Search

Recently, the unbalanced water demand and supply have been occurred globally due to the climate change, industrialization, and population increase. The water shortage problem became much severe in the developing countries because the low-stage of infrastructure might result in frequent limitation of water and transportation of water. Therefore, sustainable water supplying option is the key for the boosting the local economy. The Mekong Delta area produce most of the products in agriculture and fishery in Vietnam. However, due to low income and vast area, the infrastructure for water supply has been limited in the Mekong Delta area. Moreover, due to reduced Mekong-river flow by the dam construction in upper reaches, increased seawater level by the global warming, and increased unexpected drought and flooding by the climate change, the water supply in the Mekong delta is becoming hard. Therefore, sustainable water supplying option was required. In this study, the reverse osmosis-based desalination process could use undrinkable saline water as the feed water coupled with photovoltaic panel was suggested. In order to install the decentralized and off-grid type water production plant in the proper place in the Mekong Delta, the KIST, VKIST, and central and local governments in Vietnam collaborated. The RO plant could be successfully installed by the help of the partners and the operational conditions have been monitored. If the water providing option could be separated according to the purposes such as drinking usage and other water usage, the more technical options could be applied for the water production.

Training deep learning models on medical images heavily depends on experts’ expensive and laborious manual labels. In addition, these images, labels, and even models themselves are not widely publicly accessible and suffer from various kinds of bias and imbalances. In this paper, chest X-ray pre-trained model via self-supervised contrastive learning (CheSS) was proposed to learn models with various representations in chest radiographs (CXRs). Our contribution is a publicly accessible pretrained model trained with a 4.8-M CXR dataset using self-supervised learning with a contrastive learning and its validation with various kinds of downstream tasks including classification on the 6-class diseases in internal dataset, diseases classification in CheXpert, bone suppression, and nodule generation. When compared to a scratch model, on the 6-class classification test dataset, we achieved 28.5% increase in accuracy. On the CheXpert dataset, we achieved 1.3% increase in mean area under the receiver operating characteristic curve on the full dataset and 11.4% increase only using 1% data in stress test manner. On bone suppression with perceptual loss, we achieved improvement in peak signal to noise ratio from 34.99 to 37.77, structural similarity index measure from 0.976 to 0.977, and root-square-mean error from 4.410 to 3.301 when compared to ImageNet pretrained model. Finally, on nodule generation, we achieved improvement in Fréchet inception distance from 24.06 to 17.07. Our study showed the decent transferability of CheSS weights. CheSS weights can help researchers overcome data imbalance, data shortage, and inaccessibility of medical image datasets. CheSS weight is available at https://github.com/mi2rl/CheSS.

Water scarcity is a severe issue for humans owing to global climate change. Water reuse via reverse osmosis (RO) has been widely used to stabilize water supplies; however, RO membrane fouling increases operational costs and necessitates intermittent operation. The biofouling study on the RO pilot was operated either intermittently or continuously using collected brackish water. The environmental microbiome was incubated under oxic and anoxic conditions to simulate surface and groundwater conditions as the feed water. The least fouling was observed on the RO membrane surface when the RO system was operated intermittently using feed water incubated under anoxic conditions. The microbial results showed that specific biofilm communities were formed on RO membrane surface after brackish water RO (BWRO) operation. The major biofilm-forming bacteria distinctively differed with influent water conditions (i.e., oxic vs. anoxic), whereas they were less different with operational strategies (i.e., continuous vs. intermittent). Intermittent operation could compensate for physical cleaning. However, chemical cleaning showed the most effective results for microbe removal on the membrane surface. Therefore, intermittent operation using anoxic feed water can mitigate fouling formation on RO membranes.

