GLE-net: global-local information enhancement for semantic segmentation of remote sensing images.

Remote sensing (RS) images contain a wealth of information with expansive potential for applications in image segmentation. However, Convolutional Neural Networks (CNN) face challenges in fully harnessing the global contextual information. Leveraging the formidable capabilities of global information modeling with Swin-Transformer, a novel RS images segmentation model with CNN (GLE-Net) was introduced. This integration gives rise to a revamped encoder structure. The subbranch initiates the process by extracting features at varying scales within the RS images using the Multiscale Feature Fusion Module (MFM), acquiring rich semantic information, discerning localized finer features, and adeptly handling occlusions. Subsequently, Feature Compression Module (FCM) is introduced in main branch to downsize the feature map, effectively reducing information loss while preserving finer details, enhancing segmentation accuracy for smaller targets. Finally, we integrate local features and global features through Spatial Information Enhancement Module (SIEM) for comprehensive feature modeling, augmenting the segmentation capabilities of model. We performed experiments on public datasets provided by ISPRS, yielding notably remarkable experimental outcomes. This underscores the substantial potential of our model in the realm of RS image segmentation within the context of scientific research.

Multisample lipidomic profiles of irritable bowel syndrome and irritable bowel syndrome-like symptoms in patients with inflammatory bowel disease: new insight into the recognition of the same symptoms in different diseases.

BACKGROUND: Overlapping clinical manifestations of irritable bowel syndrome (IBS) and IBS-like symptoms in patients with inflammatory bowel disease (IBD-IBS) present challenges in diagnosis and management. Both conditions are associated with alterations in metabolites, but few studies have described the lipid profiles. Our aim was to pinpoint specific lipids that contribute to the pathogenesis of IBS and IBD-IBS by analyzing multiple biologic samples. METHODS: Diarrhea-predominant IBS (IBS-D) patients (n = 39), ulcerative colitis in remission with IBS-like symptoms patients (UCR-IBS) (n = 21), and healthy volunteers (n = 35) were recruited. IBS-D patients meet the Rome IV diagnostic criteria, and UCR-IBS patients matched mayo scores ≤ two points and Rome IV diagnostic criteria. Serum, feces, and mucosa were collected for further analysis. Lipid extraction was carried out by ultra-performance liquid chromatography-high resolution mass spectrometry (UPLC-HRMS). RESULTS: Lipidomics of mucosa and serum samples significantly differed among the three groups. Feces showed the most altered lipid species, and the enrichment analysis of 347 differentially abundant metabolites via KEGG pathway analysis revealed that alpha-linolenic acid metabolism was significantly altered in the two groups (P < 0.01). The ratio of omega-6/omega-3 fatty acid were imbalance in serum samples. CONCLUSIONS: This study revealed a comprehensive lipid composition pattern between IBS-D patients and UCR-IBS patients. We found several distinctive lipids involved in alpha-linolenic acid metabolism, reflecting an imbalance in the omega-6/omega-3 fatty acid ratio. Compared to mucosa and serum samples, fecal samples might have more advantages in lipidomics studies due to the convenience of sample collection and effectiveness in reflecting metabolic information.

Good syndrome combined with multiple microbial pulmonary infections: case report and review of the literature.

Good syndrome (GS), a rare acquired immunodeficiency disorder characterized by thymoma and hypogammaglobulinemia, predisposes individuals to recurrent infections. This study reports a case of a 37-year-old male GS with multiple pulmonary infections and reviews relevant literature. The patient, with a history of thymoma resection, experienced multiple hospitalizations due to lung infections and neutropenia. The alveolar lavage fluid was detected by macro-genomic sequencing (NGS) to detect multiple pathogens, and targeted anti-infective and immunity-enhancing treatments led to improved symptoms and normal neutrophil counts. A literature review of 98 case reports from 2000 to 2023 was conducted, summarizing the associated diseases and pathogens in GS patients. Regular immunoglobulin monitoring in thymoma patients is essential for early GS diagnosis. When empirical antimicrobial therapy fails, mNGS for pathogen detection and targeted therapy are crucial, and regular IVIG injections can reduce infection rates in GS patients.

