Utilising intraoperative respiratory dynamic features for developing and validating an explainable machine learning model for postoperative pulmonary complications.

BACKGROUND: Timely detection of modifiable risk factors for postoperative pulmonary complications (PPCs) could inform ventilation strategies that attenuate lung injury. We sought to develop, validate, and internally test machine learning models that use intraoperative respiratory features to predict PPCs. METHODS: We analysed perioperative data from a cohort comprising patients aged 65 yr and older at an academic medical centre from 2019 to 2023. Two linear and four nonlinear learning models were developed and compared with the current gold-standard risk assessment tool ARISCAT (Assess Respiratory Risk in Surgical Patients in Catalonia Tool). The Shapley additive explanation of artificial intelligence was utilised to interpret feature importance and interactions. RESULTS: Perioperative data were obtained from 10 284 patients who underwent 10 484 operations (mean age [range] 71 [65-98] yr; 42% female). An optimised XGBoost model that used preoperative variables and intraoperative respiratory variables had area under the receiver operating characteristic curves (AUROCs) of 0.878 (0.866-0.891) and 0.881 (0.879-0.883) in the validation and prospective cohorts, respectively. These models outperformed ARISCAT (AUROC: 0.496-0.533). The intraoperative dynamic features of respiratory dynamic system compliance, mechanical power, and driving pressure were identified as key modifiable contributors to PPCs. A simplified model based on XGBoost including 20 variables generated an AUROC of 0.864 (0.852-0.875) in an internal testing cohort. This has been developed into a web-based tool for further external validation (https://aorm.wchscu.cn/). CONCLUSIONS: These findings suggest that real-time identification of surgical patients' risk of postoperative pulmonary complications could help personalise intraoperative ventilatory strategies and reduce postoperative pulmonary complications.

Design, synthesis and antifungal activity of novel α-methylene-γ-butyrolactone derivatives containing benzothiophene moiety.

BACKGROUND: The discovery of agricultural fungicide candidates from natural products is one of the key strategies for developing environment friendly agricultural fungicides with high efficiency, high selectivity and unique modes-of-action. Based on previous work, a series of novel α-methylene-γ-butyrolactone (MBL) derivatives containing benzothiophene moiety were designed and synthesized. RESULTS: The majority of the proposed compounds displayed moderate to considerable antifungal efficacy against the tested pathogenic fungi and oomycetes, some exhibiting broad spectrum antifungal activity. Notably, compounds 2 (3-F-Ph) and 7 (4-Cl-Ph) showed excellent antifungal activity against Rhizoctonia with half maximal effective concentration (EC50) values of 0.94 and 0.99 mg L-1, respectively, comparable to the commercial fungicide tebuconazole (EC50 = 0.96 mg L-1), and also displayed significant inhibitory effects against V alsa mali with EC50 values of 2.26 and 1.67 mg L-1, respectively - better than famoxadone and carabrone. The in vivo protective and curative effects against R. solani of compound 2 were 57.2% and 53.7% at 100 mg L-1, respectively, which were equivalent to tebuconazole (51.6% and 52.4%). Further investigations found that compound 2 altered the ultrastructure of R. solani cell, significantly increased the relative conductivity of the cells, and reduced the activity of complex III in a dose-dependent manner. Molecular docking results showed that compound 2 matched well with the Qo pocket. CONCLUSION: The results revealed that MBL derivatives containing benzothiophene moiety are promising antifungal candidates and provide a new backbone structure for further optimization of novel fungicides. © 2024 Society of Chemical Industry.

Impacts of emissions and meteorological conditions in three different phases of aerosol pollution during 2013-2022 in Anhui, China.

