Real-time eutrophication status evaluation of coastal waters using support vector machine with grid search algorithm

In situ H2O2 treatment of blue-green algae contaminated reservoirs causes significant improvement in drinking water treatability

The evaluation of water quality improvement brought about by in situ treatment of eutrophic water bodies, especially those used for human supply is a challenging task since each water system responds differently. To overcome this challenge, we applied exploratory factor analysis (EFA) to understand the effects of using hydrogen peroxide (H₂O₂) on eutrophic water used as a drinking water supply. This analysis was used to identify the main factors that described the water treatability after exposing blue-green algae (cyanobacteria) contaminated raw water to H₂O₂ at both 5 and 10 mg L^-1. Cyanobacterial chlorophyll-a was undetectable following the application of both concentrations of H₂O₂ after four days, while not causing relevant changes to green algae and diatoms chlorophyll-a concentrations. EFA demonstrated that the main factors affected by both H₂O₂ concentrations were turbidity, pH, and cyanobacterial chlorophyll-a concentration, which are important variables for a drinking water treatment plant. The H₂O₂ caused significant improvement in water treatability by decreasing those three variables. Finally, the use of EFA was demonstrated to be a promising tool in identifying which limnological variables are most relevant concerning the efficacy of water treatment, which in turn can make water quality monitoring more efficient and less costly.

A soft sensor for simulating algal cell density based on dynamic response to environmental changes in a eutrophic shallow lake

There is a great need for timely monitoring and rapid water quality assessment to control the algal blooms that often occur in eutrophic lakes. While algal cell density (ACD) is a critical indicator of algal growth, field monitoring is laborious and time-consuming, and rapid assessment of algal blooms based on ACD is often not possible. To address the limitations of conventional ACD detection, we proposed a soft sensor approach that uses surrogate indicators to simulate ACD in machine learning models. We conducted a case study using monitoring data from Chaohu Lake collected between 2016 and 2019. We found that ensemble learning models, especially extreme gradient boosting (XGBoost), outperformed traditional machine learning algorithms by comparing various machine learning algorithms. Also, considering the influence of input variable selection on model performance, we combined the results of different filter methods in the multi-stage variable selection process. Finally, we screened out seven key variables out of the 43 initial candidate variables, including dissolved oxygen (DO), chlorophyll-a (Chl-a), Secchi disk depth (SD), pH, permanganate index (COD_Mn), week of the year (WOY), and wind velocity (WV). Their inclusion substantially improved data accessibility and supported the development of a rapid simulation model. The final model was capable of reliable spatiotemporal generalization, with an overall R² value of 0.761. On the theoretical side, our study makes a new attempt to simulate ACD values in a eutrophic lake. For practical purposes, the soft sensor can facilitate the rapid assessment of bloom conditions, which helps the local administration with emergency prevention and control.

Comprehensive modelling and cost-benefit optimization for joint regulation of algae in urban water system

Algal blooms in urban water system is an international concern, which especially in China, have become a major obstacle to the urban water environment improvement since the preliminary achievements were made in the treatment of black and odorous water bodies. The complex blooming mechanisms require a joint regulation plan. This study established a framework that consisted of three steps, i.e., simulation, optimization, and verification, to build an optimal joint regulation plan. By taking the urban river network in Suzhou Pingjiang Xincheng as a case study, the cost-benefits of six alternative regulation measures were assessed using an algal bloom mechanism model and the discounted cash flow model based on 70 regulation scenarios. The joint regulation plan was optimized using the marginal-cost-based greedy strategy on the basis of the cost-benefits of different measures. The optimized joint plans, which were verified to be global optima, were more cost-effective than the designed regulation scenarios, and reduced the average chlorophyll-a concentrations by 55.3%-60.1% compared with the status quo. Applying the optimized cost allocation ratios of each measure to adjust the existing regulation scheme of another similar case verified that the optimization results had great generalizability.

Solid-Phase Extraction of Aquatic Organic Matter: Loading-Dependent Chemical Fractionation and Self-Assembly

Dissolved organic matter (DOM) is an important component in marine and freshwater environments and plays a fundamental role in global biogeochemical cycles. In the past, optical and molecular-level analytical techniques evolved and improved our mechanistic understanding about DOM fluxes. For most molecular chemical techniques, sample desalting and enrichment is a prerequisite. Solid-phase extraction has been widely applied for concentrating and desalting DOM. The major aim of this study was to constrain the influence of sorbent loading on the composition of DOM extracts. Here, we show that increased loading resulted in reduced extraction efficiencies of dissolved organic carbon (DOC), fluorescence and absorbance, and polar organic substances. Loading-dependent optical and chemical fractionation induced by the altered adsorption characteristics of the sorbent surface (styrene divinylbenzene polymer) and increased multilayer adsorption (DOM self-assembly) can fundamentally affect biogeochemical interpretations, such as the source of organic matter. Online fluorescence monitoring of the permeate flow allowed to empirically model the extraction process and to assess the degree of variability introduced by changing the sorbent loading in the extraction procedure. Our study emphasizes that it is crucial for sample comparison to keep the relative DOC loading (DOC_load [wt %]) on the sorbent always similar to avoid chemical fractionation.

