KBMeasure: small sample in-situ damage automatic measurement method based on the combination of keypoints detection and binocular 3D reconstruction for aero-engine blades.

Size is one of the important bases for the level assessment of aero-engine blade damage and the disposal method selection for damaged blades. Therefore, research on in-situ damage measurement of aero-engine blades is conducted in this paper. We break the inherent pipeline of "3D reconstruction and manual annotation of keypoints" in traditional damage measurement methods, and propose an in-situ damage automatic measurement method (KBMeasure) based on the combination of damage keypoints intelligent detection and binocular 3D reconstruction. KBMeasure replaces the manual annotation of damage keypoints, improves the damage measurement efficiency, and reduces the dependence on professional inspectors. The proposed method also overcomes the problem of high computational cost and low efficiency caused by redundant 3D reconstruction of the entire damaged area. For the characteristics of large changes in damage scale, low image resolution, the requirement of high-precision keypoints positioning, limited annotated data, and lightweight deployment in aero-enginge blade damage measurement task, a novel blade damage keypoints detection model (DKeyDet) with top-down framework is designed by introducing coordinate classification, semi-supervised learning, and knowledge distillation. Then, intersecting optical axis binocular model is used to estimate the spatial coordinates of the detected keypoints and compute the size of damage. The keypoints detection average precision (AP) and average recall (AR) of our method are 87.6 and 91.3, and the damage measurement size error (SE) is 0.08, which is superior to existing methods. This research provides a new theoretical support for in-situ damage automatic measurement for aero-engine in service, and provides what we believe is a novel idea for damage measurement of industrial components in other fields.

Immunotherapy-related cognitive impairment after CAR T cell therapy in mice.

Persistent central nervous system (CNS) immune dysregulation and consequent dysfunction of multiple neural cell types is central to the neurobiological underpinnings of a cognitive impairment syndrome that can occur following traditional cancer therapies or certain infections. Immunotherapies have revolutionized cancer care for many tumor types, but the potential long-term cognitive sequelae are incompletely understood. Here, we demonstrate in mouse models that chimeric antigen receptor (CAR) T cell therapy for both CNS and non-CNS cancers can impair cognitive function and induce a persistent CNS immune response characterized by white matter microglial reactivity and elevated cerebrospinal fluid (CSF) cytokines and chemokines. Consequently, oligodendroglial homeostasis and hippocampal neurogenesis are disrupted. Microglial depletion rescues oligodendroglial deficits and cognitive performance in a behavioral test of attention and short-term memory function. Taken together, these findings illustrate similar mechanisms underlying immunotherapy-related cognitive impairment (IRCI) and cognitive impairment following traditional cancer therapies and other immune challenges.

Nf1 mutation disrupts activity-dependent oligodendroglial plasticity and motor learning in mice.

Neurogenetic disorders, such as neurofibromatosis type 1 (NF1), can cause cognitive and motor impairments, traditionally attributed to intrinsic neuronal defects such as disruption of synaptic function. Activity-regulated oligodendroglial plasticity also contributes to cognitive and motor functions by tuning neural circuit dynamics. However, the relevance of oligodendroglial plasticity to neurological dysfunction in NF1 is unclear. Here we explore the contribution of oligodendrocyte progenitor cells (OPCs) to pathological features of the NF1 syndrome in mice. Both male and female littermates (4-24 weeks of age) were used equally in this study. We demonstrate that mice with global or OPC-specific Nf1 heterozygosity exhibit defects in activity-dependent oligodendrogenesis and harbor focal OPC hyperdensities with disrupted homeostatic OPC territorial boundaries. These OPC hyperdensities develop in a cell-intrinsic Nf1 mutation-specific manner due to differential PI3K/AKT activation. OPC-specific Nf1 loss impairs oligodendroglial differentiation and abrogates the normal oligodendroglial response to neuronal activity, leading to impaired motor learning performance. Collectively, these findings show that Nf1 mutation delays oligodendroglial development and disrupts activity-dependent OPC function essential for normal motor learning in mice.