The PM2.5 concentrations in Anhui, which links the Yangtze River Delta region, China's fastest growing economy area, with the Beijing-Tianjin-Hebei (BTH) region, China's most polluted region, are influenced not only by emissions, but also by variation of meteorological conditions. A comprehensive understanding of the relative impacts of meteorology and emissions on heavy pollution in Anhui at three phases (i.e., phase1: from 2013 to 2017; phase2: from 2018 to 2020; phase 3: from 2021 to 2022) from 2013 to 2022, which can provide suggestions for pollution prevention and control in the future. The decrease in pollutant concentrations from 2013 to 2022 is mainly attributed to the continued reduction in emissions, while the year-to-year fluctuations in pollutant concentrations are largely influenced by meteorological conditions. Although emissions are decreasing, the proportions of residential biofuel combustion and cement are increasing. In addition to the effects of prevailing northeasterly and northwesterly winds (i.e., Type1 and Type2), there is also concern about the influences of static weather and neighboring regional transport (i.e., Type5 and Type6), especially in 2016. The contribution of emissions is greater in phase 2 and phase 3, with a 17 % increase compared to phase 1. Overall, approximately 57 % of explosive growth in PM2.5 concentration during the cumulative stage (CS) can be regarded as the feedback effect of the deteriorating meteorological conditions. Therefore, statistical analyses show that limiting PM2.5 concentrations below about 73 µg m-3 would weaken the feedback effects, which in turn would avoid most of the explosive growth processes in the CS of the 60 heavy pollution processes, which can provide a reference for the government to set a target for sustained emission reduction.

Evaluation of MERRA-2 and CAMS reanalysis for black carbon aerosol in China.

Black carbon (BC) constitutes a pivotal component of atmospheric aerosols, significantly impacting regional and global radiation balance, climate, and human health. In this study, we evaluated BC data in two prominent atmospheric composition reanalysis datasets: the Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2) and the Copernicus Atmosphere Monitoring Service (CAMS), and analyzed the causes of their deviations. This assessment is based on observational data collected from 34 monitoring stations across China from 2006 to 2022. Our research reveals a significant and consistent decline in BC concentrations within China, amounting to a reduction exceeding 67.33%. However, both MERRA-2 and CAMS reanalysis data fail to capture this declining trend. The average annual decrease of BC in MERRA-2 from 2006 to 2022 is only 0.06 µg/m3 per year, while the BC concentration in CAMS even increased with an average annual value of 0.014 µg/m3 per year. In 2022, MERRA-2 had overestimated BC concentration by 20% compared to observational data, while CAMS had overestimated it by approximately 66%. In the regional BC concentration analysis, the data quality of the reanalysis data is better in the South China (RM = 0.59, RC = 0.53), followed by the North China (RM = 0.50, RC = 0.42). Reanalysis BC data in Northwest China and the Tibetan Plateau are difficult to use for practical analysis due to their big difference with observation. In a comparison of the anthropogenic BC emissions inventory used in the two atmospheric composition reanalysis datasets with the Multi-resolution Emission Inventory model for Climate and air pollution research (MEIC) emissions inventory, we found that: Despite the significant decline in China's BC emissions, MERRA-2 still relies on the 2006 emissions inventory, while CAMS utilizes emission inventories that even show an increasing trend. These factors will undoubtedly lead to greater deviations between reanalysis and observational data.

Outstanding Synergy of Sensitivity and Linear Range Enabled by Multigradient Architectures.

Flexible pressure sensors are devices that mimic the sensory capabilities of natural human skin and enable robots to perceive external stimuli. One of the main challenges is maintaining high sensitivity over a broad linear pressure range due to poor structural compressibility. Here, we report a flexible pressure sensor with an ultrahigh sensitivity of 153.3 kPa-1 and linear response over an unprecedentedly broad pressure range from 0.0005 to 1300 kPa based on interdigital-shaped, multigradient architectures, featuring modulus, conductivity, and microstructure gradients. Such multigradient architectures and interdigital-shaped configurations enable effective stress transfer and conductivity regulation, evading the pressure sensitivity-linear range trade-off dilemma. Together with high pressure resolution, high frequency response, and good reproducibility over the ultrabroad linear range, proof-of-concept applications such as acoustic wave detection, high-resolution pressure measurement, and healthcare monitoring in diverse scenarios are demonstrated.

