A hybrid model for tuberculosis forecasting based on empirical mode decomposition in China.

BACKGROUND: Pulmonary Tuberculosis is a major public health problem endangering people's health, a scientifically accurate predictive model is of great practical significance for the prevention and treatment of pulmonary tuberculosis. METHODS: The reported incidence data of pulmonary tuberculosis were from the National Public Health Science Data Center ( https://www.phsciencedata.cn/ ). The ARIMA, LSTM, EMD-SARIMA, EMD-LSTM, EMD-ARMA-LSTM models were established using the reported monthly incidence of tuberculosis reported in China from January 2008 to December 2018. The MSE, MAE, RMSE and MAPE were used to evaluate the performance of the models to determine the best model. RESULTS: Comparing decomposition-based single model with undecomposed single model, it was found that: when predicting the incidence trend in the next year, compared with SARIMA model, the MSE, MAE, RMSE and MAPE of EMD-SARIMA decreased by 39.3%, 19.0%, 22.1% and 19.8%, respectively. The MSE, MAE, RMSE and MAPE of EMD-LSTM were reduced by 40.5%, 12.8%, 22.9% and 12.7%, respectively, compared with the LSTM model; Comparing the decomposition-based hybrid model with the decomposition-based single model, it was found that: when predicting the incidence trend in the next year, compared with EMD-SARIMA model, the MSE, MAE, RMSE and MAPE of EMD-ARMA-LSTM model decreased by 21.7%, 10.6%, 11.5% and 11.2%, respectively. The MSE, MAE, RMSE and MAPE of EMD-ARMA-LSTM were reduced by 16.7%, 9.6%, 8.7% and 12.3%, respectively, compared with EMD-LSTM model. Furthermore, the performance of the model were consistent when predicting the incidence trend in the next 3 months, 6 months and 9 months. CONCLUSION: The prediction performance of the decomposition-based single model is better than that of the undecomposed single model, and the prediction performance of the combined model using the advantages of different models is better than that of the decomposition-based single model, so the EMD-ARMA-LSTM combination model can improve the prediction accuracy better than other models, which can provide a theoretical basis for predicting the epidemic trend of pulmonary tuberculosis and formulating prevention and control policies.

Construction and application of sponge city resilience evaluation system: a case study in Xi'an, China.

Urban vulnerability is evident when highly complex flood risks overlap with diverse cities, and it is important to enhance the resilience of cities to flood shocks. In this study, a sponge city resilience assessment system is established considering engineering, environmental and social indicators, and the grey relational analysis method (GRA) is used to quantify sponge city resilience. At the same time, a multi-objective optimization model is established based on the three dimensions of water ecological environment, drainage safety, and waterlogging safety. The optimal configuration of grey-green infrastructure is weighed by combining the ideal point method, aiming to ensure that cities effectively reduce flood risk through the optimal configuration scheme. Taking the Xiaozhai area in Xi'an as the study area, the evaluation results show that the grey relational degree (GRD) of the resilience indexes of the original scheme is between 0.390 and 0.661 under the seven different return periods, while the optimization scheme ranges from 0.648 to 0.765, with the best sponge city resilience at a return period of 2a. Compared with the original scheme, the optimized sponge city resilience level increases from level II to nearly level I in the low return period and from level IV to level II in the high return period, indicating that city's ability to cope with waterlogging and pollution is enhanced significantly. Besides, the main factor affecting the sponge city resilience is the runoff control rate, followed by pollutant load reduction rate, which can provide a methodological framework for the assessment and improvement of sponge city resilience.

Study on the prediction effect of a combined model of SARIMA and LSTM based on SSA for influenza in Shanxi Province, China.