BPMTrack: Multi-Object Tracking With Detection Box Application Pattern Mining.

The key to multi-object tracking is its stability and the retention of identity information. A common problem with most detection-based approaches is trusting and using all the detector outputs for the association. However, some settings of detectors can affect stable long-range tracking. Based on the principle of reducing the association noise in the detection processing step, we propose a new framework, the Box application Pattern Mining Tracker (BPMTrack), to address this issue. Specifically, we worked on three main aspects: output threshold, association strategy, and motion model. Due to the problem of inconsistency between classification scores and localization accuracy, we propose the Box Quality Estimation Network (BQENet) to predict the localization quality scores of all detections in the current frame, reserving high-quality boxes for the tracker. In addition, based on observations of intensive scenarios, we propose a simple and effective data association method, the Non-Maximum Suppression Integration (NMSI) matching strategy. It recovers the Non-Maximum Suppression (NMS) detection, inputs them into BQENet, and then performs hierarchical matching with reasonable control of box priority to alleviate the problem of absent objects caused by occlusion. Finally, we propose an improved Measurement Correct and Noise Scale (MCNS) Kalman algorithm to improve the prediction accuracy of object positions and, thus, the association quality. We performed an extensive ablation evaluation of the proposed framework to prove its effectiveness. Moreover, the three tracking benchmarks show our method's accuracy and long-distance performance.

Isorhamnetin Regulates Programmed Death Ligand-1 Expression by Suppressing the EGFR-STAT3 Signaling Pathway in Canine Mammary Tumors.

Programmed death ligand-1 (PD-L1) is highly expressed in a variety of cancer cells and suggests a poorer prognosis for patients. The natural compound isorhamnetin (ISO) shows promise in treating cancers and causing damage to canine mammary tumor (CMT) cells. We investigated the mechanism of ISO in reducing PD-L1 expression in CMT cells. Clustered, regularly interspaced short palindromic repeat-associated protein 9 (CRISPR/Cas9) was used to mediate CD274 knockout in U27 cells. Then, monoclonal cells were screened and cultured. Nucleotide sequencing and expression of PD-L1 were detected. Additionally, we examined cell migration, invasion, and damage. Immunofluorescent staining of PD-L1 was examined in U27 cells. The signaling pathways were measured by Western blotting. Murine xenotransplantation models and murine immunocompetent allograft mammary tumor models were established to evaluate the effect of ISO therapy. Expression of Ki-67, caspase3, and PD-L1 were analyzed by immunohistochemistry. A pull-down assay was used to explore which proteins could bind to ISO. Canine EGFR protein was purified and used to detect whether it directly binds to ISO using a surface plasmon resonance assay. ISO inhibited the EGFR-STAT3-PD-L1 signaling pathway and blocked cancer growth, significantly increasing the survival rate of healthy cells. The cell membrane receptor EGFR was identified as a direct target of ISO. ISO could be exploited as an antineoplastic treatment of CMT by targeting EGFR to suppress PD-L1 expression.

Single-cell landscape identifies the immunophenotypes and microenvironments of HBV-positive and HBV-negative liver cancer.

