Explore the vast range of titles available.

Environmental sound classification is one of the important issues in the audio recognition field. Compared with structured sounds such as speech and music, the time–frequency structure of environmental sounds is more complicated. In order to learn time and frequency features from Log-Mel spectrogram more effectively, a temporal-frequency attention based convolutional neural network model (TFCNN) is proposed in this paper. Firstly, an experiment that is used as motivation in proposed method is designed to verify the effect of a specific frequency band in the spectrogram on model classification. Secondly, two new attention mechanisms, temporal attention mechanism and frequency attention mechanism, are proposed. These mechanisms can focus on key frequency bands and semantic related time frames on the spectrogram to reduce the influence of background noise and irrelevant frequency bands. Then, a feature information complementarity is formed by combining these mechanisms to more accurately capture the critical time–frequency features. In such a way, the representation ability of the network model can be greatly improved. Finally, experiments on two public data sets, UrbanSound 8 K and ESC-50, demonstrate the effectiveness of the proposed method.

Programmed death receptor-1 (PD-1) blockade have achieved some efficacy but only in a fraction of patients with hepatocellular carcinoma (HCC). Programmed cell death 1 ligand 1 (PD-L1) binds to its receptor PD1 on T cells to dampen antigen-tumor immune responses. However, the mechanisms underlying PD-L1 regulation are not fully elucidated. Herein, we identify that tumoral Prdm1 overexpression inhibits cell growth in immune-deficient mouse models. Further, tumoral Prdm1 overexpression upregulates PD-L1 levels, dampening anti-tumor immunity in vivo, and neutralizes the anti-tumor efficacy of Prdm1 overexpression in immune-competent mouse models. Mechanistically, PRDM1 enhances USP22 transcription, thus reducing SPI1 protein degradation through deubiquitination, which enhances PD-L1 transcription. Functionally, PD-1 mAb treatment reinforces the efficacy of Prdm1 -overexpressing HCC immune-competent mouse models. Collectively, we demonstrate that the PRDM1-USP22-SPI1 axis regulates PD-L1 levels, resulting in infiltrated CD8 + T cell exhaustion. Furthermore, PRDM1 overexpression combined with PD-(L)1 mAb treatment provides a therapeutic strategy for HCC treatment. Members of the PRDI-BF1 and RIZ homology domain (PRDM) family have been involved in the regulation of several pathological conditions, including cancer. Here the authors show that PRDM1/BLIMP1 promotes immune evasion by regulating PD-L1 expression in hepatocellular carcinoma cells.

The objectives are to improve the diagnostic efficiency and accuracy of epidemic pulmonary infectious diseases and to study the application of artificial intelligence (AI) in pulmonary infectious disease diagnosis and public health management. The computer tomography (CT) images of 200 patients with pulmonary infectious disease are collected and input into the AI-assisted diagnosis software based on the deep learning (DL) model, \"UAI, pulmonary infectious disease intelligent auxiliary analysis system\", for lesion detection. By analyzing the principles of convolutional neural networks (CNN) in deep learning (DL), the study selects the AlexNet model for the recognition and classification of pulmonary infection CT images. The software automatically detects the pneumonia lesions, marks them in batches, and calculates the lesion volume. The result shows that the CT manifestations of the patients are mainly involved in multiple lobes and density, the most common shadow is the ground-glass opacity. The detection rate of the manual method is 95.30%, the misdetection rate is 0.20% and missed diagnosis rate is 4.50%; the detection rate of the DL-based AI-assisted lesion method is 99.76%, the misdetection rate is 0.08%, and the missed diagnosis rate is 0.08%. Therefore, the proposed model can effectively identify pulmonary infectious disease lesions and provide relevant data information to objectively diagnose pulmonary infectious disease and manage public health.

The non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio (NHHR) is a new composite lipid index. Blood lipid levels are closely associated with the severity of acute pancreatitis (AP), but the relationship between NHHR and mortality in AP patients remains unclear. Therefore, the purpose of this study is to investigate the connection between NHHR and the mortality of patients with AP within 90 days. A total of 454 adult participants with AP from the MIMIC-IV database were categorized into three groups based on their NHHR levels. We employed a multivariate Cox proportional hazards model to assess the relationship between NHHR and 90-day mortality in AP patients. Additionally, we utilized the random forest method to identify the ten most significant risk factors associated with mortality in AP, which was subsequently used to build a prediction model. Out of the patients with AP, 27 died within 90 days. After adjusting for various factors, the hazard ratios for mortality across the tertiles of the NHHR (from the lowest to the highest tertiles: Q1–Q3) were as follows: 1.00 (reference), 1.47 (95% CI: 0.47–4.66), and 4.07 (95% CI: 1.43–11.60). Additionally, the area under the curve (AUC) for our AP mortality prediction model is 0.867 (95% CI: 0.769–0.964). Our findings indicate that a high level of NHHR is closely associated with the mortality of AP, serving as a simple index to predict the prognosis of AP patients.