A Study of Improved Two-Stage Dual-Conv Coordinate Attention Model for Sound Event Detection and Localization.

Sound Event Detection and Localization (SELD) is a comprehensive task that aims to solve the subtasks of Sound Event Detection (SED) and Sound Source Localization (SSL) simultaneously. The task of SELD lies in the need to solve both sound recognition and spatial localization problems, and different categories of sound events may overlap in time and space, making it more difficult for the model to distinguish between different events occurring at the same time and to locate the sound source. In this study, the Dual-conv Coordinate Attention Module (DCAM) combines dual convolutional blocks and Coordinate Attention, and based on this, the network architecture based on the two-stage strategy is improved to form the SELD-oriented Two-Stage Dual-conv Coordinate Attention Model (TDCAM) for SELD. TDCAM draws on the concepts of Visual Geometry Group (VGG) networks and Coordinate Attention to effectively capture critical local information by focusing on the coordinate space information of the feature map and dealing with the relationship between the feature map channels to enhance the feature selection capability of the model. To address the limitation of a single-layer Bi-directional Gated Recurrent Unit (Bi-GRU) in the two-stage network in terms of timing processing, we add to the structure of the two-layer Bi-GRU and introduce the data enhancement techniques of the frequency mask and time mask to improve the modeling and generalization ability of the model for timing features. Through experimental validation on the TAU Spatial Sound Events 2019 development dataset, our approach significantly improves the performance of SELD compared to the two-stage network baseline model. Furthermore, the effectiveness of DCAM and the two-layer Bi-GRU structure is confirmed by performing ablation experiments.

Identifying MSMO1, ELOVL6, AACS, and CERS2 related to lipid metabolism as biomarkers of Parkinson's disease.

The mechanisms underlying lipid metabolic disorders in Parkinson's diseases (PD) remain unclear. Weighted Gene Co-Expression Network Analysis (WGCNA) was conducted to identify PD-related modular genes and differentially expressed genes (DEGs). Lipid metabolism-related genes (LMRGs) were extracted from Molecular Signatures Database. Candidate genes were assessed with overlapping modular genes, DEGs, and LMRGs for the purpose of building protein-protein interaction (PPI) networks. Then, biomarkers were generated by machine learning and Backpropagation Neural Network development according to candidate genes. Biomarker-based enrichment and network modulation analyses were executed to investigate related signaling pathways. Following dimensionality reduction clustering and annotation, scRNA-seq was submitted to cellular interactions and trajectory analysis to analyze regulatory mechanisms of critical cells. Finally, qRT-PCR was conducted to confirm the expression of biomarkers in PD patients. Four biomarkers (MSMO1, ELOVL6, AACS, and CERS2) were obtained and highly predictive after analysis mentioned above. Then, OPC, Oli, and Neu cells were the primary expression sites for biomarkers according to scRNA-seq studies. Finally, we confirmed mRNA of MSMO1, ELOVL6 and AACS were downregulated in PD patients comparing with control, while CERS2 was upregulated. In conclusion, MSMO1, ELOVL6, AACS, and CERS2 related to LMRGs could be new biomarkers for diagnosing and treating PD.

Identification of CREB5 as a prognostic and immunotherapeutic biomarker in glioma through multi-omics pan-cancer analysis.