Despite its environmental significance, little is known about denitrification in vadose zones owing to the complexity of such environments. Here, we investigated denitrification in unsaturated soils with different pore distributions. To this end, we performed batch-type denitrification experiments and analyzed microbial community shifts before and after possible reactions with nitrates to clarify the relevant denitrifying mechanism in the microcosms. For quantitative comparison, pore distribution in the test soil samples was characterized based on the uniformity coefficient (C

Purposefully-designed, science content courses have the potential to help prepare future elementary teachers by helping to develop their understanding of the practices of science. We extend this area of research by investigating how students’ experiences in such a course contributes to their understanding of the coherence of science practices within a science investigation. We investigated U.S. college students’ understanding of coherent science inquiry investigations after completing an inquiry-based astronomy course designed for preservice elementary teachers. We assessed preservice teachers' (N=63) understanding of the coherence between question, data gathering, and evidence-based explanations using a novel format: student-generated astronomy-based children’s storybooks. Most students (59%) wrote storybooks featuring coherent investigations; in other words, their stories featured characters who linked an investigation question to data collection and to an evidence-based explanation. Over the three years of data collection, the percentage of preservice teachers who wrote coherent investigations in their final storybooks increased from 35% to 71% suggesting that additional scaffolding provided by the faculty in years 2 and 3 helped students understand these practices. Our findings suggest that purposefully-designed, science content courses can help preservice teachers learn about coherent science inquiry in astronomy. We also suggest that projects tied to the students’ own future careers, such as creating children’s science storybooks, can be used as assessment tools by faculty to assess preservice teachers’ development of science practices.

Microplastics (MPs) provide habitats to microorganisms in aquatic environments; distinct microbial niches have recently been elucidated. However, there is little known about the microbial communities on MPs under urban riverine conditions, in which environmental factors fluctuate. Therefore, this study investigated MP biofilm communities under various urban riverine conditions (i.e., organic content, salinity, and dissolved oxygen (DO) concentration) and evaluated the prioritized factors affecting plastisphere communities. Nine biofilm-forming reactors were operated under various environmental conditions. Under all testing conditions, biofilms grew on MPs with decreasing bacterial diversity. Interestingly, biofilm morphology and bacterial populations were driven by the environmental parameters. We found that plastisphere community structures were grouped according to the environmental conditions; organic content in the water was the most significant factor determining MP biofilm communities, followed by salinity and DO concentration. The principal plastisphere communities were Proteobacteria, Bacteroidetes, Cyanobacteria, and Firmicutes phyla. In-depth analyses of plastisphere communities revealed that biofilm-forming and plastic-degrading bacteria were the predominant microbes. In addition, potential pathogens were majorly discovered in the riverine waters with high organic content. Our results suggest that distinct plastisphere communities coexist with MP particles under certain riverine water conditions, implying that the varied MP biofilm communities may affect urban riverine ecology in a variety of ways.

With the recent development of deep learning, the classification and segmentation tasks of computer-aided diagnosis (CAD) using non-contrast head computed tomography (NCCT) for intracranial hemorrhage (ICH) has become popular in emergency medical care. However, a few challenges remain, such as the difficulty of training due to the heterogeneity of ICH, the requirement for high performance in both sensitivity and specificity, patient-level predictions demanding excessive costs, and vulnerability to real-world external data. In this study, we proposed a supervised multi-task aiding representation transfer learning network (SMART-Net) for ICH to overcome these challenges. The proposed framework consists of upstream and downstream components. In the upstream, a weight-shared encoder of the model is trained as a robust feature extractor that captures global features by performing slice-level multi-pretext tasks (classification, segmentation, and reconstruction). Adding a consistency loss to regularize discrepancies between classification and segmentation heads has significantly improved representation and transferability. In the downstream, the transfer learning was conducted with a pre-trained encoder and 3D operator (classifier or segmenter) for volume-level tasks. Excessive ablation studies were conducted and the SMART-Net was developed with optimal multi-pretext task combinations and a 3D operator. Experimental results based on four test sets (one internal and two external test sets that reflect a natural incidence of ICH, and one public test set with a relatively small amount of ICH cases) indicate that SMART-Net has better robustness and performance in terms of volume-level ICH classification and segmentation over previous methods. All code is available at https://github.com/babbu3682/SMART-Net.

