Machine learning-based prediction of seasonal hypoxia in eutrophic estuary using capacitive potentiometric sensor

A hypoxia occurred in eutrophic estuary was predicted using long short-term memory (LSTM) model with prediction time steps (PTSs) of 0, 1, 12, and 24 h. A capacitive potential (CP), which provides quantitative information on dissolved oxygen (DO) concentration, was used as a predictor along with precipitation, tide level, salinity, and water temperature. First, annual changes in DO concentration were clustered in three phases of annual DO trends (oversaturation, depletion, and stable) using k-means clustering. CP was the most influential variable in clustering the DO phases. The LSTM was implemented to predict the DO phases and hypoxia occurrences. In the simultaneous prediction of the depletion phase and hypoxia occurrence with a 12 h PTS, the accuracy was 92.1% using CP along with other variables; it was 3.3% higher than that achieved using variables other than CP. In the case of predicting the depletion phase and hypoxia non-occurrence using CP along with other variables, the accuracy was 61.1%, which was 5.5% higher than that when CP was not used. When using CP along with other variables, the total accuracy was highest for all PTS. Overall, the utilization of CP and machine learning techniques enables accurate predictions of both short-term and long-term hypoxia occurrences, providing us with the opportunity to proactively respond to disasters in aquaculture and environmental management due to hypoxia.

Machine learning estimation of biodegradable organic matter concentrations in municipal wastewater

This study investigates whether a novel estimation method based on machine learning can feasibly predict the readily biodegradable chemical oxygen demand (RB-COD) and slowly biodegradable COD (SB-COD) in municipal wastewater from the oxidation-reduction potential (ORP) data of anoxic batch experiments. Anoxic batch experiments were conducted with highly mixed liquor volatile suspended solids under different RB-COD and SB-COD conditions. As the RB-COD increased, the ORP breakpoint appeared earlier, and fermentation occurred in the interior of the activated sludge, even under anoxic conditions. Therefore, the ORP decline rates before and after the breakpoint were significantly correlated with the RB-COD and SB-COD, respectively (p < 0.05). The two biodegradable CODs were estimated separately using six machine learning models: an artificial neural network (ANN), support vector regression (SVR), an ANN-based AdaBoost, a SVR-based AdaBoost, decision tree, and random forest. Against the ORP dataset, the RB-COD and SB-COD estimation correlation coefficients of SVR-based AdaBoost were 0.96 and 0.88, respectively. To identify which ORP data are useful for estimations, the ORP decline rates before and after the breakpoint were separately input as datasets to the estimation methods. All six machine learning models successfully estimated the two biodegradable CODs simultaneously with accuracies of ≥0.80 from only ORP time-series data. Sensitivity analysis using the Shapley additive explanation method demonstrated that the ORP decline rates before and after the breakpoint obviously contributed to the estimation of RB-COD and SB-COD, respectively, indicating that acquiring the ORP data with various decline rates before and after the breakpoint improved the estimations of RB-COD and SB-COD, respectively. This novel estimation method for RB-COD and SB-COD can assist the rapid control of biological wastewater treatment when the biodegradable organic matter concentration dynamically changes in influent wastewater.

Design and analysis of water quality monitoring and filtration system for different types of water in Malaysia

Water pollution always occurs in Malaysia due to industrial, construction, agriculture, and household activities. River pollution can disturb water supply that eventually affects business and household activities. Thus, water quality monitoring system is needed to detect contaminated water. We developed a water quality monitoring and filtration system controlled by Arduino. The proposed system was designed in Proteus software and ThingSpeak platform was used for real-time monitoring. The main objective of the study was to compare water quality of river, lake and tap water in terms of pH, temperature, turbidity, electrical conductivity and oxidation-reduction potential. If the water quality was not satisfied, the water sample would be filtered through filtration system. Water turbidity level, pH, temperature, electrical conductivity, and oxidation-reduction potential for filtered and nonfiltered water were compared and analyzed according to international and national water quality standard. Besides that, statistical analysis such as box plot and one-way analysis of variance test was applied to validate data from the system. The real-time water quality monitoring system was implemented through data storage, data transfer, and data processing. The system was connected to wireless fidelity whereas the output data was sent to the user and monitored by ThingSpeak. The system can be further upgraded and scaled up to be applied in the main tank at our home or factory. The outcome of this research can be used as a reference for further study on lake and river pollution monitoring system.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s13762-022-04192-x.

Rapid screening of critically ill patients for low plasma vitamin C concentrations using a point-of-care oxidation-reduction potential measurement

Background

Hypovitaminosis C and vitamin C deficiency are common in critically ill patients and associated with organ dysfunction. Low vitamin C status often goes unnoticed because determination is challenging. The static oxidation reduction potential (sORP) reflects the amount of oxidative stress in the blood and is a potential suitable surrogate marker for vitamin C. sORP can be measured rapidly using the RedoxSYS system, a point-of-care device. This study aims to validate a model that estimates plasma vitamin C concentration and to determine the diagnostic accuracy of sORP to discriminate between decreased and higher plasma vitamin C concentrations.