Product Service System Configuration Based on a PCA-QPSO-SVM Model

To achieve sustainable development and improve market competitiveness, many manufacturers are transforming from traditional product manufacturing to service manufacturing. In this trend, the product service system (PSS) has become the mainstream of supply to satisfy customers with individualized products and service combinations. The diversified customer requirements can be realized by the PSS configuration based on modular design. PSS configuration can be deemed as a multi-classification problem. Customer requirements are input, and specific PSS is output. This paper proposes an improved support vector machine (SVM) model optimized by principal component analysis (PCA) and the quantum particle swarm optimization (QPSO) algorithm, which is defined as a PCA-QPSO-SVM model. The model is used to solve the PSS configuration problem. The PCA method is used to reduce the dimension of the customer requirements, and the QPSO is used to optimize the internal parameters of the SVM to improve the prediction accuracy of the SVM classifier. In the case study, a dataset for central air conditioning PSS configuration is used to construct and test the PCA-QPSO-SVM model, and the optimal PSS configuration can be predicted well for specific customer requirements.

Identify specific gene pairs for subarachnoid hemorrhage based on wavelet analysis and genetic algorithm

Subarachnoid hemorrhage (SAH) is a fatal stroke caused by bleeding in the brain. SAH can be caused by a ruptured aneurysm or head injury. One-third of patients will survive and recover. One-third will survive with disability; one-third will die. The focus of treatment is to stop bleeding, restore normal blood flow, and prevent vasospasm. Treatment for SAH varies, depending on the bleeding’s underlying cause and the extent of damage to the brain. Treatment may include lifesaving measures, symptom relief, repair of the bleeding vessel, and complication prevention. However, the useful diagnostic biomarkers of SAH are still limited due to the instability of gene marker expression. To overcome this limitation, we developed a new protocol pairing genes and screened significant gene pairs based on the feature selection algorithm. A classifier was constructed with the selected gene pairs and achieved a high performance.

Estimating Coastal Chlorophyll-A Concentration from Time-Series OLCI Data Based on Machine Learning

Chlorophyll-a (chl-a) is an important parameter of water quality and its concentration can be directly retrieved from satellite observations. The Ocean and Land Color Instrument (OLCI), a new-generation water-color sensor onboard Sentinel-3A and Sentinel-3B, is an excellent tool for marine environmental monitoring. In this study, we introduce a new machine learning model, Light Gradient Boosting Machine (LightGBM), for estimating time-series chl-a concentration in Fujian’s coastal waters using multitemporal OLCI data and in situ data. We applied the Case 2 Regional CoastColour (C2RCC) processor to obtain OLCI band reflectance and constructed four spectral indices based on OLCI feature bands as supplementary input features. We also used root-mean-square error (RMSE), mean absolute error (MAE), median absolute percentage error (MAPE), and R2 as performance indicators. The results indicate that the addition of spectral indices can easily improve the prediction accuracy of the model, and normalized fluorescence height index (NFHI) has the best performance, with an RMSE of 0.38 µg/L, MAE of 0.22 µg/L, MAPE of 28.33%, and R2 of 0.785. Moreover, we used the well-known band ratio and three-band methods for chl-a estimation validation, and another two OLCI chl-a products were adopted for comparison (OC4Me chl-a and Inverse Modelling Technique (IMT) Neural Net chl-a). The results confirmed that the LightGBM model outperforms the traditional methods and OLCI chl-a products. This study provides an effective remote sensing technique for coastal chl-a concentration estimation and promotes the advantage of OLCI data in ocean color remote sensing.

Face Recognition Based on SVM Optimized by the Improved Bacterial Foraging Optimization Algorithm

Support vector machine (SVM) is always used for face recognition. However, kernel function selection (kernel selection and its parameters selection) is a key problem for SVMs, and it is difficult. This paper tries to make some contributions to this problem with focus on optimizing the parameters in the selected kernel function. Bacterial foraging optimization algorithm, inspired by the social foraging behavior of Escherichia coli, has been widely accepted as a global optimization algorithm of current interest for distributed optimization and control. Therefore, we proposed to optimize the parameters in SVM by an improved bacterial foraging optimization algorithm (IBFOA). In the improved version of bacterial foraging optimization algorithm, a dynamical elimination-dispersal probability in the elimination-dispersal step and a dynamical step size in the chemotactic step are used to improve the performance of bacterial foraging optimization algorithm. Then the optimized SVM is used for face recognition. Simultaneously, an improved local binary pattern is proposed to extract features of face images in this paper to improve the accuracy rate of face recognition. Numerical results show the advantage of our algorithm over a range of existing algorithms.