Highly stable and strain-insensitive metal film conductors via manipulating strain distribution.

Metal film-based stretchable conductors are essential elements of flexible electronics for wearable, biomedical, and robotic applications, which require strain-insensitive high conductivity over a wide strain range and excellent cyclic stability. However, they suffer from serious electrical failure under monotonic and cyclic tensile loading at a small strain due to the uncontrolled film cracking behavior. Here, we propose a novel in-plane crack control strategy of engineering hierarchical microstructures to achieve outstanding electromechanical performance via harnessing the strain distribution in metal films. The wrinkles delay the crack initiation at undercuts which should be the most vulnerable sites during the stretching process. The surface protrusions/grooves/undercuts inhibit the crack propagation because of the effective strain redistribution. In addition, hierarchical microstructures significantly improve cyclic stability due to the strong interfacial adhesion and stable crack patterns. The metal film-based conductors exhibit ultrahigh strain-insensitive conductivity (1.7 × 107 S m-1), negligible resistance change (ΔR/R0 = 0.007) over an ultra-wide strain range (>200%), and excellent cyclic strain durability (>15 000 cycles at 100% strain). A range of metal films was explored to establish the universality of this strategy, including ductile copper and silver, as well as brittle molybdenum and high entropy alloy. We demonstrate the strain-insensitive electrical functionality of a metal film-based conductor in a flexible light-emitting diode circuit.

Superior Carrier Mobility Enabled by the Charge Channel Leads to Enhanced Thermoelectric Performance in BiCuSeO Composites.

BiCuSeO oxyselenides possess a highlighted thermoelectric performance among oxides, which originates from their intrinsically low thermal conductivity. However, intrinsic factors causing low thermal transport are also detrimental to carrier transport, leading to ultralow carrier mobility and relatively low electrical transport properties. Here, high-conductivity single-wall carbon nanotubes (SWCNTs) are adopted as the charge channels to be embedded in a BiCuSeO-based matrix, providing a transport pathway for charge carriers. The results show that carrier mobility is increased to 188 cm2 V-1 s-1 due to the SWCNTs composited, triggering an enhancement in electrical transport properties. Besides, the SWCNTs embedded in the matrix introduce abundant interfaces, suppressing phonon transport and depressing lattice thermal conductivity. With these achievements, a maximum zT of 0.84 at 818 K is realized in the composite with 0.1 wt% SWCNTs. The mechanical property of the composites is strengthened as well because of the SWCNTs. The work indicates that the SWCNTs, as the charge channels, propose an effective approach for enhancing carrier mobility in BiCuSeO-based materials, finally optimizing the thermoelectric performance as well as the mechanical property.

Structure of LARP7 Protein p65-telomerase RNA Complex in Telomerase Revealed by Cryo-EM and NMR.

La-related protein 7 (LARP7) are a family of RNA chaperones that protect the 3'-end of RNA and are components of specific ribonucleoprotein complexes (RNP). In Tetrahymena thermophila telomerase, LARP7 protein p65 together with telomerase reverse transcriptase (TERT) and telomerase RNA (TER) form the core RNP. p65 has four known domains-N-terminal domain (NTD), La motif (LaM), RNA recognition motif 1 (RRM1), and C-terminal xRRM2. To date, only the xRRM2 and LaM and their interactions with TER have been structurally characterized. Conformational dynamics leading to low resolution in cryo-EM density maps have limited our understanding of how full-length p65 specifically recognizes and remodels TER for telomerase assembly. Here, we combined focused classification of Tetrahymena telomerase cryo-EM maps with NMR spectroscopy to determine the structure of p65-TER. Three previously unknown helices are identified, one in the otherwise intrinsically disordered NTD that binds the La module, one that extends RRM1, and another preceding xRRM2, that stabilize p65-TER interactions. The extended La module (αN, LaM and RRM1) interacts with the four 3' terminal U nucleotides, while LaM and αN additionally interact with TER pseudoknot, and LaM with stem 1 and 5' end. Our results reveal the extensive p65-TER interactions that promote TER 3'-end protection, TER folding, and core RNP assembly and stabilization. The structure of full-length p65 with TER also sheds light on the biological roles of genuine La and LARP7 proteins as RNA chaperones and core RNP components.