BACKGROUND: The functional relevance of cyclic adenosine monophosphate (cAMP)-response element-binding protein 5 (CREB5) in cancers remains elusive, despite its significance as a member of the CREB family. The current research aims to explore the role of CREB5 in multiple cancers. METHODS: Pan-cancer analysis was performed to explore the expression patterns, prognostic value, mutational landscape as well as single-cell omic, immunologic, and drug sensitivity profiles of CREB5. Furthermore, we incorporated five distinct machine learning algorithms and determined that the least absolute shrinkage and selection operator-COX (LASSO-COX) algorithm, which exhibited the highest C index, was the optimal selection. Subsequently, we constructed a prognostic model centered around CREB5-associated genes. To elucidate the biological function of CREB5 in glioma cells, several assays including cell counting kit-8 (CCK-8), wound healing, transwell, flow cytometric were performed. RESULTS: CREB5 was overexpressed in pan-cancer and was linked to unfavorable prognosis, particularly in glioma. Furthermore, genetic alterations were determined in various types of cancer, and modifications in the CREB5 gene were linked to the prognosis. The single-cell omics and enrichment analyses showed that CREB5 was predominantly expressed in malignant glioma cells and was critically involved in the regulation of various oncogenic processes. Elevated levels of CREB5 were strongly linked with the infiltration of cancer-associated fibroblasts and the Th1 subset of CD4+ T cells. The validated CREB5-associated prognostic model reliably predicted the prognosis and drug response of glioma patients. The in vitro experiments showed that CREB5 promoted glioma cell proliferation, invasion, migration, and gap phase 2/mitotic (G2/M) phase arrest and recruited M2 macrophages into glioma cells. CONCLUSION: CREB5 has the potential to act as an oncogene and a biological marker in multiple cancers, particularly glioma.

Development of plasma technology for the preparation and modification of energy storage materials.

The development of energy storage material technologies stands as a decisive measure in optimizing the structure of clean and low-carbon energy systems. The remarkable activity inherent in plasma technology imbues it with distinct advantages in surface modification, functionalization, synthesis, and interface engineering of materials. This review systematically expounds upon the principles, classifications, and application scenarios of plasma technology, while thoroughly discussing its unique merits in the realm of modifying electrode materials, solid-state electrolytes, and conductive carbon materials, which are widely used in lithium-ion batteries, sodium ion batteries, metal air batteries and other fields. Finally, considering the existing constraints associated with lithium-ion batteries, some application prospects of plasma technology in the energy storage field are suggested. This work is of great significance for the development of clean plasma technology in the field of energy storage.

Effect of Fat Content on Rice Taste Quality through Transcriptome Analysis.

Rice is an important crop in the word, and fat is one of the main important nutrient components of rice. The lipid content and fatty acid composition of grains significantly influences the quality of rice. In this study, 94 homozygous recombination inbred lines (RILs) were developed and the crude fat content of them displayed a normal distribution ranging from 0.44% to 2.62%. Based on their taste quality, a positive association between fat content and eating quality was revealed. Then, two lines (FH and FL) were selected with similar agronomic characteristics and different lipid content and taste quality for RNA sequencing analysis, and a total of 619 differentiable expressed genes were detected, primarily enriched in metabolic pathways such as starch and sucrose metabolism, fatty acid metabolism, and amino acid metabolism. The expression of two genes related to fatty acid synthesis and elongation was significantly up-regulated, while the expression of three genes related to fatty acid degradation was significantly down-regulated in FH grains. By using liquid chromatography, the relative levels of palmitic acid and oleic acid were discovered significantly higher in FH grains. Additionally, the comparative genomic analysis was conducted to visualize genomic differences of five genes. Ultimately, two genes (Os07g0417200 and Os12g0102100) were selected to be the key gene to affect the lipid metabolism, especially for the synthesis of unsaturated fatty acids, significantly changing the eating quality of rice. These results provide a theoretical basis for improving the taste quality of rice.

Lack of causal association between epilepsy and dementia: A Mendelian randomization analysis.

