Rapid identification of medicinal plants via visual feature-based deep learning.

BACKGROUND: Traditional Chinese Medicinal Plants (CMPs) hold a significant and core status for the healthcare system and cultural heritage in China. It has been practiced and refined with a history of exceeding thousands of years for health-protective affection and clinical treatment in China. It plays an indispensable role in the traditional health landscape and modern medical care. It is important to accurately identify CMPs for avoiding the affected clinical safety and medication efficacy by the different processed conditions and cultivation environment confusion. RESULTS: In this study, we utilize a self-developed device to obtain high-resolution data. Furthermore, we constructed a visual multi-varieties CMPs image dataset. Firstly, a random local data enhancement preprocessing method is proposed to enrich the feature representation for imbalanced data by random cropping and random shadowing. Then, a novel hybrid supervised pre-training network is proposed to expand the integration of global features within Masked Autoencoders (MAE) by incorporating a parallel classification branch. It can effectively enhance the feature capture capabilities by integrating global features and local details. Besides, the newly designed losses are proposed to strengthen the training efficiency and improve the learning capacity, based on reconstruction loss and classification loss. CONCLUSIONS: Extensive experiments are performed on our dataset as well as the public dataset. Experimental results demonstrate that our method achieves the best performance among the state-of-the-art methods, highlighting the advantages of efficient implementation of plant technology and having good prospects for real-world applications.

Effects of Fe(III) on the formation and toxicity alteration of halonitromethanes, dichloroacetonitrile, and dichloroacetamide from polyethyleneimine during UV/chlorine disinfection.

Trace iron ions (Fe(III)) are commonly found in water and wastewater, where free chlorine is very likely to coexist with Fe(III) affecting the disinfectant's stability and N-DBPs' fate during UV/chlorine disinfection, and yet current understanding of these mechanisms is limited. This study investigates the effects of Fe(III) on the formation and toxicity alteration of halonitromethanes (HNMs), dichloroacetonitrile (DCAN), and dichloroacetamide (DCAcAm) from polyethyleneimine (PEI) during UV/chlorine disinfection. Results reveal that the maxima concentrations of HNMs, DCAN, and DCAcAm during UV/chlorine disinfection with additional Fe(III) were 1.39, 1.38, and 1.29 times higher than those without additional Fe(III), instead of being similar to those of Fe(III) inhibited the formation of HNMs, DCAN and DCAcAm during chlorination disinfection. Meanwhile, higher Fe(III) concentration, acidic pH, and higher chlorine dose were more favorable for forming HNMs, DCAN, and DCAcAm during UV/chlorine disinfection, which were highly dependent on the involvement of HO· and Cl·. Fe(III) in the aquatic environment partially hydrolyzed to the photoactive Fe(III)­hydroxyl complexes Fe(OH)2+ and [Fe(H2O)6]3+, which undergone UV photoactivation and coupling reactions with HOCl to achieve effective Fe(III)/Fe(II) interconversion, a process that facilitated the sustainable production of HO·. Extensive product analysis and comparison verified that the HO· production enhanced by the Fe(III)/Fe(II) internal cycle played a primary role in increasing HNMs, DCAN, and DCAcAm productions during UV/chlorine disinfection. Note that the incorporation of Fe(III) increased the cytotoxicity and genotoxicity of HNMs, DCAN, and DCAcAm formed during UV/chlorine disinfection, and yet Fe(III) did not have a significant effect on the acute toxicity of water samples before, during, and after UV/chlorine disinfection. The new findings broaden the knowledge of Fe(III) affecting HNMs, DCAN, and DCAcAm formation and toxicity alteration during UV/chlorine disinfection.

Reaction mechanisms of chlorinated disinfection byproducts formed from nitrophenol compounds with different structures during chlor(am)ination and UV/post-chlor(am)ination.