The effects of C/N ratio in mainstream partial nitritation (PN)-anaerobic ammonia oxidation (ANAMMOX) considering competitive relationship of aerobic microbes competing for oxygen were investigated. Thy system was operated for 501 d with various C/N ratio. Competitive growth of aerobic heterotrophic bacteria (AHB) at ≥ 1 of C/N ratio acted effectively on the selective inhibition of nitrite-oxidizing bacteria (NOB) while contributing to stable PN-A. In-depth kinetic analysis indicated oxygen affinity of aerobic microbes was in the order of AHB ammonia-oxidizing bacteria (AOB) NOB. In addition, potential of denitritation by AHB could contributed to improving nitrogen removal up to 87.5 ± 4.3%. AHB was comparatively clustered into two groups with a C/N ratio of 1. Nitrosomonas sp. PY1 became predominant while Nitrospira spp. were the major NOB. The potential of AHB in establishing selective inhibition of NOB was identified, which could be a novel approach to stabilze the mainstream PN-A.

Bone suppression images (BSIs) of chest radiographs (CXRs) have been proven to improve diagnosis of pulmonary diseases. To acquire BSIs, dual-energy subtraction (DES) or a deep-learning-based model trained with DES-based BSIs have been used. However, neither technique could be applied to pediatric patients owing to the harmful effects of DES. In this study, we developed a novel method for bone suppression in pediatric CXRs.First, a model using digitally reconstructed radiographs (DRRs) of adults, which were used to generate pseudo-CXRs from computed tomography images, was developed by training a 2-channel contrastive-unpaired-image-translation network. Second, this model was applied to 129 pediatric DRRs to generate the paired training data of pseudo-pediatric CXRs. Finally, by training a U-Net with these paired data, a bone suppression model for pediatric CXRs was developed.The evaluation metrics were peak signal to noise ratio, root mean absolute error and structural similarity index measure at soft-tissue and bone region of the lung. In addition, an expert radiologist scored the effectiveness of BSIs on a scale of 1-5. The obtained result of 3.31 ± 0.48 indicates that the BSIs show homogeneous bone removal despite subtle residual bone shadow.Our method shows that the pixel intensity at soft-tissue regions was preserved, and bones were well subtracted; this can be useful for detecting early pulmonary disease in pediatric CXRs.

A partial nitritation (PN)/anaerobic ammonium oxidation (ANAMMOX) process in sequencing batch reactor (SBR) was successfully developed to treat high-strength ammonium wastewater. The feed distribution in the SBR cycle and sub-cycles was considered as the main operating strategy, and was optimized using a response surface methodology (RSM)-based optimization technique. In the SBR cycle, the maximum nitrogen removal rate (NRR) of 0.79 ± 0.01 kg m−3 d−1 was achieved by applying a feed distribution strategy that considered the kinetic characteristics of ANAMMOX and ammonia oxidizing bacteria (AOB). However, this strategy negatively affected the nitrogen removal efficiency (NRE) due to alkalinity loss. Therefore, the feed distribution in the SBR sub-cycles with respect to the NRE and the NRR was further studied. The nitrogen removal performance was optimized in the optimum region and an NRE of 88% and an NRR of 0.84 kg m−3 d−1 were achieved. The optimized model was verified in confirmation test. The RSM-based optimization results provide insights into the feed distribution strategy for achieving single-stage PN/ANAMMOX SBR operation.

Foxp3＋ regulatory CD4＋ T (Treg) cells play an essential role in preventing overt immune responses against self and innocuous foreign antigens. Selective expansion of endogenous Treg cells in response to the administration of interleukin (IL)-2/antibody complex, such as the IL-2/JES6-1 complex (IL-2C) in mice, is considered an attractive therapeutic approach to various immune disorders. Here, we investigated the therapeutic potential of IL-2C in allergic airway inflammation models. IL-2C treatment ameliorated Th17-mediated airway inflammation; however, unexpectedly, IL-2C treatment exacerbated Th2-mediated allergic airway inflammation by inducing the selective expansion of Th2 cells and type-2 innate lymphoid cells. We also found that IL-2 signaling is required for the expansion of Th2 cells in lymphoproliferative disease caused by Treg cell depletion. Our data suggest that IL-2C is selectively applicable to the treatment of allergic airway diseases depending on the characteristics of airway inflammation. [BMB Reports 2019; 52(4): 283-288].