Quantifying the contribution of local drivers to observed weakening of spring dust storm frequency over northern China (1982-2017).

Recent studies have suggested that spring dust storm (SDS) events in northern China (NC) have exhibited substantial decline over the past 30 years. However, it is unclear which local factors are most responsible for the decline in SDS events, and the contribution of each dominant factor remains to be determined. This study utilized high-density DS records and collocated homogenized surface meteorological observations from 1982 to 2017, in conjunction with land surface products, to examine the local drivers that influence the long-term variation in SDS frequency (SDSF) over the entire NC area and its three dust-source areas: northwestern China (NWC), north-central China (NCC), and northeastern China (NEC). Results indicated that the observed SDSF averaged over NC, NWC, NCC, and NEC has decreased by 144.4 %, 109.3 %, 166.4 %, and 92.2 %, respectively, during 1982-2017. The variation in SDSF is largely explained by variation in wind speed (WS), precipitation, volumetric soil moisture, and surface bareness. A multivariable linear regression model incorporating these local drivers accounted for 81.0 %, 74.0 %, and 46.9 % of the variance in SDSF in NWC, NCC, and NEC, respectively. Statistical analyses on the local drivers suggested that weakening of WS was the dominant factor in the reduction in SDSF over recent decades, contributing 76.9 %, 54.7 %, and 33.6 % of the variation in NWC, NCC, and NEC, respectively. More importantly, we revealed that the interannual variation in regional SDSF was not only controlled by local drivers, but also influenced by cross-regional transport of dust aerosols emitted from upstream source areas.

Impact of the COVID-19 pandemic on cancer healthcare utilization in southwestern China on March 2021.

BACKGROUND: Oncological care has been disrupted worldwide during the COVID-19 pandemic. We aimed to quantify the long-term impact of the pandemic on cancer care utilization and to examine how this impact varied by sociodemographic and clinical factors in southwestern China, where the Dynamic Zero-COVID Strategy was implemented. This strategy mainly included lockdowns, stringent testing, and travel restrictions to prevent the spread of COVID-19. METHOD: We identified 859,497 episodes of the utilization of cancer care from electronic medical records between January 1, 2019, and March 31, 2021, from the cancer center of a tertiary hospital serving an estimated population of 8.4 million in southwestern China. Changes in weekly utilization were evaluated via segmented Poisson regression across service categories, stratified by cancer type and sociodemographic factors. RESULTS: A sharp reduction in utilization of in-person cancer services occurred during the first week of the pandemic outbreak in January 2020, followed by a quick rebound in February 2020. Although there were few COVID-19 cases from March 2020 until this analysis, the recovery of most in-person services was slow and remained incomplete as of March 31, 2021. The exceptions were outpatient radiation and surgery, which increased and exceeded pre-pandemic levels, particularly among lung cancer patients; meanwhile, telemedicine utilization increased substantially after the onset of the pandemic. Care disruptions were most prominent for women, rural residents, uninsured, and breast cancer patients. CONCLUSIONS: As of March 2021, despite few COVID-19 cases, the COVID-19 pandemic has had a strong and continuing impact on in-person oncology care utilization in southwestern China under the Dynamic Zero-COVID Strategy. Equitable and timely access to cancer care requires adjustment in strict policies for COVID-19 prevention and control, as well as targeted remedies for the most vulnerable populations during and beyond the pandemic. Future studies should monitor the long-term effects of the COVID-19 pandemic and response strategies on cancer care and outcomes.