Nitrophenol compounds (NCs) have high formation potentials of disinfection byproducts (DBPs) in water disinfection processes, however, the reaction mechanisms of DBPs formed from different NCs are not elucidated clearly. Herein, nitrobenzene, phenol, and six representative NCs were used to explore the formation mechanisms of chlorinated DBPs (Cl-DBPs) during chlor(am)ination and UV/post-chlor(am)ination. Consequently, the coexistence of nitro and hydroxy groups in NCs facilitated the electrophilic substitution to produce intermediates of Cl-DBPs, and the different positions of nitro and hydroxy groups also induced different yields and formation mechanisms of Cl-DBPs during the chlorination and UV/post-chlorination processes. Besides, the amino, chlorine, and methyl groups significantly influenced the formation mechanisms of Cl-DBPs during the chlorination and UV/post-chlorination processes. Furthermore, the total Cl-DBPs yields from the six NCs followed a decreasing order of 2-chloro-3-nitrophenol, 3-nitrophenol, 2-methyl-3-nitrophenol, 2-amino-4-nitrophenol, 2-nitrophenol, and 4-nitrophenol during chlorination and UV/post-chlorination. However, the total Cl-DBPs yields from the six NCs during chloramination and UV/post-chloramination followed a quite different order, which might be caused by additional reaction mechanisms, e.g., nucleophilic substitution or addition might occur to NCs in the presence of monochloramine (NH2Cl). This work can offer deep insights into the reaction mechanisms of Cl-DBPs from NCs during the chlor(am)ination and UV/post-chlor(am)ination processes.

Comparative analysis of chlorinated disinfection byproducts formation from 4-nitrophenol and 2-amino-4-nitrophenol during UV/post-chlorination.

Nitrophenol compounds (NCs) are widely distributed in water environments and regarded as important precursors of disinfection byproducts (DBPs). Herein, 4-nitrophenol and 2-amino-4-nitrophenol were selected as representative NCs to explore chlorinated DBPs (Cl-DBPs) formation during UV/post-chlorination. Dichloronitromethane (DCNM), trichloronitromethane (TCNM), dichloroacetonitrile (DCAN), and trichloromethane (TCM) were formed from 4-nitrophenol and 2-amino-4-nitrophenol during UV/post-chlorination, and the yields of individual Cl-DBPs from 2-amino-4-nitrophenol were higher than those from 4-nitrophenol. Meantime, increasing chlorine contact time, UV fluence, and free chlorine dose could enhance Cl-DBPs formation, while much higher values of the three factors might decrease the yields of Cl-DBPs. Besides, alkaline pH could decrease the yields of halonitromethane (HNMs) and DCAN but increase the yields of TCM. Also, higher concentrations of 4-nitrophenol and 2-amino-4-nitrophenol would induce more Cl-DBPs formation. Subsequently, the possible formation pathways of DCNM, TCNM, DCAN, and TCM form 4-nitrophenol and 2-amino-4-nitrophenol during UV/post-chlorination were proposed according to transformation products (TPs) and density functional theory (DFT) calculation. Notably, Cl-DBPs formed from 2-amino-4-nitrophenol presented higher toxicity than those from 4-nitrophenol. Among these generated Cl-DBPs, DCAN and TCNM posed higher cytotoxicity and genotoxicity, respectively. Furthermore, 4-nitrophenol, 2-amino-4-nitrophenol, and their TPs exhibited ecotoxicity. Finally, 4-nitrophenol and 2-amino-4-nitrophenol presented a high potential to produce DCNM, TCNM, DCAN, and TCM in actual waters during UV/post-chlorination, but the Cl-DBPs yields were markedly different from those in simulated waters. This work can help better understand Cl-DBPs formation from different NCs during UV/post-chlorination and is conducive to controlling Cl-DBPs formation.

Effects of cupric ions on the formation of chlorinated disinfection byproducts from nitrophenol compounds during UV/post-chlorination.