OBJECTIVE: Epidemiological studies have reported an association between epilepsy and dementia. However, the causal relationship between epilepsy and the risk of dementia is not clear. We aimed to inspect the causal effect of epilepsy on memory loss and dementia. METHODS: We analyzed summary data of epilepsy, memory loss, and dementia from the genome-wide association study (GWAS) using the two-sample Mendelian randomization (MR) method. We used the estimated odds ratio of memory loss and dementia associated with each of the genetically defined traits to infer evidence for a causal relationship with the following exposures: all epilepsy, focal epilepsy (including focal epilepsy with hippocampal sclerosis, lesion-negative focal epilepsy, and focal epilepsy with other lesions), and genetic generalized epilepsy (including childhood absence epilepsy, generalized tonic-clonic seizures alone, Juvenile absence epilepsy, and Juvenile myoclonic epilepsy). RESULTS: According to the result of MR using the inverse variance weighted method (IVW), we found that genetically predicted epilepsy did not causally increase the risk of memory loss and dementia (p > 0.05). Results of the MR-Egger and weighted median method were consistent with the IVW method. CONCLUSIONS: No evidence has been found to support the notion that epilepsy can result in memory loss and dementia. The associations observed in epidemiological studies could be attributed, in part, to confounding or nongenetic determinants.

Case Report: Leiomyosarcoma of the right external iliac artery: a diagnostic-based study on a rare case.

Leiomyosarcoma (LMS) is an uncommon and aggressive form of cancer that originates in the smooth muscles. It possesses the capacity for rapid growth and often manifests with general, nonspecific symptoms arising from the displacement of nearby structures rather than direct invasion. In this particular instance, an 81-year-old woman presented with right lower abdominal pain, leading to the discovery of a mass adjacent to the right external iliac artery. In this case, the patient was misdiagnosed initially because of her nonspecific and poorly distinguished clinical symptoms. The laboratory results were within normal ranges. A well-defined tumor was detected through laparoscopic operation and subsequently surgically excised. The histopathological analysis of the tumor revealed the presence of malignant spindle cells, nuclear pleomorphism, and tumor giant cells. Immunohistochemistry tests indicated positive results for CD34 and Desmin, while CD117 and DOG1 showed adverse effects. It is worth noting that LMS of the right external iliac artery is an infrequent occurrence, potentially resulting in delayed diagnosis and misidentification. To enhance our comprehension of this uncommon cancer, more cases with detailed information are essential.

The causal relationship between circulating biomarkersand the risk of bipolar disorder: A two-sample Mendelian randomization study.

OBJECTIVE: To identify susceptible biomarkers for the development of bipolar disorder (BD), we conducted a Mendelian Randomization (MR) design to screen circulating proteins for the potential risk of bipolar disorder systematically. METHODS: We performed a two-sample Mendelian randomization (MR) analysis to estimate the causality of 4782 human circulating proteins on the risk of bipolar disorder. 376 circulating biomarkers were selected in MR estimation (4406 circulating proteins with less than 3 SNPs were excluded) with 5368 European descents. GWAS meta-analysis of the potential role of all-cause bipolar disorder arose from the Psychiatric Genomics Consortium (41,917 cases, 371,549 controls). RESULTS: After IVW and sensitivity analysis, 4 circulating proteins having causal effects on bipolar disorder were identified. ISG15, as a key player in the innate immune response, decreased the risk of bipolar disorder causally (OR = 0.92, 95% CI = 0.89-0.94, P = 1.46e-09). Furthermore, MLN decreased the risk of bipolar disorder causally (OR = 0.94, 95% CI = 0.91-0.97, P = 1.04e-04). In addition, SFTPC (OR = 0.91, 95% CI = 0.86-0.96, P = 4.47e-04) and VCY (OR = 0.86, 95% CI = 0.77-0.96, P = 8.55e-03) presented a suggestive association with bipolar disorder. CONCLUSIONS: Our findings indicated that ISG15 and MLN showed evidence of causality in bipolar disorder and provided a promising target for the diagnosis and treatment of diseases.

Influence of cryopreservation methods of ex vivo rat and pig skin on the results of in vitro permeation test.