BACKGROUND: Influenza is an acute respiratory infectious disease that is highly infectious and seriously damages human health. Reasonable prediction is of great significance to control the epidemic of influenza. METHODS: Our Influenza data were extracted from Shanxi Provincial Center for Disease Control and Prevention. Seasonal-trend decomposition using Loess (STL) was adopted to analyze the season characteristics of the influenza in Shanxi Province, China, from the 1st week in 2010 to the 52nd week in 2019. To handle the insufficient prediction performance of the seasonal autoregressive integrated moving average (SARIMA) model in predicting the nonlinear parts and the poor accuracy of directly predicting the original sequence, this study established the SARIMA model, the combination model of SARIMA and Long-Short Term Memory neural network (SARIMA-LSTM) and the combination model of SARIMA-LSTM based on Singular spectrum analysis (SSA-SARIMA-LSTM) to make predictions and identify the best model. Additionally, the Mean Squared Error (MSE), Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) were used to evaluate the performance of the models. RESULTS: The influenza time series in Shanxi Province from the 1st week in 2010 to the 52nd week in 2019 showed a year-by-year decrease with obvious seasonal characteristics. The peak period of the disease mainly concentrated from the end of the year to the beginning of the next year. The best fitting and prediction performance was the SSA-SARIMA-LSTM model. Compared with the SARIMA model, the MSE, MAE and RMSE of the SSA-SARIMA-LSTM model decreased by 38.12, 17.39 and 21.34%, respectively, in fitting performance; the MSE, MAE and RMSE decreased by 42.41, 18.69 and 24.11%, respectively, in prediction performances. Furthermore, compared with the SARIMA-LSTM model, the MSE, MAE and RMSE of the SSA-SARIMA-LSTM model decreased by 28.26, 14.61 and 15.30%, respectively, in fitting performance; the MSE, MAE and RMSE decreased by 36.99, 7.22 and 20.62%, respectively, in prediction performances. CONCLUSIONS: The fitting and prediction performances of the SSA-SARIMA-LSTM model were better than those of the SARIMA and the SARIMA-LSTM models. Generally speaking, we can apply the SSA-SARIMA-LSTM model to the prediction of influenza, and offer a leg-up for public policy.

3D Bacterial flagella as both synthetic biotemplates and ultrathin spacers for enhanced inter-particle coupling and solar energy harvesting.

Linear light-absorbing nanomaterials are ideal for film-based solar harvesting applications as they form porous structures that can maximize the absorption and minimize the reflection of the solar light. Conventional 1D nanochains of plasmonic nanoparticle assemblies can achieve significantly broadened optical absorption through surface plasmon coupling, but their optical bands are still not broad enough to absorb through the solar spectrum and thus are not efficient solar absorbers. Here we discovered first by simulation that 3D structured nanochains of plasmonic nanoparticles presented a remarkably increased optical broadening effect and much longer redshift of the optical peaks due to the enhanced inter-particle coupling effect. Then we fabricated 3D nanochains by assembling gold nanoparticles (AuNPs) around 14 nm ultrathin bionanofibers, the bacterial flagella. The ultrathin biotemplates enabled the 3D arrangement of 50 nm AuNPs along the nanofiber with a very small inter-particle gap, allowing the strong coupling of surface plasmons in a 3D manner. Consistent with the theoretical prediction, the 3D nanochains, when assembled into films, could effectively convert nearly the full spectrum of solar energy into heat, which was further efficiently converted into electricity through a thermoelectric generation unit. Our work represents a nanobiomaterial approach to highly efficient solar thermal power generation.

Application of a novel hybrid algorithm of Bayesian network in the study of hyperlipidemia related factors: a cross-sectional study.

BACKGROUND: This article aims to understand the prevalence of hyperlipidemia and its related factors in Shanxi Province. On the basis of multivariate Logistic regression analysis to find out the influencing factors closely related to hyperlipidemia, the complex network connection between various variables was presented through Bayesian networks(BNs). METHODS: Logistic regression was used to screen for hyperlipidemia-related variables, and then the complex network connection between various variables was presented through BNs. Since some drawbacks stand out in the Max-Min Hill-Climbing (MMHC) hybrid algorithm, extra hybrid algorithms are proposed to construct the BN structure: MMPC-Tabu, Fast.iamb-Tabu and Inter.iamb-Tabu. To assess their performance, we made a comparison between these three hybrid algorithms with the widely used MMHC hybrid algorithm on randomly generated datasets. Afterwards, the optimized BN was determined to explore to study related factors for hyperlipidemia. We also make a comparison between the BN model with logistic regression model. RESULTS: The BN constructed by Inter.iamb-Tabu hybrid algorithm had the best fitting degree to the benchmark networks, and was used to construct the BN model of hyperlipidemia. Multivariate logistic regression analysis suggested that gender, smoking, central obesity, daily average salt intake, daily average oil intake, diabetes mellitus, hypertension and physical activity were associated with hyperlipidemia. BNs model of hyperlipidemia further showed that gender, BMI, and physical activity were directly related to the occurrence of hyperlipidemia, hyperlipidemia was directly related to the occurrence of diabetes mellitus and hypertension; the average daily salt intake, daily average oil consumption, smoking, and central obesity were indirectly related to hyperlipidemia. CONCLUSIONS: The BN of hyperlipidemia constructed by the Inter.iamb-Tabu hybrid algorithm is more reasonable, and allows for the overall linking effect between factors and diseases, revealing the direct and indirect factors associated with hyperlipidemia and correlation between related variables, which can provide a new approach to the study of chronic diseases and their associated factors.