Description and evaluation of a newly developed emission inventory processing system (EMIPS).

Commonly available emission inventories are often incompatible with the input requirements of atmospheric chemistry models due to temporal and spatial resolution, pollutant types, etc. We present the Emission Inventory Processing System (EMIPS) version 1, an open-source and multi-scale atmospheric emission modeling framework that prepares specific emissions inputs for atmospheric chemistry models. EMIPS is a multifunctional and user-friendly system, coded in Jython and developed on the MeteoInfo software platform. It allows users to freely combine and process emission inventories to generate model-ready emissions data. The core functions of EMIPS include preprocessing, temporal allocation, spatial allocation, chemical speciation, and vertical allocation. We detail the implementation of each function in the body of this paper, and several examples are provided for illustration. The emission outputs obtained with EMIPS have been evaluated by simulating four pollutants (PM2.5, PM10, NO2, and O3) concentrations in January 2017 using the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem), and comparison of model results with the observations indicates the model can reproduce the temporal and spatial patterns of pollutants, suggesting that EMIPS is capable of supporting atmospheric chemistry modeling. We expect this work could help to improve air quality research and forecast.

Seasonal cycles and long-term trends of arctic tropospheric aerosols based on CALIPSO lidar observations.

Notable warming trends have been observed in the Arctic, with tropospheric aerosols being one of the key drivers. Here the seasonal cycles of three-dimensional (3D) distributions of aerosol extinction coefficients (AECs) and frequency of occurrences (FoOs) for different aerosol subtypes in the troposphere over the Arctic from 2007 to 2019 are characterized capitalizing on Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) Level-3 gridded aerosol profile product. Seasonal contributions of total and type-dependent aerosols through their partitioning within the planetary boundary layer (PBL) and free troposphere (FT) are also quantified utilizing the Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2) PBL height data. The results show substantial seasonal and geographical dependence in the distribution of aerosols over the Arctic. Sulfate, black carbon (BC), and organic carbon (OC) contribute most of the total AEC, with Eurasia being the largest contributor. The vertical structure of AECs and FoOs over the Arctic demonstrates that the vertical influence of aerosols is higher in eastern Siberia and North America than in northern Eurasia and its coasts. When the total aerosol optical depth (TAOD) is partitioned into the PBL and FT, results indicate that the contributions of TAOD within the FT tend to be more significant, especially in summer, with the FT contributes 64.2% and 69.2% of TAOD over the lower (i.e., 60° N-70° N) and high (i.e., north of 70° N) Arctic, respectively. Additionally, seasonal trend analyses suggest Arctic TAOD exhibits a multi-year negative trend in winter, spring, and autumn and a positive trend in summer during 2007-2019, due to an overall decrease in sulfate from weakened anthropogenic emissions and a significant increase in BC and OC from enhanced biomass burning activities. Overall, this study has potential implications for understanding the seasonal cycles and trends in Arctic aerosols.

A Curriculum Batching Strategy for Automatic ICD Coding with Deep Multi-Label Classification Models.

The International Classification of Diseases (ICD) has an important role in building applications for clinical medicine. Extremely large ICD coding label sets and imbalanced label distribution bring the problem of inconsistency between the local batch data distribution and the global training data distribution into the minibatch gradient descent (MBGD)-based training procedure for deep multi-label classification models for automatic ICD coding. The problem further leads to an overfitting issue. In order to improve the performance and generalization ability of the deep learning automatic ICD coding model, we proposed a simple and effective curriculum batching strategy in this paper for improving the MBGD-based training procedure. This strategy generates three batch sets offline through applying three predefined sampling algorithms. These batch sets satisfy a uniform data distribution, a shuffling data distribution and the original training data distribution, respectively, and the learning tasks corresponding to these batch sets range from simple to complex. Experiments show that, after replacing the original shuffling algorithm-based batching strategy with the proposed curriculum batching strategy, the performance of the three investigated deep multi-label classification models for automatic ICD coding all have dramatic improvements. At the same time, the models avoid the overfitting issue and all show better ability to learn the long-tailed label information. The performance is also better than a SOTA label set reconstruction model.