In this study, the stability of the total nitrogen removal efficiency (TNRE) was modeled using an artificial neural network (ANN)-based binary classification model for the anaerobic ammonium oxidation (AMX) process under saline conditions. The TNRE was stabilized to 80.2 ± 11.4% at the final phase under the salinity of 1.0 ± 0.02%. The results of terminal restriction fragment length polymorphism (T-RFLP) analysis showed the predominance of Candidatus Jettenia genus. Real-time quantitative PCR analysis revealed the average abundance of Ca. Jettenia and Kuenenia spp. increased in 3.2 ± 5.4 × 10

The protocol for placing anatomical side markers (L/R markers) in chest radiographs varies from one hospital or department to another. However, the markers have strong signals that can be useful for deep learning-based classifier to predict diseases. We aimed to enhance the performance of a deep learning-based classifiers in multi-center datasets by inpainting the L/R markers.The L/R marker was detected with using the EfficientDet detection network; only the detected regions were inpainted using a generative adversarial network (GAN). To analyze the effect of the inpainting in detail, deep learning-based classifiers were trained using original images, marker-inpainted images, and original images clipped using the min-max value of the marker-inpainted images. Binary classification, multi-class classification, and multi-task learning with segmentation and classification were developed and evaluated. Furthermore, the performances of the network on internal and external validation datasets were compared using DeLong's test for two correlated receiver operating characteristic (ROC) curves in binary classification and Stuart-Maxwell test for marginal homogeneity in multi-class classification and multi-task learning. In addition, the qualitative results of activation maps were evaluated using the gradient-class activation map (Grad-CAM).Marker-inpainting preprocessing improved the classification performances. In the binary classification based on the internal validation, the area under the curves (AUCs) and accuracies were 0.950 and 0.900 for the model trained on the min-max clipped images and 0.911 and 0.850 for the model trained on the original images, respectively (P-value=0.006). In the external validation, the AUCs and accuracies were 0.858 and 0.677 for the model trained using the inpainted images and 0.723 and 0.568 for the model trained using the original images (P-value0.001), respectively. In addition, the models trained using the marker inpainted images showed the best performance in multi-class classification and multi-task learning. Furthermore, the activation maps obtained using the Grad-CAM improved with the proposed method. The 5-fold validation results also showed improvement trend according to the preprocessing strategies.Inpainting an L/R marker significantly enhanced the classifier's performance and robustness, especially in internal and external studies, which could be useful in developing a more robust and accurate deep learning-based classifier for multi-center trials. The code for detection is available at: https://github.com/mi2rl/MI2RLNet. And the code for inpainting is available at: https://github.com/mi2rl/L-R-marker-inpainting.

This paper describes the new concept of the mainstream partial nitritation (PN)-anaerobic ammonium oxidation (ANAMMOX) combined with a high-temperature shock strategy for the selective recovery of ammonia-oxidizing bacteria (AOB) activity. In the preliminary test, the temperature shock condition for PN was optimized (60 °C and > 20 min). Based on this, the implementation strategy in a continuous stirred tank reactor (CSTR) system was studied further, and the polyvinyl alcohol (PVA)/sodium alginate carrier exposure ratio (ER) and dissolved oxygen (DO) concentration were considered as primary variables. The AOB activity was recovered selectively when the ER of the carrier ranged from 20 to 40%, and the DO was higher than 2.3 mg O2/L. This was not the case for nitrite-oxidizing bacteria (NOB) (AOB: 1.17±0.1 gNH4+-N/LCarrier/d, NOB: 0.34±0.1 gNO3--N/LCarrier/d). As a result, the activity of AOB was recovered selectively with a decrease in Nitrospira spp., which was verified by kinetic and microbial analyses for the AOB (KS, DO = 3.89 mgO2/L) and NOB (KS, DO = 1.14 mgO2/L). Eventually, the mainstream PN-ANAMMOX was achieved with a nitrogen removal efficiency of 81.5±3.3% for 95 days. The findings provide insight to establishing a stable mainstream PN-ANAMMOX process using a high-temperature shock strategy.