Cupric ions (Cu2+) are ubiquitous in surface waters and can influence disinfection byproducts (DBPs) formation in water disinfection processes. This work explored the effects of Cu2+ on chlorinated DBPs (Cl-DBPs) formation from six representative nitrophenol compounds (NCs) during UV irradiation followed by a subsequent chlorination (i.e., UV/post-chlorination), and the results showed Cu2+ enhanced chlorinated halonitromethane (Cl-HNMs) formation from five NCs (besides 2-methyl-3-nitrophenol) and dichloroacetonitrile (DCAN) and trichloromethane (TCM) formation from six NCs. Nevertheless, excessive Cu2+ might reduce Cl-DBPs formation. Increasing UV fluences displayed different influences on total Cl-DBPs formation from different NCs, and increasing chlorine dosages and NCs concentrations enhanced that. Moreover, a relatively low pH (5.8) or high pH (7.8) might control the yields of total Cl-DBPs produced from different NCs. Notably, Cu2+ enhanced Cl-DBPs formation from NCs during UV/post-chlorination mainly through the catalytic effect on nitro-benzoquinone production and the conversion of Cl-DBPs from nitro-benzoquinone. Additionally, Cu2+ could increase the toxicity of total Cl-DBPs produced from five NCs besides 2-methyl-3-nitrophenol. Finally, the impacts of Cu2+ on Cl-DBPs formation and toxicity in real waters were quite different from those in simulated waters. This study is conducive to further understanding how Cu2+ affected Cl-DBPs formation and toxicity in chlorine disinfection processes and controlling Cl-DBPs formation in copper containing water.

An unsupervised water quality anomaly detection method based on a combination of time-frequency analysis and clustering.

Anomalies in water quality, which frequently arise due to pollution, constitute a substantial menace to human health. The preservation of public welfare critically entails the timely recognition of abnormal water quality. Conventional techniques for detecting water quality anomalies face obstacles such as the necessity of expert knowledge, limited accuracy in detection, and delays in identification. In this paper, we proposed an original unsupervised technique for identifying water quality anomalies combined with time-frequency analysis and clustering (TCAD). We chose time-frequency analysis because it effectively evaluates water quality changes, generating distinct multi-band signals that reflect different aspects of water quality dynamics. We also proposed a clustering technique which can identify water quality markers and amalgamate data from multi-band signals for accurate anomaly detection. We seek to clarify the reasoning behind our methodology by portraying how time-frequency analysis and clustering address the deficiencies of conventional methods. Our experiments evaluated various indicators of water quality, and the effectiveness of our proposed approach was supported by comparative analyses with commonly used models for detecting anomalies in water quality.

Bromide induced the formation of brominated halonitromethanes from aspartic acid in the UV/chlorine disinfection process.

Brominated halonitromethanes (Br-HNMs) are generated in water disinfection processes and present high toxicity to human health. This work used aspartic acid (ASP) as the precursor to reveal that bromide (Br-) induced the production of Br-HNMs in the UV/chlorine disinfection process. Consequently, six Br-HNMs were identified, and their yields presented an increasing and then declining evolution over the reaction time from 0 to 15 min. Also, the total Br-HNMs yield reached the maximum of 251.1 µg L-1 at 5 min and then declined to 107.1 µg L-1. The total Br-HNMs yield increased from 2.40 to 251.14 µg L-1 with the increase of Cl2:Br- ratios from 0.25 to 3.0 by increasing free chlorine dosage with a fixed Br- concentration, and it increased from 207.59 to 251.14 µg L-1 and then decreased to 93.44 µg L-1 with the increase of Cl2:Br- ratio from 1.0 to 3.6 by increasing Br- concentration with a fixed free chlorine dosage. Besides, the total Br-HNMs yield reached the highest value (251.14 µg L-1) at pH 7.0 and the lowest value (74.20 µg L-1) at pH 8.0. Subsequently, the possible reaction mechanism of Br-HNMs generated from ASP was deduced, and the changes in toxicity of Br-HNMs also followed an increasing and then declining trend, closely relating to Br-HNMs yields and Br- utilization. This work explored and illustrated the yields, influence factors, reaction mechanisms, and toxicity of Br-HNMs formed from Br- containing ASP water during UV/chlorine disinfection, which might help to control Br-HNMs formation.

Selective degradation of organic micropollutants by activation of peroxymonosulfate by Se@NC: Role of Se doping and nonradical pathway mechanism.