In vitro permeation test (IVPT) is a frequently used method for in vitro assessment of topical preparations and transdermal drug delivery systems. However, the storage of ex vivo skin for IVPT remains a challenge. Here, two cryopreservation media were chosen to preserve rat and pig skin at -20 °C and -80 °C for further IVPT, namely, 10 % DMSO and 10 % GLY. The skin viability test confirmed that the skin protective capacity of 10 % DMSO and 10 % GLY was almost equal. The results of skin viability and IVPT showed that the skin viability and permeability of rat skin in 10 %DMSO or 10 % GLY were maintained for at least 7 and 30 days at -20 °C and -80 °C compared to fresh skin, respectively; in contrast, those of porcine skin were just maintained for <7 days at -20 °C and -80 °C. These results indicated that ex vivo skin for IVPT preserved at -80 °C in 10 % DMSO or 10 % GLY was optimal. Furthermore, skin permeability was independent of skin barrier integrity. Our study provides reference conditions for preserving IVPT skin, and skin viability can be a potential indicator of IVPT skin.

LDH@Boronate Polymer Core-Shell Nanoparticles: Nanostructure Design for Synergistically Enhancing the Flame Retardancy of Epoxy Resin.

As a promising nanofiller, layered double hydroxides (LDHs) can advance the fire safety of epoxy resin (EP), but so far, due to the problems of dispersion and low efficiency, it has still been a challenge to incorporate the flame retardancy and mechanical properties of EP nanocomposites effectively under the circumstance of a low additive amount. In this work, we take LDHs as the template, via the adsorption of a catechol group and the condensation polymerization between catechol groups and phenylboric acid groups, to prepare a core-shell structured nanoparticle LDH@BP, which contains rich flame-retardant elements. EP/LDH@BP nanocomposites were prepared by introducing LDH@BP into EP. The experimental results indicate that, compared with the original LDH, LDH@BP disperses uniformly in the EP matrix, and the flame retardancy and mechanical properties of EP/LDH@BP are significantly improved. At a relatively low content (5 wt%), EP/LDH@BP reached the rating of V-0 in the UL-94 test, LOI was increased to 29.1%, and peak heat release rate (PHRR) was reduced by 35.9% in cone calorimeter tests, which effectively inhibited the release of heat and toxic smoke during the combustion process of EP. Simultaneously, the mechanical properties of EP/LDH@BP have been improved satisfactorily. The above results derive from the reasonable architectural design of organic-inorganic nano-hybrid flame retardants and provide a novel method for the construction of efficient and balanced EP nanocomposite system with LDHs.

Decreased SPTBN2 expression regulated by the ceRNA network is associated with poor prognosis and immune infiltration in low­grade glioma.

The majority of low-grade gliomas (LGGs) in adults invariably progress to glioblastoma over time. Spectrin ß non-erythrocytic 2 (SPTBN2) is detected in numerous tumors and is involved in tumor occurrence and metastasis. However, the specific roles and detailed mechanisms of SPTBN2 in LGG are largely unknown. The present study performed pan-cancer analysis for the expression and prognosis of SPTBN2 in LGG using The Cancer Genome Atlas and The Genotype-Tissue Expression. Western blotting was used to detect the amount of SPTBN2 between glioma tissues and normal brain tissues. Subsequently, based on expression, prognosis, correlation and immune infiltration, non-coding RNAs (ncRNAs) were identified that regulated SPTBN2 expression. Finally, tumor immune infiltrates associated with SPTBN2 and prognosis were performed. Lower expression of SPTBN2 was correlated with an unfavorable outcome in LGG. A significant correlation between the low SPTBN2 mRNA expression and poor clinicopathological features was observed, including wild-type isocitrate dehydrogenase status (P<0.001), 1p/19q non-codeletion (P<0.001) and elders (P=0.019). The western blotting results revealed that, compared with normal brain tissues, the amount of SPTBN2 was significantly lower in LGG tissues (P=0.0266). Higher expression of five microRNAs (miRs/miRNAs), including hsa-miR-15a-5p, hsa-miR-15b-5p, hsa-miR-16-5p, hsa-miR-34c-5p and hsa-miR-424-5p, correlated with poor prognosis by targeting SPTBN2 in LGG. Subsequently, four long ncRNAs (lncRNAs) [ARMCX5-GPRASP2, BASP1-antisense RNA 1 (AS1), EPB41L4A-AS1 and LINC00641] were observed in the regulation of SPTBN2 via five miRNAs. Moreover, the expression of SPTBN2 was significantly correlated with tumor immune infiltration, immune checkpoint expression and biomarkers of immune cells. In conclusion, SPTBN2 was lowly expressed and correlated with an unfavorable prognosis in LGG. A total of six miRNAs and four lncRNAs were identified as being able to modulate SPTBN2 in a lncRNA-miRNA-mRNA network of LGG. Furthermore, the current findings also indicated that SPTBN2 possessed anti-tumor roles by regulating tumor immune infiltration and immune checkpoint expression.