Research on the predictive effect of a combined model of ARIMA and neural networks on human brucellosis in Shanxi Province, China: a time series predictive analysis.

BACKGROUND: Brucellosis is a major public health problem that seriously affects developing countries and could cause significant economic losses to the livestock industry and great harm to human health. Reasonable prediction of the incidence is of great significance in controlling brucellosis and taking preventive measures. METHODS: Our human brucellosis incidence data were extracted from Shanxi Provincial Center for Disease Control and Prevention. We used seasonal-trend decomposition using Loess (STL) and monthplot to analyse the seasonal characteristics of human brucellosis in Shanxi Province from 2007 to 2017. The autoregressive integrated moving average (ARIMA) model, a combined model of ARIMA and the back propagation neural network (ARIMA-BPNN), and a combined model of ARIMA and the Elman recurrent neural network (ARIMA-ERNN) were established separately to make predictions and identify the best model. Additionally, the mean squared error (MAE), mean absolute error (MSE) and mean absolute percentage error (MAPE) were used to evaluate the performance of the model. RESULTS: We observed that the time series of human brucellosis in Shanxi Province increased from 2007 to 2014 but decreased from 2015 to 2017. It had obvious seasonal characteristics, with the peak lasting from March to July every year. The best fitting and prediction effect was the ARIMA-ERNN model. Compared with those of the ARIMA model, the MAE, MSE and MAPE of the ARIMA-ERNN model decreased by 18.65, 31.48 and 64.35%, respectively, in fitting performance; in terms of prediction performance, the MAE, MSE and MAPE decreased by 60.19, 75.30 and 64.35%, respectively. Second, compared with those of ARIMA-BPNN, the MAE, MSE and MAPE of ARIMA-ERNN decreased by 9.60, 15.73 and 11.58%, respectively, in fitting performance; in terms of prediction performance, the MAE, MSE and MAPE decreased by 31.63, 45.79 and 29.59%, respectively. CONCLUSIONS: The time series of human brucellosis in Shanxi Province from 2007 to 2017 showed obvious seasonal characteristics. The fitting and prediction performances of the ARIMA-ERNN model were better than those of the ARIMA-BPNN and ARIMA models. This will provide some theoretical support for the prediction of infectious diseases and will be beneficial to public health decision making.

Exploratory study on classification of diabetes mellitus through a combined Random Forest Classifier.

BACKGROUND: Diabetes Mellitus (DM) has become the third chronic non-communicable disease that hits patients after tumors, cardiovascular and cerebrovascular diseases, and has become one of the major public health problems in the world. Therefore, it is of great importance to identify individuals at high risk for DM in order to establish prevention strategies for DM. METHODS: Aiming at the problem of high-dimensional feature space and high feature redundancy of medical data, as well as the problem of data imbalance often faced. This study explored different supervised classifiers, combined with SVM-SMOTE and two feature dimensionality reduction methods (Logistic stepwise regression and LAASO) to classify the diabetes survey sample data with unbalanced categories and complex related factors. Analysis and discussion of the classification results of 4 supervised classifiers based on 4 data processing methods. Five indicators including Accuracy, Precision, Recall, F1-Score and AUC are selected as the key indicators to evaluate the performance of the classification model. RESULTS: According to the result, Random Forest Classifier combining SVM-SMOTE resampling technology and LASSO feature screening method (Accuracy = 0.890, Precision = 0.869, Recall = 0.919, F1-Score = 0.893, AUC = 0.948) proved the best way to tell those at high risk of DM. Besides, the combined algorithm helps enhance the classification performance for prediction of high-risk people of DM. Also, age, region, heart rate, hypertension, hyperlipidemia and BMI are the top six most critical characteristic variables affecting diabetes. CONCLUSIONS: The Random Forest Classifier combining with SVM-SMOTE and LASSO feature reduction method perform best in identifying high-risk people of DM from individuals. And the combined method proposed in the study would be a good tool for early screening of DM.

Human Mesenchymal Stem Cell Derived Exosomes Enhance Cell-Free Bone Regeneration by Altering Their miRNAs Profiles.