BACKGROUND: HBV infection leads to HCC and affects immunotherapy. We are exploring the tumor ecosystem in HCC to help gain a deeper understanding and design more effective immunotherapy strategies for patients with HCC with or without HBV infection. METHODS: Single-cell RNA sequencing series were integrated as a discovery cohort to interrogate the tumor microenvironment of HBV-positive (HBV+) HCC and HBV-negative (HBV-) HCC. We further dissect the intratumoral immune status of HBV+ HCC and HBV- HCC. An independent cohort, including samples treated with immune checkpoint blockade therapy, was used to validate the major finding and investigate the effect of HBV infection on response to immunotherapy. RESULTS: The interrogation of tumor microenvironment indicated that regulatory T cells, exhausted CD8+ T cells, and M1-like Macrophage_MMP9 were enriched in HBV+ HCC, while mucosa-associated invariant T cells were enriched in HBV- HCC. All subclusters of T cells showed high expression of immune checkpoint genes in HBV+ HCC. Regulatory T cells enriched in HBV+ HCC also showed more robust immunosuppressive properties, which was confirmed by cross talk between immune cell subsets. The ability of antigen presentation with major histocompatibility complex-II was downregulated in HBV+ HCC and this phenomenon can be reversed by immunotherapy. Two types of HCC also present different responses to immunotherapy. CONCLUSIONS: There is a more immunosuppressive and exhausted tumor microenvironment in HBV+ HCC than in HBV- HCC. This in-depth immunophenotyping strategy is critical to understanding the impact of HBV and the HCC immune microenvironment and helping develop more effective treatments in patients with HCC.

PLOD3 facilitated T cell activation in the colorectal tumor microenvironment and liver metastasis by the TNF-α/ NF-κB pathway.

BACKGROUND: Colorectal cancer (CRC) has been the third most prevalent cancer worldwide. Liver metastasis is the critical factor for the poor prognosis of CRC. Here, we investigated the expression and role of PLOD3 in CRC. METHODS: Different liver metastasis models were established by injecting PLOD3 stable knockdown or overexpression CT26 or MC38 mouse CRC cells into the spleen of mice to verify the tumorigenicity and metastasis ability in vivo. RESULTS: We identified PLOD3 is significantly overexpressed in liver metastasis samples of CRC. High expression of PLOD3 was significantly associated with poor survival of CRC patients. The knockdown of PLOD3 exhibited remarkable inhibition of proliferation, migration, and invasion in CRC cells, while the opposite results could be found in different PLOD3-overexpressed CRC cells. Stable knockdown of PLOD3 also significantly inhibited liver metastasis of CRC cells in different xenografts models, while stable overexpression of PLOD3 promotes liver metastasis and tumor progression. Further studies showed that PLOD3 facilitated the T cell activation in the tumor microenvironment and affected the TNF-α/ NF-κB pathway. CONCLUSIONS: This study revealed the essential biological functions of PLOD3 in colon cancer progression and metastasis, suggesting that PLOD3 is a promising translational medicine target and bioengineering targeting PLOD3 overcomes CRC liver metastasis.

Homoharringtonine induces apoptosis of mammary carcinoma cells by inhibiting the AKT/mTOR signaling pathway.

Mammary tumour is the most common type of tumour in dogs, especially in unneutered female dogs. Homoharringtonine (HHT) is a natural alkaloid that can be used to treat various types of human tumour. However, the inhibitory effect and mechanism of HHT on canine mammary carcinomas (CMC) remain unclear. This study aimed to evaluate the inhibitory effect of HHT on CMC in vitro and determine its underlying molecular mechanism. The effects of HHT on the cytotoxicity of CMC U27 cells were evaluated by the cell counting kit-8, wound healing, and Transwell assays. HHT-induced apoptosis of U27 cells was detected by JC-1 and terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) assay. Moreover, the gene expression of B-cell lymphoma-2 (Bcl-2) and Bcl-2 associated X protein (Bax) were analysed using quantitative reverse transcription-polymerase chain reaction (RT-qPCR), and the protein expression of protein kinase B/mammalian target of rapamycin (AKT/mTOR) and mitochondrial apoptosis proteins were determined by western blotting. Furthermore, mammary tumour-bearing mouse models were established using 4T1 cells to evaluate the therapeutic effect of HHT. It was found that HHT could significantly down-regulated the protein expression of p-AKT, p-mTOR, and Bcl-2, and up-regulated the protein expression of P53, Bax, cleaved caspase-3, and cleaved caspase-9. In addition, HHT significantly suppressed both tumour volume and mass in mammary tumour mice. In conclusion, HHT damages CMC cells by inhibiting the AKT/mTOR signalling pathway and inducing mitochondrial apoptosis. Such findings lay a theoretical foundation for the clinical treatment of CMC and provide more options for clinical medication.

