A Joint Classification Method for COVID-19 Lesions Based on Deep Learning and Radiomics.

Pneumonia caused by novel coronavirus is an acute respiratory infectious disease. Its rapid spread in a short period of time has brought great challenges for global public health. The use of deep learning and radiomics methods can effectively distinguish the subtypes of lung diseases, provide better clinical prognosis accuracy, and assist clinicians, enabling them to adjust the clinical management level in time. The main goal of this study is to verify the performance of deep learning and radiomics methods in the classification of COVID-19 lesions and reveal the image characteristics of COVID-19 lung disease. An MFPN neural network model was proposed to extract the depth features of lesions, and six machine-learning methods were used to compare the classification performance of deep features, key radiomics features and combined features for COVID-19 lung lesions. The results show that in the COVID-19 image classification task, the classification method combining radiomics and deep features can achieve good classification results and has certain clinical application value.

Interface engineering based NiCoMoO4/Ti3C2Tx MXene heterostructure for high-performance flexible supercapacitors.

The advancement of interface engineering has demonstrated remarkable efficacy in overcoming the primary impediment associated with sluggish reaction kinetics in supercapacitor electrodes. In this investigation, we employed a facile co-precipitation method to synthesize NiCoMoO4/MXene heterostructures utilizing Ti3C2Tx MXene nanosheets as carriers. This heterostructure inhibits the restacking of MXene nanosheets and simultaneously enhances the exposure of electrochemically active sites in NiCoMoO4 nanorods, thereby mitigating the reduction in specific capacitance resulting from volumetric fluctuations. The NiCoMoO4/MXene electrode, possessing pseudo-capacitance properties, demonstrates an impressive level of specific capacitance, exceptional performance across various charging rates, and consistent behavior throughout repeated cycles. By optimizing the mass ratio, this electrode achieves a specific capacity of 1900 F/g under a current density of 1 A/g. Even after enduring 10,000 cycles at a significantly higher current density of 5 A/g, it still maintains an impressive retention rate of 94.73 %. Our density functional theory (DFT) calculations indicate that the enhanced electrochemical performance can be attributed to the improved electronic coupling within the NiCoMoO4/MXene heterostructure. The integration of NiCoMoO4/MXene cathode and activated carbon (AC) anode with an alkaline gel electrolyte containing potassium ferricyanide in flexible quasi-solid-state supercapacitors (FSSCs) results in exceptional electrochemical performance and flexibility. These FSSCs demonstrate a maximum energy density of 72.89 Wh kg-1 at a power density of 850 W kg-1, while maintaining an impressive power output of 16,780 W kg-1 with an energy density of 37.28 Wh kg-1. Based on these outstanding properties, it is evident that the NiCoMoO4/MXene heterojunction possesses significant advantages as electrode material for supercapacitors, and the fabricated FSSCs devices pave a new pathway for flexible electronic devices.

Prediction Model of hospitalization time of COVID-19 patients based on Gradient Boosted Regression Trees.

When an outbreak of COVID-19 occurs, it will cause a shortage of medical resources and the surge of demand for hospital beds. Predicting the length of stay (LOS) of COVID-19 patients is helpful to the overall coordination of hospital management and improves the utilization rate of medical resources. The purpose of this paper is to predict LOS for patients with COVID-19, so as to provide hospital management with auxiliary decision-making of medical resource scheduling. We collected the data of 166 COVID-19 patients in a hospital in Xinjiang from July 19, 2020, to August 26, 2020, and carried out a retrospective study. The results showed that the median LOS was 17.0 days, and the average of LOS was 18.06 days. Demographic data and clinical indicators were included as predictive variables to construct a model for predicting the LOS using gradient boosted regression trees (GBRT). The MSE, MAE and MAPE of the model are 23.84, 4.12 and 0.76 respectively. The importance of all the variables involved in the prediction of the model was analyzed, and the clinical indexes creatine kinase-MB (CK-MB), C-reactive protein (CRP), creatine kinase (CK), white blood cell count (WBC) and the age of patients had a higher contribution to the LOS. We found our GBRT model can accurately predict the LOS of COVID-19 patients, which will provide good assistant decision-making for medical management.

Yersinia enterocolitica-Derived Outer Membrane Vesicles Inhibit Initial Stage of Biofilm Formation.

Yersinia enterocolitica (Y. enterocolitica) is an important food-borne and zoonotic pathogen. It can form biofilm on the surface of food, increasing the risk to food safety. Generally, outer membrane vesicles (OMVs) are spherical nanostructures secreted by Gram-negative bacteria during growth. They play a role in biological processes because they contain biologically active molecules. Several studies have reported that OMVs secreted by various bacteria are associated with the formation of biofilms. However, the interactions between Y. enterocolitica OMVs and biofilm are unknown. This study aims to investigate the effect of Y. enterocolitica OMVs on biofilm formation. Firstly, OMVs were extracted from Y. enterocolitica Y1083, which has a strong biofilm-forming ability, at 15 °C, 28 °C and 37 °C and then characterized. The characterization results showed differences in the yield and protein content of three types of OMVs. Next, by co-culturing the OMVs with Y. enterocolitica, it was observed that the OMVs inhibited the initial stage of Y. enterocolitica biofilm formation but did not affect the growth of Y. enterocolitica. Furthermore, biofilm formation by Salmonella enteritidis and Staphylococcus aureus were also inhibited by OMVs. Subsequently, it was proved that lipopolysaccharides (LPS) in OMVs inhibited biofilm formation., The proteins, DNA or RNA in OMVs could not inhibit biofilm formation. Bacterial motility and the expression of the biofilm-related genes pgaABC, motB and flhBD were inhibited by LPS. LPS demonstrated good anti-biofilm activity against various bacteria. This study provides a new approach to the prevention and control of pathogenic bacterial biofilm.

Dicaffeoyltartaric acid analogues inhibit human immunodeficiency virus type 1 (HIV-1) integrase and HIV-1 replication at nontoxic concentrations.

The human immunodeficiency virus type 1 (HIV-1) is a major health problem worldwide. In this study, 17 analogues of L-chicoric acid, a potent inhibitor of HIV integrase, were studied. Of these analogues, five submicromolar inhibitors of integrase were discovered and 13 compounds with activity against integrase at less than 10 microM were identified. Six demonstrated greater than 10-fold selectivity for HIV replication over cellular toxicity. Ten analogues inhibited HIV replication at nontoxic concentrations. Alteration of the linkages between the two bis-catechol rings, including the use of amides, mixed amide esters, cholate, and alkyl bridges, was explored. Amides were as active as esters but were more toxic in tissue culture. Alkyl and cholate bridges were significantly less potent against HIV-1 integrase in vitro and were inactive against HIV-1 replication. Two amino acid derivates and one digalloylderivative of L-chicoric acid (L-CA) showed improved selectivity over L-CA against integration in cell culture. These data suggest that in addition to the bis-catechols and free carboxylic acid groups reported previously, polar linkages are important constituents for optimal activity against HIV-1 integrase and that new derivatives can be developed with increased specificity for integration over HIV entry in vivo.