Methods

Plasma vitamin C concentrations and sORP were measured in a mixed intensive care (IC) population. Our model estimating vitamin C from sORP was validated by assessing its accuracy in two datasets. Receiver operating characteristic (ROC) curves with areas under the curve (AUC) were constructed to show the diagnostic accuracy of sORP to identify and rule out hypovitaminosis C and vitamin C deficiency. Different cut-off values are provided.

Results

Plasma vitamin C concentration and sORP were measured in 117 samples in dataset 1 and 43 samples in dataset 2. Bias and precision (SD) were 1.3 ± 10.0 µmol/L and 3.9 ± 10.1 µmol/L in dataset 1 and 2, respectively. In patients with low plasma vitamin C concentrations, bias and precision were − 2.6 ± 5.1 µmol/L and − 1.1 ± 5.4 µmol in dataset 1 (n = 40) and 2 (n = 20), respectively. Optimal sORP cut-off values to differentiate hypovitaminosis C and vitamin C deficiency from higher plasma concentrations were found at 114.6 mV (AUC 0.91) and 124.7 mV (AUC 0.93), respectively.

Conclusion

sORP accurately estimates low plasma vitamin C concentrations and can be used to screen for hypovitaminosis C and vitamin C deficiency in critically ill patients. A validated model and multiple sORP cut-off values are presented for subgroup analysis in clinical trials or usage in clinical practice.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40635-021-00403-w.

Optimization of a diagnostic platform for oxidation-reduction potential (ORP) measurement in human plasma

Objectives: Oxidation–reduction potential (ORP) measurement can demonstrate the extent of oxidative stress in patients with severe illness and/or injury. A novel ORP diagnostic platform using disposable sensors (RedoxSYS) has been validated by comparison to mass spectrometry, but the optimal methods of sample handling for best performance of the device have not been described.

Methods: We sought to optimize ORP measurement in human plasma under controlled conditions. We hypothesized that the anticoagulant, freeze–thawing, and storage duration would influence measured ORP levels.

Results: The platform was sensitive to exogenous oxidation with hydrogen peroxide and reduction with ascorbic acid. Plasma anticoagulated with heparin was more sensitive to differences in ORP than plasma prepared in citrate. ORP measurements decreased slightly after a freeze–thaw cycle, but once frozen, ORP was stable for up to one month.

Discussion: We confirm that ORP detects oxidative stress in plasma samples. Optimal measurement of plasma ORP requires blood collection in heparin anticoagulant tubes and immediate analysis without a freeze–thaw cycle.

Stimulative mineralization of p-fluoronitrobenzene in biocathode microbial electrolysis cell with an oxygen-limited environment

p-Fluoronitrobenzene (p-FNB) is a toxic compound and tends to accumulate in the environment. p-FNB can be effectively removed and defluorinated in a single-chamber bioelectrochemical system (BES). To verify the suppositionally integrated reductive and oxidative metabolism mechanism in the BES, an oxygen-limited environment was used, with pure oxygen and nitrogen environments used as two controls. Under the oxygen-limited condition, the most excellent performance was achieved. The defluorination rate and mineralization efficiency were 0.0132h(-1) and 72.99±5.68% after 96h, with 75.4% of fluorine in the form of the fluoride ion. This resulted from the unique environment that allowed conventionally integrated reductive and oxidative catabolism. Moreover, the oxidation-reduction potential (ORP) had a significant effect on microbial communities, which was also an important reason for performance diversity. These results provide a new method for complete p-FNB treatment and a control strategy by ORP regulation for optimal system performance.

Effects of nitrate on the treatment of lead contaminated groundwater by nanoscale zerovalent iron

Nanoscale zerovalent iron (nZVI) is efficient for removing Pb(2+) and nitrate from water. However, the influence of nitrate, a common groundwater anion, on Pb(2+) removal by nZVI is not well understood. In this study, we showed that under excess Fe(0) conditions (molar ratio of Fe(0)/nitrate>4), Pb(2+) ions were immobilized more quickly (<5 min) than in nitrate-free systems (∼ 15 min) due to increasing pH. With nitrate in excess (molar ratio of Fe(0)/nitrate<4), nitrate stimulated the formation of crystal PbxFe3-xO4 (ferrite), which provided additional Pb(2+) removal. However, ∼ 7% of immobilized Pb(2+) ions were released into aqueous phase within 2h due to ferrite deformation. Oxidation-reduction potential (ORP) values below -600 mV correlated with excess Fe(0) conditions (complete Pb(2+) immobilization), while ORP values ≥-475 mV characterized excess nitrate conditions (ferrite process and Pb(2+) release occurrence). This study indicates that ORP monitoring is important for proper management of nZVI-based remediation in the subsurface to avoid lead remobilization in the presence of nitrate.