Columnar optical, microphysical and radiative properties of the 2022 Hunga Tonga volcanic ash plumes.

The Hunga Tonga-Hunga Ha'apai eruption on January 15, 2022 was one of the most explosive volcanic eruptions of the 21st century and has attracted global attention. Here we show that large numbers of the volcanic aerosols from the eruption broke through the tropopause into the lower stratosphere, forming an ash plume with an overshooting top at 25-30 km altitude. In the four days following the eruption, the ash plume moved rapidly westward for nearly 10,000 km under stable stratospheric conditions characterized by strong tropical easterlies, weak meridional winds and weak vertical motion. The intrusion of the ash plume into the stratosphere resulted in a marked increase in atmospheric aerosol loading across northern Australia, with the aerosol optical depth (AOD) observed by satellites and sun-photometers peaking at 1.5 off the coast of northeastern Australia; these effects lasted for nearly three days. The ash plume was characterized by fine-mode particles clustered at a radius of about 0.26 µm, with an observed peak volume of 0.25 µm3 µm-2. The impact of the ash plume associated with the Hunga Tonga eruption on the stratospheric AOD and radiative balance in the tropical southern hemisphere is remarkable, with an observed volcanic-induced perturbation of the regional stratospheric AOD of up to 0.6. This perturbation largely explains an instantaneous bottom (top) of the atmosphere radiative forcing of -105.0 (-65.0) W m-2 on a regional scale.

A multicenter prospective study on postoperative pulmonary complications prediction in geriatric patients with deep neural network model.

Aim: Postoperative pulmonary complications (PPCs) can increase the risk of postoperative mortality, and the geriatric population has high incidence of PPCs. Early identification of high-risk geriatric patients is of great value for clinical decision making and prognosis improvement. Existing prediction models are based purely on structured data, and they lack predictive accuracy in geriatric patients. We aimed to develop and validate a deep neural network model based on combined natural language data and structured data for improving the prediction of PPCs in geriatric patients. Methods: We consecutively enrolled patients aged ≥65 years who underwent surgery under general anesthesia at seven hospitals in China. Data from the West China Hospital of Sichuan University were used as the derivation dataset, and a deep neural network model was developed based on combined natural language data and structured data. Data from the six other hospitals were combined for external validation. Results: The derivation dataset included 12,240 geriatric patients, and 1949(15.9%) patients developed PPCs. Our deep neural network model outperformed other machine learning models with an area under the precision-recall curve (AUPRC) of 0.657(95% confidence interval [CI], 0.655-0.658) and an area under the receiver operating characteristic curve (AUROC) of 0.884(95% CI, 0.883-0.885). The external dataset included 7579 patients, and 776(10.2%) patients developed PPCs. In external validation, the AUPRC was 0.632(95%CI, 0.632-0.633) and the AUROC was 0.889(95%CI, 0.888-0.889). Conclusions: This study indicated that the deep neural network model based on combined natural language data and structured data could improve the prediction of PPCs in geriatric patients.

Topological Gradients for Metal Film-Based Strain Sensors.

Metal film-based stretchable strain sensors hold great promise for applications in various domains, which require superior sensitivity-stretchability-cyclic stability synergy. However, the sensitivity-stretchability trade-off has been a long-standing dilemma and the metal film-based strain sensors usually suffer from weak cyclic durability, both of which significantly limit their practical applications. Here, we propose an extremely facile, low-cost and spontaneous strategy that incorporates topological gradients in metal film-based strain sensors, composed of intrinsic (grain size and interface) and extrinsic (film thickness and wrinkle) microstructures. The topological gradient strain sensor exhibits an ultrawide stretchability of 100% while simultaneously maintaining a high sensitivity at an optimal topological gradient of 4.5, due to the topological gradients-induced multistage film cracking. Additionally, it possesses a decent cyclic stability for >10 000 cycles between 0 and 40% strain enabled by the gradient-mixed metal/elastomer interfaces. It can monitor the full-range human activities from subtle pulse signals to vigorous joint movements.

