Rapid Triage of Children with Suspected COVID-19 Using Laboratory-Based Machine-Learning Algorithms.

In order to limit the spread of the novel betacoronavirus (SARS-CoV-2), it is necessary to detect positive cases as soon as possible and isolate them. For this purpose, machine-learning algorithms, as a field of artificial intelligence, have been recognized as a promising tool. The aim of this study was to assess the utility of the most common machine-learning algorithms in the rapid triage of children with suspected COVID-19 using easily accessible and inexpensive laboratory parameters. A cross-sectional study was conducted on 566 children treated for respiratory diseases: 280 children with PCR-confirmed SARS-CoV-2 infection and 286 children with respiratory symptoms who were SARS-CoV-2 PCR-negative (control group). Six machine-learning algorithms, based on the blood laboratory data, were tested: random forest, support vector machine, linear discriminant analysis, artificial neural network, k-nearest neighbors, and decision tree. The training set was validated through stratified cross-validation, while the performance of each algorithm was confirmed by an independent test set. Random forest and support vector machine models demonstrated the highest accuracy of 85% and 82.1%, respectively. The models demonstrated better sensitivity than specificity and better negative predictive value than positive predictive value. The F1 score was higher for the random forest than for the support vector machine model, 85.2% and 82.3%, respectively. This study might have significant clinical applications, helping healthcare providers identify children with COVID-19 in the early stage, prior to PCR and/or antigen testing. Additionally, machine-learning algorithms could improve overall testing efficiency with no extra costs for the healthcare facility.

Multivariate analysis of water quality parameters in Lake Palic, Serbia.

This study presents a comprehensive investigation of water quality parameters in the fourth sector of Lake Palic in Serbia, which has a regional strategic importance. Namely, it is designated as a tourist destination. What is perhaps even more important is that its surplus water ends up in Lake Ludas, a significant habitat for migrating and aquatic bird species, and it is a RAMSAR site. The conducted analysis points to the major conclusion that the reasons for very high Chlorophyll-a values can be found in considerable anthropogenic pressures exerted on the studied area. Due to these pressures, the lake is not in ecological equilibrium. To support this conclusion, an in-depth analysis was conducted using water quality measurements for 9 years, from 2011 to 2019. The data was subject to principal component analysis (PCA) and machine learning classification algorithms that identified a seasonal character regarding the lake's water quality. Water quality indexes (WQI) were determined using two approaches to provide a more general insight into the lake's overall quality. Keeping in mind the large number of data gathered monthly within the Palic-Ludas Lake system, fitted models for estimating certain water quality parameters were also developed. This was accomplished via multivariate regression, resulting in a number of equations that can, using a few basic input parameters, predict values of ammonium nitrogen, Chlorophyll-a, and 5-day biological oxygen demand. The fitted models were obtained for relatively homogeneous periods within a year identified by cluster analysis.

Field measurements on alluvial watercourses in light of numerical modeling: case studies on the Danube River.

Adequate monitoring and data acquisition of proper hydraulic, sediment, and constituent parameters in alluvial watercourses have become crucial aspects of human interaction with the environment. Conducting well-organized, comprehensive, and meaningful field measurements on natural watercourses are of great importance when assessing its hydraulic, morphological, and ecological state. However, this paper presents a methodology for field measurements on alluvial watercourses in light of numerical modeling. The proposed methodology focuses on collecting field data sets to calibrate numerical models for flow, sediment, and heavy metal transport. The proposed approach targets the simultaneous measurement of hydraulic, sediment transport, and heavy metal transport parameters that are key for calibrating constants and exchange mechanisms in contemporary numerical models. Using the principles laid out in this paper, two sets of measurements were carried out on the Danube River, one on a reach near Mohács in Hungary and the other on a reach near Belgrade in Serbia. The first case study discusses the measurement and results of comprehensive hydraulic and sediment parameters. The second case study considers hydraulic and sediment measurements complemented with trace metal measurements for zinc, lead, and mercury. These measurements were used for calibrating numerical models for flow, sediment, and heavy metal transport, as a proof of concept. It has been demonstrated that the gathered data sets contain key parameters that are strongly linked through physical laws and are needed for calibration purposes, as well as parameters that can allow the newly calibrated coefficients to be confirmed through other measured phenomena. Therefore, the proposed methodology provides minimal data sets with detailed measurements for calibrating numerical models for flow, sediment, and heavy metal transport. Guidelines for future measurements that can suffice the increasing need for numerical modeling and monitoring of natural watercourses are also offered.

Sediment transport and bed evolution model for complex river systems.