ALKBH5 Stabilized N6-Methyladenosine-Modified LOC4191 to Suppress E. coli-Induced Apoptosis.

E. coli is a ubiquitous pathogen that is responsible for over one million fatalities worldwide on an annual basis. In animals, E. coli can cause a variety of diseases, including mastitis in dairy cattle, which represents a potential public health hazard. However, the pathophysiology of E. coli remains unclear. We found that E. coli could induce global upregulation of m6A methylation and cause serious apoptosis in bovine mammary epithelial cells (MAC-T cells). Furthermore, numerous m6A-modified lncRNAs were identified through MeRIP-seq. Interestingly, we found that the expression of LOC4191 with hypomethylation increased in MAC-T cells upon E. coli-induced apoptosis. Knocking down LOC4191 promoted E. coli-induced apoptosis and ROS levels through the caspase 3-PARP pathway. Meanwhile, knocking down ALKBH5 resulted in the promotion of apoptosis through upregulated ROS and arrested the cell cycle in MAC-T cells. ALKBH5 silencing accelerated LOC4191 decay by upregulating its m6A modification level, and the process was recognized by hnRNP A1. Therefore, this indicates that ALKBH5 stabilizes m6A-modified LOC4191 to suppress E. coli-induced apoptosis. This report discusses an initial investigation into the mechanism of m6A-modified lncRNA in cells under E. coli-induced apoptosis and provides novel insights into infectious diseases.

RHOJ Induces Epithelial-to-Mesenchymal Transition by IL-6/STAT3 to Promote Invasion and Metastasis in Gastric Cancer.

Background: Recently, the molecular classification of gastric cancer (GC) promotes the advances of GC patients' precision therapy and prognosis prediction. According to the Asian Cancer Research Group (ACRG), GC is classified as microsatellite instable (MSI) subtype GC, microsatellite stable/epithelial-to-mesenchymal transition (MSS/EMT) subtype GC, MSS/TP53- subtype GC, and MSS/TP53+ subtype GC. Due to the easy metastasis of EMT-subtype GC, it has the worst prognosis, the highest recurrence rate, and the tendency to occur at a younger age. Therefore, it is curious and crucial for us to understand the molecular basis of EMT-subtype GC. Methods: The expression of RHOJ was detected by quantitative real-time PCR (qPCR) and immunohistochemistry (IHC) in GC cells and tissues. Western blotting and immunofluorescence (IF) were conducted to examine the effects of RHOJ on the EMT markers' expression of GC cells. The GC cells' migration and invasion were investigated by transwell assay. The tumor growth and metastasis were demonstrated correspondingly in different xenograft models. Results: Firstly, it was noticed that RHOJ was significantly upregulated in EMT-subtype GC and RHOJ has close relationships with the EMT process of GC, based on the Gene Expression Omnibus (GEO) and the Cancer Genome Atlas (TCGA) databases. Next, transwell assay and tail vein metastasis models were conducted to verify that RHOJ mediates the EMT to regulate the invasion and metastasis of GC in vitro and in vivo. In addition, weakened tumor angiogenesis was observed after RHOJ knockdown by the angiogenesis assay of HUVEC. RNA-seq and further study unveiled that RHOJ aggravates the malignant progression of GC by inducing EMT through IL-6/STAT3 to promote invasion and metastasis. Finally, blocking the IL-6/STAT3 signaling overcame RHOJ-mediated GC cells' growth and migration. Conclusions: These results indicate that the upregulation of RHOJ contributes to EMT-subtype GC invasion and metastasis via IL-6/STAT3 signaling, and RHOJ is expected to become a promising biomarker and therapeutic target for EMT-subtype GC patients.

Identification and validation of a novel necroptosis-related molecular signature to evaluate prognosis and immune features in breast cancer.