In this study, Se@NC-x decorated with Se was successfully prepared via two-step calcination with zeolitic imidazole framework (ZIF) as a precursor. Mechanistic studies show that PMS would be adsorbed onto the surface of Se@NC-900 to form an active complex (Se@NC-900/PMS*), and the active Se@NC-900/PMS* could oxidize phenol by the rapid decomposition of PMS. Specifically, electrons are extracted by Se@NC-900/PMS* and then transferred to the surface of Se@NC-900, which can trigger the degradation of phenol. Notably, it is found that the local charge redistribution caused by the doping of Se can activate the catalytic potential of the intrinsically inert carbon skeleton through density flooding theory (DFT) calculations. The XLogP, ΔE, VIP, and ELUMO (Se@NC/PMS)-HOMO (pollutants) and degradation rate constants of different micropollutants were correlated well linearly. This indicates that the Se@NC-900/PMS system has a great selectivity for the degradation of pollutants. Overall, these findings not only illustrate the role of Se in tuning the electronic structure of Se@NC-x to enhance the activation of PMS, but also bridge the gap in our knowledge about the physicochemical properties and degradation performance of Se@NC catalysts.

Formation, toxicity, and mechanisms of halonitromethanes from poly(diallyl dimethyl ammonium chloride) during UV/monochloramine disinfection in the absence and presence of bromide ion.

Bromide ion (Br-) is known as a prevalent component in water environments, which exhibits significant impacts on halonitromethanes (HNMs) formation. This study was performed to explore and compare the formation, toxicity, and mechanisms of HNMs from poly(diallyl dimethyl ammonium chloride) (PDDACl) in the absence and presence of Br- in the UV/monochloramine (UV/NH2Cl) disinfection process. The results showed that chlorinated HNMs were found in the absence of Br-, while brominated (chlorinated) HNMs and brominated HNMs were found in the presence of Br-. Furthermore, the peaks of total HNMs were promoted by 2.0 and 2.4 times, respectively when 1.0 and 2.0 mg L-1 Br- were added. Also, the peaks of total HNMs were enhanced with the increase of the NH2Cl dosage, which were reduced with the increase of pH. It should be noted that Br- induced higher toxicity of HNMs, and the cytotoxicity and genotoxicity of HNMs with the addition of 2.0 mg L-1 Br- were 78.0 and 3.7 times those without the addition of Br-, respectively. Meanwhile, both the reaction mechanisms of HNMs produced from PDDACl were speculated in the absence and presence of Br-. Finally, different HNMs species and yields were discovered in these two real water samples compared to those in simulated waters. These findings of this work will be conducive to understanding the significance of Br- affecting HNMs formation and toxicity in the disinfection process.

Analysis of degradation kinetic modeling and mechanism of chlorinated-halonitromethanes under UV/monochloramine treatment.

Chlorinated-halonitromethanes (Cl-HNMs) including chloronitromethane (CNM), dichloronitromethane (DCNM), and trichloronitromethane (TCNM) are nitrogenous disinfection by-products, which have high cytotoxicity and genotoxicity to human. This study aimed to investigate the degradation kinetic modeling and mechanism of Cl-HNMs under monochloramine activated by ultraviolet of 254 nm (UV/NH2Cl) treatment. The first-principle kinetic model of UV/NH2Cl process was developed to simulate Cl-HNMs degradation. Of note, the second-order rate constants of Cl-HNMs reacting with HO• (∼108 M-1 s-1), Cl• (kCl•,CNM or DCNM = âˆ¼1010 M-1 s-1, kCl•,TCNM = âˆ¼102 M-1 s-1), Cl2•- (kCl•,CNM or DCNM = âˆ¼109 M-1 s-1, kCl•,TCNM = âˆ¼101 M-1 s-1), ClO• (∼105-106 M-1 s-1) and CO3•- (∼106-107 M-1 s-1) were obtained by the first-principle kinetic model. Overall, Cl-HNMs degradation under UV/NH2Cl treatment was successfully predicted by the kinetic model under various conditions. It was found that UV (>60%) was dominant in Cl-HNMs degradation, followed by HO• (3.8%-24.5%), reactive chlorine species (RCS, 0.9%-28.8%) and CO3•- (0-26.1%). Among the contributions of RCS, Cl• and Cl2•- were main radicals in the degradation of CNM and DCNM, while ClO• was responsible for the abatement of TCNM. The minimum EE/O values under UV/NH2Cl treatment were approximately 30% lower than those under UV treatment. Finally, the possible degradation pathways were proposed, including hemolytic/heterolytic cleavage of Cl-HNMs by UV irradiation, hydrogen abstraction/electron transfer of CNM and DCNM and adduct reaction of TCNM by free radicals. This study based on the kinetic model is beneficial to predict and control the concentrations of Cl-HNMs under UV/NH2Cl treatment.