Structure of Tetrahymena telomerase-bound CST with polymerase α-primase.

Telomeres are the physical ends of linear chromosomes. They are composed of short repeating sequences (such as TTGGGG in the G-strand for Tetrahymena thermophila) of double-stranded DNA with a single-strand 3' overhang of the G-strand and, in humans, the six shelterin proteins: TPP1, POT1, TRF1, TRF2, RAP1 and TIN21,2. TPP1 and POT1 associate with the 3' overhang, with POT1 binding the G-strand3 and TPP1 (in complex with TIN24) recruiting telomerase via interaction with telomerase reverse transcriptase5 (TERT). The telomere DNA ends are replicated and maintained by telomerase6, for the G-strand, and subsequently DNA polymerase α-primase7,8 (PolαPrim), for the C-strand9. PolαPrim activity is stimulated by the heterotrimeric complex CTC1-STN1-TEN110-12 (CST), but the structural basis of the recruitment of PolαPrim and CST to telomere ends remains unknown. Here we report cryo-electron microscopy (cryo-EM) structures of Tetrahymena CST in the context of the telomerase holoenzyme, in both the absence and the presence of PolαPrim, and of PolαPrim alone. Tetrahymena Ctc1 binds telomerase subunit p50, a TPP1 orthologue, on a flexible Ctc1 binding motif revealed by cryo-EM and NMR spectroscopy. The PolαPrim polymerase subunit POLA1 binds Ctc1 and Stn1, and its interface with Ctc1 forms an entry port for G-strand DNA to the POLA1 active site. We thus provide a snapshot of four key components that are required for telomeric DNA synthesis in a single active complex-telomerase-core ribonucleoprotein, p50, CST and PolαPrim-that provides insights into the recruitment of CST and PolαPrim and the handoff between G-strand and C-strand synthesis.

Incorporation and improvement of a heterogeneous chemistry mechanism in the atmospheric chemistry model GRAPES_Meso5.1/CUACE and its impacts on secondary inorganic aerosol and PM2.5 simulations in Middle-Eastern China.

Heterogeneous chemistry is considered one of the critical pathways of secondary inorganic aerosol (SIA) productions. In this study, a heterogeneous chemistry mechanism is incorporated into the atmospheric chemistry model GRAPES_Meso5.1/CUACE. Varying uptake coefficient schemes of SO2 and NO2 are compared and the equivalent ratio of inorganic aerosol (ER)-dependent scheme for SO2 and relative humidity (RH)/ER-dependent scheme for NO2 are used to form the improved heterogeneous chemistry. Focusing on a severe haze episode in Middle-Eastern China, the impacts of heterogeneous mechanism on SIA and PM2.5 composition are investigated based on the updated model. Study results show that the differences in RH or ER uptake coefficients result in obvious differences in sulfate and nitrate concentrations, especially during the severe pollution period, because the ER schemes restrict the excessive production of sulfate and nitrate under high RH effectively by including the self-limitation of heterogeneous reactions, which shows better performance in capturing the magnitude and temporal variations of surface SIA and PM2.5. Normalized mean bias of sulfate, nitrate, ammonium, and PM2.5 in megacity Beijing decreases from -27.0, -28.3, -58.2, and -26.3 to 1.0, -2.2, -47.2, and -16.5 %, respectively. And the fractions of sulfate, nitrate, ammonium, and organics during the polluted period change from 13.7, 19.3, 6.9, and 60.1 to 16.5, 23.0, 7.6, and 52.9 %, respectively, which is more consistent with the observation (16.0, 23.2, 14.1, and 46.7 %). SIA and PM2.5 simulations in another megacity Shanghai have the similar improvements. The modeled SIA by heterogeneous processes contributes 11.7 % of total PM2.5 in Beijing and 22.5 % in Shanghai. That is 13.5 % in the Chinese megalopolis Beijing-Tianjin-Hebei and 19.8 % in Yangtze-River-Delta, indicating a considerable contribution of heterogeneous pathways to haze pollution. This work indicates the importance of detailed and reasonable heterogeneous schemes for better SIA and haze/fog prediction in the atmospheric chemistry model.