Necroptosis has been shown to play an important role in the development of tumors. However, the characteristics of the necroptosis-related subtypes and the associated immune cell infiltration in the tumor microenvironment (TME) of breast cancer (BRCA) remain unclear. In this study, we identified three clusters related to necroptosis using the expression patterns of necroptosis-relevant genes (NRGs), and found that these three clusters had different clinicopathological features, prognosis and immune cell infiltration in the TME. Cluster 2 was characterized by less infiltration of immune cells in the TME and was associated with a worse prognosis. Then, a necroptosis risk score (NRS) composed of 14 NRGs was constructed using the least absolute shrinkage and selection operator regression (LASSO) Cox regression method. Based on NRS, all BRCA patients in the TCGA datasets were classified into a low-risk group and a high-risk group. Patients in the low-risk group were characterized by longer overall survival (OS), lower mutation burden, and higher infiltration level of immune cells in the TME. Moreover, the NRS was significantly associated with chemotherapeutic drug sensitivity. Finally, the knockdown of VDAC1 reduced the proliferation and migration of BRCA cells, and promoted cell death induced by necroptosis inducer. This study identified a novel necroptosis-related subtype of BRCA, and a comprehensive analysis of NRGs in BRCA revealed its potential roles in prognosis, clinicopathological features, TME, chemotherapy, tumor proliferation, and tumor necroptosis. These results may improve our understanding of NRGs in BRCA and provide a reference for developing individualized therapeutic strategies.

Application of Levosimendan in Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Infection Patients with Myocardial Injury.

Objective: This study aims to investigate the effectiveness of levosimendan in patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection complicated by cardiac insufficiency and myocardial injury. Methods: A total of 22 patients with SARS-CoV-2 infection and myocardial injury, admitted to the Cardiology Department of our hospital between December 2022 and February 2023, are randomly divided into two groups: a dobutamine general treatment group and a levosimendan observation treatment group. The treatment outcomes of the two groups are compared and analyzed. Results: The overall improvement rate in the general treatment group is 80%, while the levosimendan treatment group shows a 100% improvement rate. There is a statistically significant difference between the two groups (P < .05). Post-treatment, the left ventricular ejection fraction for the general treatment group and the levosimendan treatment group are (48 ± 7)% and (54 ± 6)%, respectively. Additionally, the left ventricular end-diastolic diameter is (55.0 ± 3.0) mm in the general treatment group and (51 ± 5.0) mm in the levosimendan group, with a statistically significant difference (P < .05). After active treatment, the plasma levels of B-type natriuretic peptide (Brain Natriuretic Peptide, NT-proBNP) are significantly lower in the levosimendan treatment group than in the general treatment group (P < .05). Moreover, the plasma levels of interleukin-6 (IL-6) and C-reactive protein (CRP) in the levosimendan group decrease slightly faster than those in the general treatment group, with a statistically significant difference (P < .05). The length of hospital stay in the levosimendan group is (12 ± 3) days, significantly lower than the general treatment group (16 ± 5) days, with a statistically different result (P < .05). Conclusions: Levosimendan demonstrates significant efficacy in patients with novel coronavirus infection complicated by myocardial injury, resulting in improved clinical symptoms, enhanced cardiac function, shorter hospital stays, early discharge, and cost savings.

STING-IRG1 inhibits liver metastasis of colorectal cancer by regulating the polarization of tumor-associated macrophages.

The liver is the main site of colorectal cancer (CRC) metastasis. Tumor-associated macrophages (TAMs) play a key role in tumor metastasis. Therefore, modulating the function of tumor-associated macrophages is a potential therapeutic strategy to control tumor metastasis. We found in vivo experiments that the activation of STING inhibited CRC liver metastasis in model mice and affected the macrophage phenotype in the tumor microenvironment. Mechanistically, STING affects TAM polarization and regulates macrophage function through IRG1. And STING activates IRG1 to promote the nuclear translocation of TFEB, affecting the ability of macrophages to suppress tumor metastasis.Therefore, this study highlights the critical role of the STING-IRG1 axis on TAM reprogramming and its role in the process of tumor liver metastasis, which may provide a promising therapeutic strategy for CRC liver metastasis.