A large amount of arsenic-containing wastewater discharged by the non-ferrous metal industry will cause serious environmental problems if it is not properly treated. Pyrolysis char of waste tire is a kind of solid waste. Since the surface properties of tire derived char (TC) are affected by tar/ash adhesion during pyrolysis, it is necessary to modify TC to treat wastewater containing As(V) effectively as an adsorbent. At present, most studies on the modification of TC are prepared into activated carbon by high temperature activation in N 2 atmosphere. In this study, TC was modified at room temperature and air atmosphere, and Fe(OH) 3 -TC NaOH adsorbent with particle size of 61–75 μm was obtained under the premise of the removal rate of As(V) and the settling performance of the adsorbent. When the initial concentration of As(V) was 5 mg/L, the removal rate of As(V) by Fe(OH) 3 -TC NaOH with a particle size of 61–75 μm could reach 90% within 30 min under a wide pH range (3–9). The adsorption of As(V) by Fe(OH) 3 -TC NaOH was most affected by the coexistence of PO 4 3- , which resulted in the removal rate of As(V) decreased by about 20%. The adsorption mechanism shows that the significant increase in the number of 3–5 nm mesoporous pores of Fe(OH) 3 -TC NaOH and the formation of H bonds are beneficial to the adsorption of Fe(OH) 3 -TC NaOH to As(V), and improve the stability of Fe-As complex.

The irregular distribution of non-zero elements of large-scale sparse matrix leads to low data access efficiency caused by the unique architecture of the Sunway many-core processor, which brings great challenges to the efficient implementation of sparse matrix–vector multiplication (SpMV) computing by SW26010P many-core processor. To address this problem, a study of SpMV optimization strategies is carried out based on the SW26010P many-core processor. Firstly, we design a memorized data storage transformation strategy to transform the matrix in CSR storage format into BCSR (Block Compressed Sparse Row) storage. Secondly, the dynamic task scheduling method is introduced to the algorithm to realize the load balance between slave cores. Thirdly, the LDM memory is refined and designed, and the slave core dual cache strategy is optimized to further improve the performance. Finally, we selected a large number of representative sparse matrices from the Matrix Market for testing. The results show that the scheme has obviously speedup the processing procedure of sparse matrices with various sizes and sizes, and the master–slave speedup ratio can reach up to 38 times. The optimization method used in this paper has implications for other complex applications of the SW26010P many-core processor.

With the development of industry and agriculture, eutrophication caused by increasing amounts of phosphorus in the environment has attracted people's attention. On the other hand, copper tailings (CT) is a kind of solid waste with large quantity, large area, and easy to cause groundwater and soil pollution. CT is also a potential resource because of its large specific surface area. CT is intended to be used as an adsorbent for removal phosphate in water, but trace heavy metals and a small amount of phosphate in CT may bring negative effects. Calcium hydroxide (Ca(OH) 2 ) was used to modify CT (CCT), hoping to fix the heavy metals and phosphate in CT at the same time. It was found that the removal capacity of CCT was significantly higher than that of CT. The process of phosphate removal by CCT involves electrostatic sorption and surface precipitation, and there is a synergistic effect between CT and Ca(OH) 2 . The phosphate removal rate of CCT-0.4 increased with the increase of pH value under alkaline conditions. The XRD patterns of phosphate sorption by CCT mean that Ca 3 (PO 4 ) 2 , Ca 5 (PO 4 ) 3 (OH) and AlPO 4 exist in CCT after phosphate removal, indicating that surface precipitation occurs during the removal process. In summary, the removal mechanism of phosphate by CCT is mainly electrostatic attraction and surface precipitation.

With the increasing number and types of global power loads and the development and popularization of smart grid technology, a large number of researches on load-level non-intrusive load monitoring technology have emerged. However, the unique power characteristics of the load make NILM face the difficult problem of low robustness of feature extraction and low accuracy of classification and identification in the recognition stage. This paper proposes a structured V-I mapping method to address the inherent limitations of traditional V-I trajectory mapping methods from a new perspective. In addition, for the verification of the V-I trajectory mapping method proposed in this paper, the complexity of load characteristics is comprehensively considered, and a lightweight convolutional neural network is designed based on AlexNet. The experimental results on the NILM dataset show that the proposed method significantly improves recognition accuracy compared to existing VI trajectory mapping methods.

Deep learning networks have been successfully applied to transfer functions so that the models can be adapted from the source domain to different target domains. This study uses multiple convolutional neural networks to decode the electroencephalogram (EEG) of stroke patients to design effective motor imagery (MI) brain-computer interface (BCI) system. This study has introduced ‘fine-tune’ to transfer model parameters and reduced training time. The performance of the proposed framework is evaluated by the abilities of the models for two-class MI recognition. The results show that the best framework is the combination of the EEGNet and ‘fine-tune’ transferred model. The average classification accuracy of the proposed model for 11 subjects is 66.36%, and the algorithm complexity is much lower than other models.These good performance indicate that the EEGNet model has great potential for MI stroke rehabilitation based on BCI system. It also successfully demonstrated the efficiency of transfer learning for improving the performance of EEG-based stroke rehabilitation for the BCI system.

Molecular mechanisms underlying breast cancer lymph node metastasis remain unclear. Using single-cell sequencing, we investigated the transcriptome profile of 96,796 single cells from 15 paired samples of primary tumors and axillary lymph nodes. We identified nine cancer cell subclusters including CD44 + / ALDH2 + /ALDH6A1 + breast cancer stem cells (BCSCs), which had a copy-number variants profile similar to that of normal breast tissue. Importantly, BCSCs existed only in primary tumors and evolved into metastatic clusters infiltrating into lymph nodes. Furthermore, transcriptome data suggested that NECTIN2-TIGIT-mediated interactions between metastatic breast cancer cells and tumor microenvironment (TME) cells, which promoted immune escape and lymph node metastasis. This study is the first to delineate the transcriptome profile of breast cancer lymph node metastasis using single-cell RNA sequencing. Our findings offer novel insights into the mechanisms underlying breast cancer metastasis and have implications in developing novel therapies to inhibit the initiation of breast cancer metastasis.