An improved graph convolutional network with feature and temporal attention for multivariate water quality prediction.

The analysis and prediction of water quality are of great significance to water quality management and pollution control. In general, current water quality prediction methods are often aimed at single indicator, while the prediction effect is not ideal for multivariate water quality data. At the same time, there may be some correlations between multiple indicators which the conventional prediction models cannot capture. To resolve these problems, this paper proposes a deep learning model: Graph Convolutional Network with Feature and Temporal Attention (FTGCN), realizing the prediction for multivariable water quality data. Firstly, a feature attention mechanism based on multi-head self-attention is designed to capture the potential correlations between water indicators. Then, a temporal prediction module including temporal convolution and bidirectional GRU with a temporal attention mechanism is designed to deal with temporal dependencies of time series. Moreover, an adaptive graph learning mechanism is introduced to extract hidden associations between water quality indicators. An auto-regression module is also added to solve the disadvantage of non-linear nature of neural networks. Finally, an evolutionary algorithm is adopted to optimize the parameters of the proposed model. Our model is applied on four real-world water quality datasets, compared with other models for multivariate time series forecasting. Experimental results demonstrate that the proposed model has a better performance in water quality prediction than others by two indices.

Characteristics of leachate from refuse transfer stations in rural China.

The properties of leachate from refuse transfer stations (RTSs) in rural China were indefinite. In this study, a total of 14 leachate samples from RTSs in nine provinces of China were characterized for their pH, electric conductivity, chromaticity, concentration of organic substances, nitrogen distribution, volatile organic compounds (VOCs), organic phosphorous pesticide, and heavy metals. The structural composition of fluorescent dissolved organic matter (FDOM) was also determined. To evaluate the leachate pollution potential in this study, a leachate pollution index was derived and used. Chromium (Cr) was the most polluting heavy metal present in rural leachate. Ethanol and ethyl acetate were the most frequently detected VOCs at high concentrations. Three-dimensional fluorescence excitation-emission matrix spectra were used to characterize the FDOM. Three components, tryptophan (C1), tyrosine-like (C2), and humic acid- and fulvic acid-like (C3) substances, were identified from all 14 samples. Tryptophan was the major component of FDOM and present in 45.7% of the samples by calculating the fluorescence intensity percentage, on average. Pearson correlations revealed that the fluorescence intensity of C1 and C3 was strongly related to soluble chemical oxygen demand and dissolved oxygen carbon, while C2 had significant positive correlations with ammonia nitrogen and total phosphorus of the solid waste. This study provided detailed data and findings that could serve as a preliminary basis for broadening options for the treatment and management of leachate from rural RTSs in China.

The key role of crystal boron in enhanced degradation of refractory contaminants using heterogeneous Fe3+/SPC system.

An origin Fenton-like system was discussed for the abatement of refractory contaminants. Sodium percarbonate (SPC) was utilized as the source of H2O2 and crystal boron (C-boron) was applied to enhance the activation of H2O2. Under the conditions of 0.50 mM Fe3+, 0.34 mM SPC, and heterogeneous catalysis using 100 mg L-1 C-boron, four target pollutants, at the initial concentrations of 20 µM, could be efficiently degraded by the Fenton-like system, with a degradation rate within 20 min up to 81.1% (aspirin, ASA), 92.8% (nitrobenzene, NB), 94.7% (flunixin meglumine, FMME), and 94.3% (benzoic acid, BA) respectively and total organic carbon removal up to 25.0%. The increase of Fe2+ concentration indicated that the conversion of Fe2+/Fe3+ was remarkably promoted by C-boron. Degradation reactions at acidic pH were comparatively fast, with pH-dependent kobs of 9.9 × 10-2 min-1 (ASA), 1.5 × 10-1 min-1 (NB), 1.7 × 10-1 min-1 (FMME), and 1.9 × 10-1 min-1 (BA), whereas those at neutral and alkaline pH were slower. Furthermore, reactive oxygen species including ·OH, 1O2, and O2·- were identified by in-situ electron paramagnetic resonance tests. The contribution ratios of ·OH turned out to be about 71.3-86.7% for the decomposition of four contaminants. The elimination of natural organic matter and the performance of material recycling highlighted the potential for its application in water treatment. The inhibition rate of Chlorella pyrenoidosa reached 211.9% in the C-boron/Fe3+/SPC system. The relatively high algae toxicity limited its application scope, which requires additional research to resolve.