Generic User Behavior: A User Behavior Similarity-Based Recommendation Method.

Recommender system (RS) plays an important role in Big Data research. Its main idea is to handle huge amounts of data to accurately recommend items to users. The recommendation method is the core research content of the whole RS. However, the existing recommendation methods still have the following two shortcomings: (1) Most recommendation methods use only one kind of information about the user's interaction with items (such as Browse or Purchase), which makes it difficult to model complete user preference. (2) Most mainstream recommendation methods only consider the final consistency of recommendation (e.g., user preferences) but ignore the process consistency (e.g., user behavior), which leads to the biased final result. In this article, we propose a recommendation method based on the Entity Interaction Knowledge Graph (EIKG), which draws on the idea of collaborative filtering and innovatively uses the similarity of user behaviors to recommend items. The method first extracts fact triples containing interaction relations from relevant data sets to generate the EIKG; then embeds the entities and relations in the EIKG; finally, uses link prediction techniques to recommend items for users. The proposed method is compared with other recommendation methods on two publicly available data sets, Scholat and Lizhi, and the experimental result shows that it exceeds the state of the art in most metrics, verifying the effectiveness of the proposed method.

Multifunctional Protein Hybrid Nanoplatform for Synergetic Photodynamic-Chemotherapy of Malignant Carcinoma by Homologous Targeting Combined with Oxygen Transport.

Photodynamic therapy (PDT) under hypoxic conditions and drug resistance in chemotherapy are perplexing problems in anti-tumor treatment. In addition, central nervous system neoplasm-targeted nanoplatforms are urgently required. To address these issues, a new multi-functional protein hybrid nanoplatform is designed, consisting of transferrin (TFR) as the multicategory solid tumor recognizer and hemoglobin for oxygen supply (ODP-TH). This protein hybrid framework encapsulates the photosensitizer protoporphyrin IX (PpIX) and chemotherapeutic agent doxorubicin (Dox), which are attached by a glutathione-responsive disulfide bond. Mechanistically, ODP-TH crosses the blood-brain barrier (BBB) and specifically aggregated in hypoxic tumors via protein homology recognition. Oxygen and encapsulated drugs ultimately promote a therapeutic effect by down-regulating the abundance of multidrug resistance gene 1 (MDR1) and hypoxia-inducible factor-1-α (HIF-1α). The results reveal that ODP-TH achieves oxygen transport and protein homology recognition in the hypoxic tumor occupation. Indeed, compared with traditional photodynamic chemotherapy, ODP-TH achieves a more efficient tumor-inhibiting effect. This study not only overcomes the hypoxia-related inhibition in combination therapy by targeted oxygen transport but also achieves an effective treatment of multiple tumors, such as breast cancer and glioma, providing a new concept for the construction of a promising multi-functional targeted and intensive anti-tumor nanoplatform.

The Role of N6-Methyladenosine in Inflammatory Diseases.

N6-Methyladenosine (m6A) is the most abundant epigenetic RNA modification in eukaryotes, regulating RNA metabolism (export, stability, translation, and decay) in cells through changes in the activity of writers, erasers, and readers and ultimately affecting human life or disease processes. Inflammation is a response to infection and injury in various diseases and has therefore attracted significant attention. Currently, extensive evidence indicates that m6A plays an essential role in inflammation. In this review, we focus on the mechanisms of m6A in inflammatory autoimmune diseases, metabolic disorder, cardio-cerebrovascular diseases, cancer, and pathogen-induced inflammation, as well as its possible role as targets for clinical diagnosis and treatment.

Clinical Prognostic Factors and Integrated Multi-Omics Studies Identify Potential Novel Therapeutic Targets for Pediatric Desmoid Tumor.