Although murine γδ T cells are largely considered innate immune cells, they have recently been reported to form long-lived memory populations. Much remains unknown about the biology and specificity of memory γδ T cells. Here, we interrogated intestinal memory Vγ4 Vδ1 T cells generated after foodborne Listeria monocytogenes (Lm) infection to uncover an unanticipated complexity in the specificity of these cells. Deep TCR sequencing revealed that a subset of non-canonical Vδ1 clones are selected by Lm infection, consistent with antigen-specific clonal expansion. Ex vivo stimulations and in vivo heterologous challenge infections with diverse pathogenic bacteria revealed that Lm-elicited memory Vγ4 Vδ1 T cells are broadly reactive. The Vγ4 Vδ1 T cell recall response to Lm, Salmonella enterica serovar Typhimurium (STm) and Citrobacter rodentium was largely mediated by the γδTCR as internalizing the γδTCR prevented T cell expansion. Both broadly-reactive canonical and pathogen-selected non-canonical Vδ1 clones contributed to memory responses to Lm and STm. Interestingly, some non-canonical γδ T cell clones selected by Lm infection also responded after STm infection, suggesting some level of cross-reactivity. These findings underscore the promiscuous nature of memory γδ T cells and suggest that pathogen-elicited memory γδ T cells are potential targets for broad-spectrum anti-infective vaccines.

Partial nitritation (PN) is a bioprocess that is essential for developing cost-effective biological nitrogen removal processes. Understanding the abundant bacterial communities responsible for nitrification under salt stress conditions is important to achieve a stable PN system for treating saline wastewater. Therefore, in this study, we identified the core nitrifying communities and investigated their correlations with the process parameters in a nitrifying bioreactor that was used for treating saline high-strength ammonia wastewater. A PN system worked efficiently under saline conditions with varying operational factors, such as temperature, dissolved oxygen (DO), and alkalinity. Interestingly, the specific oxygen uptake rate (SOUR) became similar under salt-free and saline media after the salt adaption. Next generation sequencing results suggested that the inactivation of Nitrobacter winogradskyi was a key factor for the PN reaction under salt stress conditions. We also found that Nitrosomonas europaea, a freshwater type ammonia-oxidizing bacteria (AOB), was predominantly found under both salt-free and saline conditions, whereas other halotolerant or halophilic AOB species, including Nitrosomonas nitrosa and Nitrosomonas mobilis, became selectively abundant under saline conditions. This implies that adaptation (training of N. europaea) and selection (presence of N. nitrosa and N. mobilis) were simultaneously attributed to selective ammonia conversion for the PN reaction. The redundancy analysis showed that the salinity and ammonia loading rates were statistically significant process parameters that determined the nitrifying bacterial community, suggesting that these parameters drive the adaptation and selection of the core AOB species during the PN reaction. Furthermore, the correlation analysis revealed that the abundance of N. nitrosa and N. mobilis was critically correlated with the specific oxygen uptake rates in saline media containing ammonia.

This study evaluated the long-term organic removal performance and microbial community shift in simulated aquifer storage and recovery (ASR) conditions. For this purpose, anoxic soil box systems were operated at 15 °C for one year. The results showed that the assimilable organic carbon (AOC) concentration in the anoxic soil box systems was successfully decreased by 79.1%. The dissolved organic carbon (DOC) concentration increased during the initial operational periods; however, it subsequently decreased during long-term operation. Readily biodegradable organic fractions (i.e., low-molecular weight (LMW) neutrals and LMW acids) decreased along with time elapsed, whereas non-biodegradable fraction (i.e., humic substances) increased. Proteobacteria and Acidobacteriota were predominant in the anoxic box systems throughout the operational periods. Firmicutes and Bacteroidota suddenly increased during the initial operational period while Gemmatimonadota slightly increased during prolonged long-term operation. Interestingly, the microbial community structures were significantly shifted with respect to the operational periods while the effects of AOC/NO3− addition were negligible. Various bacterial species preferring low temperature or anoxic conditions were detected as predominant bacteria. Some denitrifying (i.e., Noviherbaspirillum denitrificans) and iron reducing bacteria (i.e., Geobacter spp.) appeared during the long-term operation; these bacterial communities also acted as organic degraders in the simulated ASR systems. The findings of this study suggest that the application of natural bioattenuation using indigenous soil microbial communities can be a promising option as an organic carbon management strategy in ASR systems.