Effects of bromide ion on the formation and toxicity alteration of halonitromethanes from nitrate containing humic acid water during UV/chlor(am)ine disinfection.

UV/chlor(am)ine are efficient for achieving multiple-barrier disinfection and maintaining residuals, while bromide (Br-) has notable impacts on the formation and toxicity of halonitromethanes (HNMs) during UV/chlor(am)ine disinfection. This study investigated the effects of Br- on HNMs formation and toxicity alteration during UV/chlor(am)ine disinfection of nitrate containing humic acid (HA) water. Results reveal that the maximum concentration of HNMs during UV/chlorine disinfection was 12.03 µg L-1 with 0.2 mg L-1 Br-, which was 22.5% higher than that without Br-, and the predominant species of HNMs were converted from trichloronitromethane (TCNM) to dibromonitromethane (DBNM) and tribromonitromethane (TBNM). However, the maximum concentration of HNMs during UV/chloramine disinfection was 3.69 µg L-1 with 0.2 mg L-1 Br-, which was increased by 26.0% than that without Br-, and the predominant species of HNMs were converted from dichloronitromethane (DCNM) to bromochloronitromethane (BCNM) and DBNM. Notably, the HNMs species and yields during UV/chloramine disinfection were less than those during UV/chlorine disinfection, primarily due to the higher concentrations of HO· and reactive chlorine/bromine species in UV/chlorine. Also, in the ranges of the Br-:Cl2 molar ratio from 0 to 0.32 and pH from 6.0 to 8.0, the Br-:Cl2 molar ratio of 0.16 and acidic pH contributed to the HNMs formation during UV/chlorine disinfection, and a high Br-:Cl2 molar ratio and neutral pH contributed to the HNMs formation during UV/chloramine disinfection. Note that the incorporation of Br- significantly improved the calculated cytotoxicity (CTI) and genotoxicity (GTI) of HNMs formed, and the calculated CTI and GTI of HNMs formed during UV/chloramine disinfection were 28.19 and 48.90% of those during UV/chlorine disinfection. Based on the diversity of nitrogen sources, the possible formation pathways of HNMs from nitrate containing HA water were proposed during UV/chlor(am)ine disinfection in the presence of Br-. Results of this study indicated that UV/chloramine can reduce the formation and toxicity of HNMs efficiently.

Biochar-seeded struvite precipitation for simultaneous nutrient recovery and chemical oxygen demand removal in leachate: From laboratory to pilot scale.

Refuse transfer station (RTS) leachate treatment call for efficient methods to increase nutrient recovery (NH4 +-N and PO4 3--P) and chemical oxygen demand (COD) removal. In this study, the effects of various operational factors (seeding dose, pH, initial NH4 +-N concentration, and reaction time) on biochar-seeded struvite precipitation were investigated at laboratory and pilot scales. Mealworm frass biochar (MFB) and corn stover biochar (CSB) were used as seeding materials to compare with traditional seed struvite. The maximum NH4 +-N and PO4 3--P recover efficiency of the MFB-seeded process reached 85.4 and 97.5%, higher than non-seeded (78.5 and 88.0%) and CSB-seeded (80.5 and 92.0%) processes and close to the struvite-seeded (84.5 and 95.1%) process. The MFB-seeded process also exhibited higher COD removal capacity (46.4%) compared to CSB-seeded (35.9%) and struvite-seeded (31.2%) processes and increased the average particle size of the struvite product from 33.7 to 70.2 µm for better sustained release. XRD, FT-IR, and SEM confirmed the orthorhombic crystal structure with organic matter attached to the struvite product. A pilot-scale test was further carried out in a custom-designed stirred tank reactor (20 L). In the pilot-scale test, the MFB-seeded process still spectacularly recovered 77.9% of NH4 +-N and 96.1% of PO4 3--P with 42.1% COD removal, which was slightly lower than the laboratory test due to insufficient and uniform agitation. On the whole, MFB-seeded struvite precipitation is considered to be a promising pretreatment method for rural RTS leachate.