This work presents the theoretical background, development, and preliminary evaluation of a one-dimensional unsteady sediment transport and bed evolution model for a looped river network. The sediment transport and bed evolution model employs a concept that differentiates sediment particles moving in the form of suspended sediment, and near bed, and bed sediment. Applying the active layer concept, the developed model utilizes the appropriate exchange mechanisms between the suspended sediment and active layer material, which are presented in great detail. The governing transport equations were solved using the split operator approach that resulted in two successive steps. The advection equations were solved using the characteristics method, whereas the diffusion equations were discretized using the Crank-Nicholson scheme. The obtained system was complemented with additional auxiliary equations in order to allow sediment transport and bed evolution simulation at nodes, thus enabling the same in a looped river network. The derived equations were applied to develop an open-channel flow, sediment transport, and bed evolution model that was subjected to a series of preliminary numerical tests.

Hydraulic analysis of a meander on the Danube River using a 2D flow model.

This paper presents the development, calibration and verification of a two-dimensional model for a Danube reach and its old cutoff meander. The considered meander was at one point separated from the main reach with a levee, which caused a series of unwanted environmental consequences. Aiming to stop the ongoing degradation of the meander, the validated model was engaged to investigate the effects different river works would have on its current state. The considered river works involved dredging in certain parts of the meander, while keeping in mind the negative effects these works can have on the environment, as well as possible widening of the existing opening in the levee. Numerical simulations showed that all the considered scenarios would result in a general increase of velocities, which is important from the aspect of sediment deposition. After a thorough evaluation of the results, it was found that by carefully selecting the locations where dredging should be conducted, no other interference is needed to put an end to the increasingly deteriorating situation in the meander. Computations confirmed that the suggested river works would increase the unit discharge in the entire meander, especially in its upstream part. The redistribution of flow around the island downstream of the levee (that formed during past years as a result of sediment deposition) was also predicted, where the flow would once again favor the left side of the island. This is a more natural path of flow that would aid the flushing of previously deposited sediment in this part of the meander.

Implementation of a monitoring approach: the Palic-Ludas lake system in the Republic of Serbia.

This paper presents the development and implementation of a monitoring procedure on the Palic-Ludas lake system in the north of the Republic of Serbia. After evaluating the system's current monitoring approach, additional measurements were conducted for both hydraulic and water quality parameters. The considered area is a complex system that incorporates lakes, hydraulic structures, and a channel, with a waste water treatment plant on its upstream, and a nature protection area on its downstream end, making its monitoring and management an interesting challenge. The available data was used to analyze both spacial and temporal changes of water quality parameters throughout the area. Conclusions were made regarding the onset of anoxia and the possible release of nutrients from sediment in some parts of the lake system. Using the available data and the Canadian Council of Ministers of the Environment Water Quality Index (CCME WQI) method, it was determined that the current status of the lakes is between poor and marginal. This result certainly adds to the importance of enhancing the current monitoring approach. Consequently, the authors proposed the implementation of a monitoring system, so that the gathered information would be useful for decision makers. The CCME WQI method proved to be a beneficial way to describe and monitor the overall state of this complex lake system; therefore, future data gathering activity should be conceived in a way to facilitate this method of computing the water quality index.

Impaired cerebrovascular reactivity in chronic obstructive pulmonary disease.

Impaired cerebrovascular reactivity (CVR) is associated with stroke. Cerebrovascular diseases are common comorbidity in chronic obstructive pulmonary disease (COPD) patients. The aim of our study was to quantify CVR in the anterior and posterior cerebral circulation during voluntary breath-holding in COPD patients according to airflow limitation severity. In this cross-sectional study, we compared 90 COPD patients without previous cerebrovascular disease and 30 age- and sex-matched healthy volunteers (mean age 67 ± 7.9, 87 males). Using transcranial Doppler ultrasound and breath-holding index (BHI), we analysed baseline mean flow velocities (MFV) and CVR of middle cerebral artery (MCA) and basilar artery (BA). Our results demonstrated that COPD patients had lower baseline MFV of both MCA and BA than controls. COPD patients had significantly lower BHImMCA and BHImBA than controls (0.8 and 0.7 versus 1.24 and 1.07, respectively; p < 0.001). With the severity of airflow obstruction, there were significant declines of BHImMCA and BHImBA in mild (0.94 and 0.83), moderate (0.8 and 0.7) and severe to very severe COPD (0.7 and 0.6), respectively (p < 0.001). For all participants, we found a significant and positive correlation between forced expiratory volume in one second (FEV1) and BHImMCA (Rho = 0.761, p < 0.001) and between FEV1 and BHImBA (Rho = 0.409, p < 0.001). COPD patients have impaired CVR in anterior and posterior cerebral circulation. Impairment of CVR increase with the airflow limitation severity. CVR is an appropriate marker to identify vulnerable COPD subjects at high risk to develop cerebrovascular disease. Prospective studies are needed for further evaluation.