BACKGROUND: Desmoid tumor (DT), also known as desmoid-type fibromatosis (DTF) or aggressive fibromatosis (AF) is a rare mesenchymal tumor affecting both children and adults. It is non-metastasis but infiltrative, growing with a high recurrence rate to even cause serious health problems. This study investigates the biology of desmoid tumors through integrated multi-omics studies. METHODS: We systematically investigated the clinical data of 98 extra-abdominal cases in our pediatric institute and identified some critical clinical prognostic factors. Moreover, our integrated multi-omics studies (Whole Exome Sequencing, RNA sequencing, and untargeted metabolomics profiling) in the paired PDT tumor/matched normal tissues identified more novel mutations, and potential prognostic markers and therapeutic targets for PDTs. RESULTS: The top mutation genes, such as CTNNB1 (p.T41A and p.S45F) and MUC4 (p.T3775T, p.S3450S, etc.), were observed with a mutation in more than 40% of PDT patients. We also identified a panel of genes that are classed as the FDA-approved drug targets or Wnt/ß-catenin signaling pathway-related genes. The integrated analysis identified pathways and key genes/metabolites that may be important for developing potential treatment of PDTs. We also successfully established six primary PDT cell lines for future studies. CONCLUSIONS: These studies may promote the development of novel drugs and therapeutic strategies for PDTs.

Surface roughness prediction of aircraft after coating removal based on optical image and deep learning.

To quickly evaluate the surface quality of aircraft after coating removal, a surface roughness prediction method based on optical image and deep learning model is proposed. In this paper, the "optical image-surface roughness" data set is constructed, and SSEResNet for regression prediction of surface roughness is designed by using feature fusion method. SSEResNet can effectively extract the detailed features of optical images, and Adam method is used for training optimization. Experiments show that the proposed model outperforms the other seven CNN backbone networks compared. This paper also investigates the effect of four different learning rate decay strategies on model training and prediction performance. The results show that the learning rate decay method of Cosine Annealing with warm restart has the best effect, its test MAE value is 0.245 µm, and the surface roughness prediction results are more consistent with the real value. The work of this paper is of great significance to the removal and repainting of aircraft coatings.

Adaptive and maladaptive myelination in health and disease.

Within the past decade, multiple lines of evidence have converged to identify a critical role for activity-regulated myelination in tuning the function of neural networks. In this Review, we provide an overview of accumulating evidence that activity-regulated myelination is required for brain adaptation and learning across multiple domains. We then discuss dysregulation of activity-dependent myelination in the context of neurological disease, a novel frontier with the potential to uncover new mechanisms of disease pathogenesis and to develop new therapeutic strategies. Alterations in myelination and neural network function can result from deficient myelin plasticity that impairs neurological function or from maladaptive myelination, in which intact activity-dependent myelination contributes to the disease process by promoting pathological patterns of neuronal activity. These emerging mechanisms suggest new avenues for therapeutic intervention that could more fully address the complex interactions between neurons and oligodendroglia.

An Object Point Set Inductive Tracker for Multi-Object Tracking and Segmentation.

Multi-object tracking and segmentation (MOTS) is a derivative task of multi-object tracking (MOT). The new setting encourages the learning of more discriminative high-quality embeddings. In this paper, we focus on the problem of exploring the relationship between the segmenter and the tracker, and propose an efficient Object Point set Inductive Tracker (OPITrack) based on it. First, we discover that after a single attention layer, the high-dimensional, key point embedding will show feature averaging. To alleviate this phenomenon, we propose an embedding generalization training strategy for sparse training and dense testing. This strategy allows the network to increase randomness in training and encourages the tracker to learn more discriminative features. In addition, to learn the desired embedding space, we propose a general Trip-hard sample augmentation loss. The loss uses patches that are not distinguishable by the segmenter to join the feature learning and force the embedding network to learn the difference between false positives and true positives. Our method was validated on two MOTS benchmark datasets and achieved promising results. In addition, our OPITrack can achieve better performance for the raw model while costing less video memory (VRAM) at training time.