Information Extraction from the Text Data on Traditional Chinese Medicine: A Review on Tasks, Challenges, and Methods from 2010 to 2021.

Background: The practice of traditional Chinese medicine (TCM) began several thousand years ago, and the knowledge of practitioners is recorded in paper and electronic versions of case notes, manuscripts, and books in multiple languages. Developing a method of information extraction (IE) from these sources to generate a cohesive data set would be a great contribution to the medical field. The goal of this study was to perform a systematic review of the status of IE from TCM sources over the last 10 years. Methods: We conducted a search of four literature databases for articles published from 2010 to 2021 that focused on the use of natural language processing (NLP) methods to extract information from unstructured TCM text data. Two reviewers and one adjudicator contributed to article search, article selection, data extraction, and synthesis processes. Results: We retrieved 1234 records, 49 of which met our inclusion criteria. We used the articles to (i) assess the key tasks of IE in the TCM domain, (ii) summarize the challenges to extracting information from TCM text data, and (iii) identify effective frameworks, models, and key findings of TCM IE through classification. Conclusions: Our analysis showed that IE from TCM text data has improved over the past decade. However, the extraction of TCM text still faces some challenges involving the lack of gold standard corpora, nonstandardized expressions, and multiple types of relations. In the future, IE work should be promoted by extracting more existing entities and relations, constructing gold standard data sets, and exploring IE methods based on a small amount of labeled data. Furthermore, fine-grained and interpretable IE technologies are necessary for further exploration.

Multi-Year Variation of Ozone and Particulate Matter in Northeast China Based on the Tracking Air Pollution in China (TAP) Data.

With the rapid development of economy and urbanization acceleration, ozone (O3) pollution has become the main factor of urban air pollution in China after particulate matter. In this study, 90th percentile of maximum daily average (MDA) 8 h O3 (O3-8h-90per) and PM2.5 data from the Tracking Air Pollution in China (TAP) dataset were used to determine the mean annual, seasonal, monthly, and interannual distribution of O3-8h-90per and PM2.5 concentrations in Northeast China (NEC). The O3-8h-90per concentration was highest in Liaoning (>100 µg/m3), whereas the highest PM2.5 concentration was observed mainly in urban areas of central Liaoning and the Harbin−Changchun urban agglomeration (approximately 60 µg/m3). The O3-8h-90per concentrations were highest in spring and summer due to more intense solar radiation. On the contrary, the PM2.5 concentration increased considerably in winter influenced by anthropogenic activities. In May and June, the highest monthly mean O3-8h-90per concentrations were observed in central and western Liaoning, about 170−180 µg/m3, while the PM2.5 concentrations were the highest in January, February, and December, approximately 100 µg/m3. The annual mean O3-8h-90per concentration in NEC showed an increasing trend, while the PM2.5 concentration exhibited an annual decline. By 2020, the annual mean O3-8h-90per concentration in southern Liaoning had increased considerably, reaching 120−130 µg/m3. From the perspective of city levels, PM2.5 and O3-8h-90per also showed an opposite variation trend in the 35 cities of NEC. The reduced tropospheric NO2 column is consistent with the decreasing trend of the interannual PM2.5, while the increased surface temperature could be the main meteorological factor affecting the O3-8h-90per concentration in NEC. The results of this study enable a comprehensive understanding of the regional and climatological O3-8h-90per and PM2.5 distribution at distinct spatial and temporal scales in NEC.