Predicting Hospital Readmission for Campylobacteriosis from Electronic Health Records: A Machine Learning and Text Mining Perspective

(1) Background: This study investigates influential risk factors for predicting 30-day readmission to hospital for Campylobacter infections (CI). (2) Methods: We linked general practitioner and hospital admission records of 13,006 patients with CI in Wales (1990–2015). An approach called TF-zR (term frequency-zRelevance) technique was presented to evaluates how relevant a clinical term is to a patient in a cohort characterized by coded health records. The zR is a supervised term-weighting metric to assign weight to a term based on relative frequencies of the term across different classes. Cost-sensitive classifier with swarm optimization and weighted subset learning was integrated to identify influential clinical signals as predictors and optimal model for readmission prediction. (3) Results: From a pool of up to 17,506 variables, 33 most predictive factors were identified, including age, gender, Townsend deprivation quintiles, comorbidities, medications, and procedures. The predictive model predicted readmission with 73% sensitivity and 54% specificity. Variables associated with readmission included male gender, recurrent tonsillitis, non-healing open wounds, operation for in-gown toenails. Cystitis, paracetamol/codeine use, age (21–25), and heliclear triple pack use, were associated with a lower risk of readmission. (4) Conclusions: This study gives a profile of clustered variables that are predictive of readmission associated with campylobacteriosis.

GENERALIZED AUTOREGRESSIVE CONDITIONAL HETEROSCEDASTICITY (GARCH) MODELS AND OPTIMAL FOR NIGERIAN STOCK EXCHANGE

This paper focused on comparative performance of GARCH models, ascertaining the best model fit, estimating the parameters and making prediction from optimal model. The study used UBA daily stock exchange prices sourced from the official websites of www.investing.com,on the daily basis of the Nigeria stock exchange rate over a period of ten years from 06/06/2012 – 04/06/2021. Five GARCH models (SGARCH, GJRGARCH or TGARCH, EGARCH, APGARCH and IGARCH) were fitted to the secondary data set of the Nigerian Stock exchange market for the period of June 2012- June 2021 and the results of the findings were obtained. The AIC results were SGARCH (1,1) (-6.1784), GJRGARCH (1,1) (-6.1778), EGARCH (1,1) (-6.1714) , APGARCH (1,1) (-6.1245) and IGARCH(1,1) with the value of AIC -6.1793. The EGARCH (1, 1) was found to be the optimal model with AIC value of -6.1714. The further findings indicated volatility clustering and leverage effect. The result of the analysis equally showed parameter estimates of the EGARCH (1,1) model and all the parameters were significant including mean and alpha. Prediction using the optimal model was made with an initial out of sample of 200 and n ahead of 200 with predicted values within the 95% confidence interval resulting there is no sign of volatility and clustering. Based on the findings of the study, other time series packages should be compared with GARCH models, data should be making available for easy access and investors should be encouraged to invest in United Bank for Africa (UBA, Nigeria).

Study on the application of BP neural network optimized based on various optimization algorithms in storm surge prediction

Storm surge prediction is of great importance to disaster prevention and mitigation. In this study, four optimization algorithms including genetic algorithm (GA), particle swarm optimization (PSO), beetle antenna search (BAS), and beetle swarm optimization (BSO) are used to optimize the back propagation neural network (BPNN), and four optimized BPNNs for storm surge prediction are proposed and applied to Yulin station and Xiuying station at Hainan, China. The optimal model parameter combination is determined by trail-and-error method for the best prediction performance. Comparisons with the single BPNN indicate that storm surge can be efficiently predicted using the optimized BPNNs. BPNN optimized by BSO has the minimum prediction error, and BPNN optimized by BAS has the minimum time cost to reduce unit prediction error.

Soil Erosion Assessment of Alpine Grassland in the Source Park of the Yellow River on the Qinghai-Tibetan Plateau, China

The source park of the Yellow River (SPYR), as a vital ecological shelter on the Qinghai-Tibetan Plateau, is suffering different degrees of degradation and desertification, resulting in soil erosion in recent decades. Therefore, studying the mechanism, influencing factors and current situation of soil erosion in the alpine grassland ecosystems of the SPYR are significant for protecting the ecological and productive functions. Based on the 137Cs element tracing technique and machine learning algorithms, five strategic variable selection algorithms based on machine learning algorithms are used to identify the minimal optimal set and analyze the main factors that influence soil erosion in the SPYR. The optimal model for estimating soil erosion in the SPYR is obtained by comparisons model outputs between the RUSLE and machine learning algorithms combined with variable selection models. We identify the spatial distribution pattern of soil erosion in the study area by the optimal model. The results indicated that: (1) A comprehensive set of variables is more objective than the RUSLE model. In terms of verification accuracy, the simulated annealing -Cubist model (R = 0.67, RMSD = 1,368 t km–2⋅a–1) simulation results represents the best while the RUSLE model (R = 0.49, RMSD = 1,769 t⋅km–2⋅a–1) goes on the worst. (2) The soil erosion is more severe in the north than the southeast of the SPYR. The average erosion modulus is 6,460.95 t⋅km–2⋅a–1 and roughly 99% of the survey region has an intensive erosion modulus (5,000–8,000 t⋅km–2⋅a–1). (3) Total erosion loss is relatively 8.45⋅108 t⋅a–1 in the SPYR, which is commonly 12.64 times greater than the allowable soil erosion loss. The economic monetization of SOC loss caused by soil erosion in the entire research area was almost $47.90 billion in 2014. These results will help provide scientific evidences not only for farmers and herdsmen but also for environmental science managers and administrators. In addition, a new ecological policy recommendation was proposed to balance grassland protection and animal husbandry economic production based on the value of soil erosion reclassification.

Designing an Optimal Model of Blood Logistics Management with the Possibility of Return in the Three-Level Blood Transfusion Network

Inventory managers in the blood supply chain always seek timely and proper response to their customers, which is essential because of the perishability and uncertainty of blood demand and the direct relationship of its presence or non-presence with human life. On the other hand, timely and regular delivery of blood to consumers is vital, as the weakness in delivery and transportation policies results in increased shortages, returns, blood loss and significant decrease in the quality of blood required by patients. Given the significance of this for the blood transfusion network, the paper tried to design a comprehensive and integrated optimal model of blood transfusion network logistics management by blood group to reduce the cost of losses, returns and blood shortages. This model is divided into two parts: Inventory management and routing. A combination of simulation techniques and neural network with several recurrent layers was used to evaluate the optimal inventory management and a multi-objective planning model was designed to determine the delivery and distribution of blood to consumers. The model designed was implemented in Khorasan Razavi Blood Transfusion Network with a main base, six central bases and 54 hospitals. Solving the model led to estimating the f consumer demand, the optimal value of target inventory and re-ordering point of central bases and hospitals, and blood distribution from the supplier to its consumers that decreased the units of blood returned to bases, increased inventory availability, and reduced costs significantly.