A selective kernel-based cycle-consistent generative adversarial network for unpaired low-dose CT denoising.

Low-dose computed tomography (LDCT) denoising is an indispensable procedure in the medical imaging field, which not only improves image quality, but can mitigate the potential hazard to patients caused by routine doses. Despite the improvement in performance of the cycle-consistent generative adversarial network (CycleGAN) due to the well-paired CT images shortage, there is still a need to further reduce image noise while retaining detailed features. Inspired by the residual encoder-decoder convolutional neural network (RED-CNN) and U-Net, we propose a novel unsupervised model using CycleGAN for LDCT imaging, which injects a two-sided network into selective kernel networks (SK-NET) to adaptively select features, and uses the patchGAN discriminator to generate CT images with more detail maintenance, aided by added perceptual loss. Based on patch-based training, the experimental results demonstrated that the proposed SKFCycleGAN outperforms competing methods in both a clinical dataset and the Mayo dataset. The main advantages of our method lie in noise suppression and edge preservation.

Formation and transformation of halonitromethanes from dimethylamine in the presence of bromide during the UV/chlorine disinfection.

Halonitromethanes (HNMs) is a typical class of nitrogenous disinfection byproducts with high toxicity. The effect of Br- on the formation and transformation of HNMs from dimethylamine (DMA) during the ultraviolet (UV)/chlorine disinfection has been investigated in current study. Results reveal that only chloronitromethane, dichloronitromethane and trichloronitromethane (TCNM) could be found during the UV/chlorine disinfection. Whereas in the presence of Br-, nine species of HNMs could be observed simultaneously. When Br- concentration increased from 0 to 15.0 mg L-1, the predominant species of HNMs were gradually changed from TCNM to dibromonitromethane and tribromonitromethane, which contributed to 23.37% and 31.07% of total HNMs concentration at 15 mg L-1 Br-, respectively. The presence of Br- not only shifted the chlorinated-HNMs (Cl-HNMs) towards brominated-HNMs (Br-HNMs) but also affected the dominant species and total concentration of HNMs. When Br- concentration was 4.0 mg L-1, the formation of HNMs decreased with the increase of pH from 6.0 to 8.0 and increased with the increase of free chlorine and DMA. When free chlorine concentration rose from 0.25 to 1.1 mmol L-1, Br-HNMs were shifted to Br(Cl)-HNMs and then to Cl-HNMs. According to the findings, possible formation and transformation pathways of HNMs from DMA were proposed in the presence of Br- during the UV/chlorine disinfection. Finally, it was proved that the effect of Br- on the trend of HNMs in real water was similar to that in deionized water, but higher HNMs concentrations and delayed peak time were observed in real water. This study can provide the scientific evidence and fundamental data for the applications of UV/chlorine disinfection in the treatment of water containing Br-.

Feasibility of micropollutants removal by solar-activated persulfate: Reactive oxygen species formation and influence on DBPs.

