmanned citrus picking technology, the detection accuracy, prediction speed, and lightweight deployment of the model are important issues. Traditional object detection methods often fail to achieve balanced effects in all aspects. Therefore, an improved YOLOv7 network model is proposed, which introduces a small object detection layer, lightweight convolution, and a CBAM (Convolutional Block Attention Module) attention mechanism to achieve multi-scale feature extraction and fusion and reduce the number of parameters of the model. The performance of the model was tested on the test set of citrus fruit. The average accuracy (mAP@0.5) reached 97.29%, the average prediction time was 69.38 ms, and the number of parameters and computation costs were reduced by 11.21 M and 28.71 G compared with the original YOLOv7. At the same time, the Citrus-YOLOv7 model's results show that it performs better compared with the current state-of-the-art network models. Therefore, the proposed Citrus-YOLOv7 model can contribute to solving the problem of citrus detection.

A ratiometric fluorescence method based on nitrogen-doped carbon quantum dots for the determination of the activity of alkaline phosphatase.

Herein, a simple and sensitive ratiometric fluorescence sensing platform to detect alkaline phosphatase (ALP) activity is developed on the basis of yellow fluorescent nitrogen-doped carbon quantum dots (YNCDs). The hydrolysis of ascorbic acid 2-phosphate (AAP) into ascorbic acid (AA) is catalyzed by ALP. Then, AA will react with o-phenylenediamine (OPD) to form 3-(1,2-dihydroxyethyl)furo[3,4b]-quinoxaline (QXD) which is a blue fluorescent quinoxaline derivative with emission at 435 nm in the presence of Cu2+. YNCDs have yellow fluorescence emission at 555 nm, and can maintain stable in QXD reaction system. Therefore, by utilizing the fluorescence of YNCDs at 555 nm as reference signal and the fluorescence of QXD at 435 nm as report signal, we can detect the ALP activity by monitoring the fluorescence ratio (F435/F555). The linear range is 0.5-5 U/L, and the limit of detection is 0.14 U/L. An application of this method for the analysis of ALP in human serum has given satisfactory results. A ratiometric fluorescent nanoprobe for ascorbic acid and alkaline phosphatase detection with excellent biocompatible and high sensitivity was successfully constructed based on YNCDs and QXD.