The objective of this study was to evaluate the effects of salt accumulation on nitrifying bacterial communities in a nitrifying bioreactor combined with forward osmosis. The conversion of nitrite to nitrate was inhibited at a total dissolved solids (TDS) concentration of 17.3 g/L, whereas conversion of ammonia to nitrite was inhibited at a higher concentration (52.8 g-TDS/L). The gene copies of ammonia-oxidizing bacteria (AOB) were more abundant than those of nitrite-oxidizing bacteria (NOB) throughout the entire operating period of 225 days. Among NOB, the number of copies of Nitrobacter spp. were 100–1000 times higher than those of Nitrospira spp. A total of 140 operational taxonomic units were identified using 454 pyrosequencing. The relative abundances of autotrophic AOB and NOB accounted for 34.1–57.8% during 225 days. Dominance of Nitrosomonas eutropha was stable as a salt-tolerant AOB, but the representative NOB, Nitrobacter winogradskyi, showed salt-sensitive variations in their relative abundance. Nonmetric multidimensional scaling and hierarchical clustering analysis clearly illustrated the shift in bacterial community due to external conditions, i.e., ammonia loading rate, alkalinity availability, and salinity. Heterotrophic bacteria contributed to changes in overall bacterial community structure in the nitrifying osmotic membrane bioreactor despite the absence of carbon sources in the influent.

This study investigated the effect of seeding source on the mature anaerobic ammonia oxidation (ANAMMOX) bacterial community niche in continuous poly(vinyl alcohol) (PVA) gel systems operated under high nitrogen loading rate (NLR) condition. Four identical column reactors packed with PVA gels were operated for 182 d using different seeding sources which had distinct community structures. The ANAMMOX reaction was achieved in all the bioreactors with comparable total and ANAMMOX bacterial 16S rRNA gene quantities. The bacterial community structure of the bioreactors became similar during operation; some major bacteria were commonly found. Interestingly, one ANAMMOX species, "Candidatus Brocadia sinica", was conclusively predominant in all the bioreactors, even though different seeding sludges were used as inoculum source, possibly due to the unique physiological characteristics of "Ca. Brocadia sinica" and the operating conditions (i.e., PVA gel-based continuous system and 1.0 kg-N/(m3·d) of NLR). The results clearly suggest that high NLR condition is a more significant factor determining the final ANAMMOX community niche than is the type of seeding source.

This study investigated the single-stage partial nitritation and anammox (S-PNA) treatment of low-strength ammonia wastewater (≤140 mg NH4+-N/L). Upflow dual-bed gel-carrier reactor (UDGR) with polyvinyl alcohol cryogel biocarriers, developed in this study, was employed for the anammox biomass enrichment from conventional activated sludge and subsequent S-PNA experiments. Anammox biomass was successfully enriched from conventional activated sludge. The enriched anammox carriers were inoculated together with gel carriers containing nitrifying sludge into the S-PNA reactors. S-PNA activity developed rapidly, and the nitrogen removal efficiency and rate reached up to 90.1% (with complete ammonia removal) and 0.15 kg N/m3⋅d, respectively, under low nitrogen loading conditions (0.10–0.17 kg N/m3⋅d). The microbial community structure changed significantly while adapting to anammox and S-PNA conditions. Anammox was likely driven solely by a Candidatus Jettenia population accounting for ≤49.4% of bacterial 16S rRNA genes. The results demonstrate that the UDGR-based S-PNA is suitable for treating low-strength wastewater.