As a natural source of visible light and a type of renewable energy, solar energy is extensively used in the field of photochemistry. In this study, solar was employed to activate persulfate (PS) to degrade typical micropollutants. The removal kinetics of aspirin (ASA) and flunixin meglumine (FMME) in the solar/PS system were well fitted by pseudo-first-order models (R2 > 0.99). In the system containing 1.0 mM PS activated by solar irradiation at a fluence of 1.14 × 10-4 E·m-2·s-1, 72.6% and 97.5% of ASA and FMME were degraded, and the corresponding kinetic constants were 6.8-9.8 × 10-2 and 1.6-9.8 × 10-1 min-1, respectively. Qualitative and quantitative analyses of the reactive oxygen species (ROS) indicated that sulfate radical (SO4·-) played a major role in degradation, with the maximum contributions of 77.7% and 88.8% for the degradation of ASA and FMME, whereas the maximum contributions of hydroxyl radical (·OH) were only 11.6% and 6.5%, respectively. The contributions of singlet oxygen (1O2) were less than 15% at pH 5.5, but increased to 25.6% and 45.5% at pH 8.5, respectively. Solar/PS pre-oxidation increased disinfection byproducts (DBPs) (95.8% for trihalomethanes (THMs) and 47.9% for haloacetic acids (HAAs) at pH 7.0) after chlorination in deionized water, and an opposite trend was found in systems coexisting with natural organic matter (NOM). Residual PS after oxidation resulted in a high aquatic toxicity, with an inhibition rate of 18.70% to algae growth. Economic analysis showed that the electrical energy per order values of the system ranged from 23.5 to 86.5 kWh·m-3·order-1, indicating that the solar/PS system shows promise for practical applications.

Kinetic removal of acetaminophen and phenacetin during LED-UV365 photolysis of persulfate system: Reactive oxygen species generation.

Acetaminophen (ACT) and phenacetin (PNT) removal during light-emitting diode (LED)-UV photolysis of persulfate (PS) was evaluated with a typical wavelength of 365 nm. Decay of PNT and ACT in pH ranges of 5.5-8.5 followed pseudo-first order kinetics. Maximum pseudo-first order rate constants (kobs) of ACT and PNT decomposition of 1.8 × 10-1 and 1.2 × 10-1 min-1, respectively, were obtained at pH 8.5. Hydroxyl radicals (·OH), sulfate radicals (SO4·-), superoxide radicals (O2-·), and singlet oxygen (1O2) were determined in-situ electron paramagnetic resonance (EPR) and alcohol scavenging tests. The average contributions of ·OH and SO4·- were 23.5% and 53.0% for PNT removal, and 15.9% and 53.0% for ACT removal at pH ranges of 5.5-8.5. In samples subjected to chlorination after LED-UV365/PS pre-oxidation, a relatively small total concentration of five halogenated disinfection by-products (DBPs) was obtained of 90.9 µg L-1 (pH 5.5) and 126.7 µg L-1 (pH 7.0), which is 58.5% and 30.2% lower than that in system without LED-UV365/PS pre-oxidation. Meanwhile, a higher maximum value of total DBP concentration was obtained at pH 8.5 (445.6 µg L-1) following LED-UV365/PS pre-oxidation. The results of economy evaluation showed that UV365 was more cost-effective in application for organic contaminant removal compared with UV254.

Cobalt doped iron oxychloride as efficient heterogeneous Fenton catalyst for degradation of paracetamol and phenacetin.

Cobalt doped iron oxychloride (Co-FeOCl) was synthesized and employed as catalyst in Fenton degradation of paracetamol (APAP) and phenacetin (PNCT) for the first time. The catalytic performance was evaluated by means of various parameters including catalyst load, hydrogen peroxide (H2O2) dose and pH value. The high removal of APAP (87.5%) and PNCT (76.0%) was obtained under conditions of 0.2 g/L Co-FeOCl and 0.5 mM H2O2 at pH 7.0, with calculated pseudo-first order kinetic constants of 0.031 min-1 for APAP and 0.023 min-1 for PNCT. Particularly, quenching tests and in situ electron spin resonance (ESR) tests were employed for the identification of the reactive oxygen species (ROS) in system. Hydroxyl radical (·OH) and superoxide radical (O2-·) were the primary ROS in Co-FeOCl/H2O2 system. A possible mechanism for H2O2 activation by Co-FeOCl catalyst was proposed as well. Finally, the formation of typical disinfection by-products (DBPs) decreased slightly in Co-FeOCl/H2O2 pre-oxidation. However, stability and reusability of Co-FeOCl were deactivated in the consecutive three cycles.