Recognition of Lung Adenocarcinoma-specific Gene Pairs Based on Genetic Algorithm and Establishment of a Deep Learning Prediction Model

AIM AND OBJECTIVE

Lung cancer is a disease with a dismal prognosis and is the major cause of cancer deaths in many countries. Nonetheless, rapid technological developments in genome science guarantees more effective prevention and treatment strategies.

MATERIALS AND METHODS

In this study, genes were pair-matched and screened for lung adenocarcinomaspecific gene relationships. False positives due to fluctuations in single gene expression were avoided and the stability and accuracy of the results was improved.

RESULTS

Finally, a deep learning model was constructed with machine learning algorithm to realize the clinical diagnosis of lung adenocarcinoma in patients.

CONCLUSION

Comparing with the traditional methods which takes ingle gene as a feature, the relative difference between gene pairs is a higher order feature, leverage high-order features to build the model can avoid instability caused by a single gene mutation, making the prediction results more reliable.

Assessing the severe eutrophication status and spatial trend in the coastal waters of Zhejiang province (China)

The eutrophication of the coastal waters of Zhejiang Province has become one of the main contamination threats to the region's coastal marine ecosystems. Accordingly, the comprehensive characterization of the eutrophication status in terms of improved quantitative methods is valuable for local risk assessment and policy making. A novelty of this work is that the spatial distributions of chemical oxygen demand, dissolved inorganic nitrogen, and dissolved inorganic phosphorus were estimated across space by the Bayesian maximum entropy (BME) method. The BME estimates were found to have the best cross‐validation performance compared to ordinary kriging and inverse distance weighted techniques. Based on the BME maps, it was found that about 25.95%, 19.18%, 20.53%, and 34.34% of these coastal waters were oligotrophic, mesotrophic, eutrophic, and hypereutrophic. Another novelty of the present work is that comprehensive stochastic site indicators (SSI) were introduced in the quantitative characterization of the eutrophication risk in the Zhejiang coastal waters under conditions of in situ uncertainty. The results showed that the level of the eutrophication index (EI) increased almost linearly with increasing threshold values; and that 71%, 51%, and 19% of coastal locations separated by various spatial lags experience considerable mesotrophic, eutrophic, and hypereutrophic risks, respectively. The average EI values over the subregions of the Zhejiang coastal waters graded as “oligotrophic or higher,” “eutrophic or higher,” and “hypereutrophic” were about 11.14, 14.28, and 25.34, respectively. Our results also revealed that the joint eutrophication strength between coastal locations in the Zhejiang region was consistently greater than the combined strength of independent eutrophications at these locations (we termed this situation “positive quadrant eutrophication dependency”). It was found that a critical eutrophication threshold ζcr ≈ 8.38 exists so that below ζcr the spatial eutrophication dependency in the Zhejiang coastal waters increases with ζ, whereas above ζcr the opposite is true. Moreover, the eutrophication dependency decreases as the separation distance δs increases. Interestingly, at distances δs smaller than a critical distance δscr ≈ 15  km, the eutrophication locations are concentrated in the coastal waters of the Zhejiang province rather than being dispersed (this observation holds even for large thresholds ζ). Elasticity analysis of eutrophication indicators offered a quantitative measure of the excess eutrophication change in the Zhejiang coastal waters caused by a threshold change (the larger the elasticity is, the more sensitive eutrophication is to threshold changes). The above findings can contribute to an improved understanding of seawater quality and provide a practical approach for the identification of critical coastal water regions.

Identification of an early diagnostic biomarker of lung adenocarcinoma based on co-expression similarity and construction of a diagnostic model

Background

The purpose of this study was to achieve early and accurate diagnosis of lung cancer and long-term monitoring of the therapeutic response.

Methods

We downloaded GSE20189 from GEO database as analysis data. We also downloaded human lung adenocarcinoma RNA-seq transcriptome expression data from the TCGA database as validation data. Finally, the expression of all of the genes underwent z test normalization. We used ANOVA to identify differentially expressed genes specific to each stage, as well as the intersection between them. Two methods, correlation analysis and co-expression network analysis, were used to compare the expression patterns and topological properties of each stage. Using the functional quantification algorithm, we evaluated the functional level of each significantly enriched biological function under different stages. A machine-learning algorithm was used to screen out significant functions as features and to establish an early diagnosis model. Finally, survival analysis was used to verify the correlation between the outcome and the biomarkers that we found.

Results

We screened 12 significant biomarkers that could distinguish lung cancer patients with diverse risks. Patients carrying variations in these 12 genes also presented a poor outcome in terms of survival status compared with patients without variations.

Conclusions

We propose a new molecular-based noninvasive detection method. According to the expression of the stage-specific gene set in the peripheral blood of patients with lung cancer, the difference in the functional level is quantified to realize the early diagnosis and prediction of lung cancer.