A Novel P/N/Si-Containing Vanillin-Based Compound for a Flame-Retardant, Tough Yet Strong Epoxy Thermoset.

It is still extremely challenging to endow epoxy resins (EPs) with excellent flame retardancy and high toughness. In this work, we propose a facile strategy of combining rigid-flexible groups, promoting groups and polar phosphorus groups with the vanillin compound, which implements a dual functional modification for EPs. With only 0.22% phosphorus loading, the modified EPs obtain a limiting oxygen index (LOI) value of 31.5% and reach V-0 grade in UL-94 vertical burning tests. Particularly, the introduction of P/N/Si-containing vanillin-based flame retardant (DPBSi) improves the mechanical properties of EPs, including toughness and strength. Compared with EPs, the storage modulus and impact strength of EP composites can increase by 61.1% and 240%, respectively. Therefore, this work introduces a novel molecular design strategy for constructing an epoxy system with high-efficiency fire safety and excellent mechanical properties, giving it immense potential for broadening the application fields of EPs.

A Novel Anderson-Type POMs-Based Hybrids Flame Retardant for Reducing Smoke Release and Toxicity of Epoxy Resins.

Smoke emission and smoke toxicity have drawn more attention to improving the fire safety of polymers. In this work, a polyoxometalates (POMs)-based hybrids flame retardant (P-AlMo6 ) epoxy resin (EP) is prepared with toxicity-reduction and smoke-suppression properties via a peptide coupling reaction between POMs and organic molecules with double DOPO (bisDOPA). It combines the good compatibility of the organic molecule and the superior catalytic performance of POMs. Compared to pure EP, the glass transition temperature and flexural modulus of EP composite with 5 wt.% P-AlMo6 (EP/P-AlMo6 -5) are raised by 12.3 °C and 57.75%, respectively. Notably, at low flame-retardant addition, the average CO to CO2 ratio (Av-COY/Av-CO2 Y) is reduced by 33.75%. Total heat release (THR) and total smoke production (TSP) are lowered by 44.4% and 53.7%, respectively. The Limited Oxygen Index (LOI) value achieved 31.7% and obtained UL-94 V-0 rating. SEM, Raman, X-ray photoelectron spectroscopy, and TG-FTIR are applied to analyze the flame-retardant mechanism in condensed and gas phase. Outstanding flame retardant, low smoke toxicity properties are attained due to the catalytic carbonization ability of metal oxides Al2 O3 and MoO3 produced from the breakdown of POMs. This work advances the development of POMs-based hybrids flame retardants with low smoke toxicity properties.

Arrhythmia Disease Diagnosis Based on ECG Time-Frequency Domain Fusion and Convolutional Neural Network.