Implantation of stem cells for tissue regeneration faces significant challenges such as immune rejection and teratoma formation. Cell-free tissue regeneration thus has a potential to avoid these problems. Stem cell derived exosomes do not cause immune rejection or generate malignant tumors. Here, exosomes that can induce osteogenic differentiation of human mesenchymal stem cells (hMSCs) are identified and used to decorate 3D-printed titanium alloy scaffolds to achieve cell-free bone regeneration. Specifically, the exosomes secreted by hMSCs osteogenically pre-differentiated for different times are used to induce the osteogenesis of hMSCs. It is discovered that pre-differentiation for 10 and 15 days leads to the production of osteogenic exosomes. The purified exosomes are then loaded into the scaffolds. It is found that the cell-free exosome-coated scaffolds regenerate bone tissue as efficiently as hMSC-seeded exosome-free scaffolds within 12 weeks. RNA-sequencing suggests that the osteogenic exosomes induce the osteogenic differentiation by using their cargos, including upregulated osteogenic miRNAs (Hsa-miR-146a-5p, Hsa-miR-503-5p, Hsa-miR-483-3p, and Hsa-miR-129-5p) or downregulated anti-osteogenic miRNAs (Hsa-miR-32-5p, Hsa-miR-133a-3p, and Hsa-miR-204-5p), to activate the PI3K/Akt and MAPK signaling pathways. Consequently, identification of osteogenic exosomes secreted by pre-differentiated stem cells and the use of them to replace stem cells represent a novel cell-free bone regeneration strategy.

Quantification of silk protein using phage nanofibers with high binding specificity.

Silk sericin (SS) has emerged as an important silk protein for use in medicine and textiles. However, no sensitive method is available for detecting it. Here, we employed phage nanofibers (∼7 nm wide) as a probe to quantify SS from a dilute aqueous solution by exploiting two properties of the bacteria-infecting phage nanofibers, its use as a platform for discovering SS-binding peptide and its ultrasensitive quantification by a simple titering assay (where the number of phage nanofibers displaying the SS-binding peptide is equal to the number of countable millimeter-sized plaques derived from the phage nanofibers by infecting bacteria through plating). We first discovered a SS-binding peptide and the phage nanofibers (SS-phage) displaying this peptide at the tip. We found that this peptide can even differentiate SS from another silk protein (silk fibroin), showing its high specificity. We then employed SS-phage nanofibers as a probe to bind the SS casted from the aqueous solution. Because SS-phage nanofibers bound to the SS and the SS in the original SS solution were numerically correlated and the number of SS-phage nanofibers can be determined by counting the plaques in a Petri dish by the titering assay, determining the number of phage-derived plaques with the naked eye led to the rapid quantification of SS concentration with a detection limit of 19.50 ng ml-1. This phage-based counting strategy can be potentially applied to the facile detection of other proteins.

Alpha-7 Nicotinic Receptor Signaling Pathway Participates in the Neurogenesis Induced by ChAT-Positive Neurons in the Subventricular Zone.

Choline acetyltransferase-positive (ChAT+) neurons within the subventricular zone (SVZ) have been shown to promote neurogenesis after stroke in mice by secreting acetylcholine (ACh); however, the mechanisms remain unclear. Receptors known to bind ACh include the nicotinic ACh receptors (nAChRs), which are present in the SVZ and have been shown to be important for cell proliferation, differentiation, and survival. In this study, we investigated the neurogenic role of the alpha-7 nAChR (α7 nAChR) in a mouse model of middle cerebral artery occlusion (MCAO) by using α7 nAChR inhibitor methyllycaconitine. Mice subjected to MCAO exhibited elevated expression of cytomembrane and nuclear fibroblast growth factor receptor 1 (FGFR1), as well as increased expression of PI3K, pAkt, doublecortin (DCX), polysialylated - neuronal cell adhesion molecule (PSA-NCAM), and mammalian achaete-scute homolog 1 (Mash1). MCAO mice also had more glial fibrillary acidic protein (GFAP)/5-bromo-2'-deoxyuridine (BrdU)-positive cells and DCX-positive cells in the SVZ than did the sham-operated group. Methyllycaconitine treatment increased cytomembrane FGFR1 expression and GFAP/BrdU-positive cells, upregulated the levels of phosphoinositide 3-kinase (PI3K) and phospho-Akt (pAkt), decreased nuclear FGFR1 expression, decreased the number of DCX-positive cells, and reduced the levels of DCX, PSA-NCAM, and Mash1 in the SVZ of MCAO mice compared with levels in vehicle-treated MCAO mice. MCAO mice treated with α7 nAChR agonist PNU-282987 exhibited the opposite effects. Our data show that α7 nAChR may decrease the proliferation of neural stem cells and promote differentiation of existing neural stem cells after stroke. These results identify a new mechanism of SVZ ChAT+ neuron-induced neurogenesis.