Electrocardiogram (ECG) signals are often used to diagnose cardiac status. However, most of the existing ECG diagnostic methods only use the time-domain information, resulting in some obviously lesion information in frequency-domain of ECG signals are not being fully utilized. Therefore, we propose a method to fuse the time and frequency domain information in ECG signals by convolutional neural network (CNN). First, we adapt multi-scale wavelet decomposition to filter the ECG signal; Then, R-wave localization is used to segment each individual heartbeat cycle; And then, the frequency domain information of this heartbeat cycle is extracted via fast Fourier transform. Finally, the temporal information is spliced with the frequency domain information and input to the neural network for classification. The experimental results show that the proposed method has the highest recognition accuracy (99.43%) of ECG singles compared with state-of-the-art methods. Clinical and Translational Impact Statement- The proposed ECG classification method provides an effective solution for ECG interrogation to quickly diagnose the presence of arrhythmia in a patient from the ECG signal. It can increase the efficiency of the interrogating physician by aiding diagnosis.

Deubiquitinase OTUD6A in macrophages promotes intestinal inflammation and colitis via deubiquitination of NLRP3.

Inflammatory bowel disease (IBD) is a chronic inflammatory disorder of the gastrointestinal tract, which has been shown to increase the incidence of colorectal cancer. Recent studies have highlighted the role of ubiquitination, a post-translational modification, in the occurrence and development of colonic inflammation. Ovarian tumor deubiquitinase 6 A (OTUD6A) is a deubiquitinating enzyme, which regulates cell proliferation and tumorigenesis. In this study, we investigated the expression and role of OTUD6A in IBD. Wide-type or Otud6a-/- mice were used to develop dextran sodium sulfate (DSS)- or 2,6,4-trinitrobenzene sulfonic acid (TNBS)-induced colitis model, as well as azoxymethane (AOM)/DSS-induced colitis-associated cancer model. Bone marrow-derived macrophages (BMDMs) were isolated from wild-type and Otud6a-/- mice to dissect molecular mechanisms. Our data show that OTUD6A deficiency attenuated DSS or TNBS-induced colitis, as well as AOM/DSS-induced colitis-related colon cancer in vivo. Bone marrow transplantation experiments further revealed that OTUD6A in myeloid cells was responsible for exacerbation of DSS-induced colitis. Mechanistically, OTUD6A directly bound to NACHT domain of NLRP3 inflammasome and selectively cleaved K48-linked polyubiquitin chains from NLRP3 at K430 and K689 to enhance the stability of NLRP3, leading to increased IL-1ß level and inflammation. Taken together, our research identifies a new function of OTUD6A in the pathogenesis of colitis by promoting NLRP3 inflammasome activation, suggesting that OTUD6A could be a potential target for the treatment of IBD.

HWA-SegNet: Multi-channel skin lesion image segmentation network with hierarchical analysis and weight adjustment.

Convolutional neural networks (CNNs) show excellent performance in accurate medical image segmentation. However, the characteristics of sample with small size and insufficient feature expression, irregular shape of the segmented target and inaccurate judgment of edge texture have always been problems to be faced in the field of skin lesion image segmentation. Therefore, in order to solve these problems, discrete Fourier transform (DFT) is introduced to enrich the input data and a CNN architecture (HWA-SegNet) is proposed in this paper. Firstly, DFT is improved to analyze the features of the skin lesions image, and multi-channel data is extended for each image. Secondly, a hierarchical dilated analysis module is constructed to understand the semantic features under multi-channel. Finally, the pre-prediction results are fine-tuned using a weight adjustment structure with fully connected layers to obtain higher accuracy prediction results. Then, 520 skin lesion images are tested on the ISIC 2018 dataset. Extensive experimental results show that our HWA-SegNet improve the average segmentation Dice Similarity Coefficient from 88.30% to 91.88%, Sensitivity from 89.29% to 92.99%, and Jaccard similarity index from 81.15% to 85.90% compared with U-Net. Compared with the State-of-the-Art method, the Jaccard similarity index and Specificity are close, but the Dice Similarity Coefficient is higher. The experimental data show that the data augmentation strategy based on improved DFT and HWA-SegNet are effective for skin lesion image segmentation.