Version Of Model Research Articles

Decadal Thermal Variability of the Upper Southern Ocean: Zonal Asymmetry

Abstract As the major sink of anthropogenic heat, the Southern Ocean has shown quasi-symmetric, deep-reaching warming since the mid-twentieth century. In comparison, the shorter-term heat storage pattern of the Southern Ocean is more complex and has notable impacts on regional climate and marine ecosystems. By analyzing observational datasets and climate model simulations, this study reveals that the Southern Ocean exhibits prominent decadal (>8 years) variability extending to ∼700-m depth and is characterized by out-of-phase changes in the Pacific and Atlantic–Indian Ocean sectors. Changes in the Pacific sector are larger in magnitude than those in the Atlantic–Indian Ocean sectors and dominate the total heat storage of the Southern Ocean on decadal time scales. Instead of heat uptake through surface heat fluxes, these asymmetric variations arise primarily from wind-driven heat redistribution. Pacemaker and preindustrial simulations of the Community Earth System Model version 1 (CESM1) suggest that these variations in Southern Ocean winds arise primarily from natural variability of the tropical Pacific, as represented by the interdecadal Pacific oscillation (IPO). Through atmospheric teleconnection, the positive phase of the IPO gives rise to higher-than-normal sea level pressure and anticyclonic wind anomalies in the 50°–70°S band of the Pacific sector. These winds lead to warming of 0–700 m by driving the convergence of warm water. The opposite processes, involving cyclonic winds and upper-layer divergence, occur in the Atlantic–Indian Ocean sector. These findings aid our understanding of the time-varying heat storage of the Southern Ocean and provide useful implications on initialized decadal climate prediction.

Investigating large language model (LLM) performance using in-context learning (ICL) for interpretation of ESMO and NCCN guidelines for lung cancer.

e13637 Background: The recent development of advanced LLMs has been suggested to improve patient care across several areas such as clinical-decision support or helping to answer patients’ questions. Hallucinations have been identified as a blocker for the use of LLMs in routine clinical practice. ICL and Retrieval Augmented Generation (RAG) could improve the LLM performance and reduce hallucinations, consecutively making the use of LLMs possible in clinical practice. Methods: A method using ICL and RAG was developed on top of health AI platform (Gosta MedKit) to interpret the most recent ESMO (Dec 2022 for NSCLC, Mar 2021 for SCLC) and NCCN (Nov 2023 for NSCLC and SCLC) clinical guidelines for lung cancer. Guidelines (including tables and diagrams) were curated into a text format, and split and stored into a vector database. OpenAI’s GPT4 Turbo model version gpt-4-1106-preview (GPT4-T), having the knowledge cutoff in April 2023, was used in all implementations. 11 questions about SCLC and 13 questions about NSCLC treatment recommendations and definitions were developed to evaluate the performance of different settings: GPT4-T (existing knowledge of the model), ICL with maximum context (ICL-MC) length (128k tokens) and ICL with RAG (ICL-RAG) heuristically including only the most relevant parts from vector database. Question prompts were generated for different settings and guidelines (ESMO, NCCN and both combined) and two oncologists evaluated 216 different responses and their alignment with ESMO and NCCN guidelines. Results: For responses using ESMO guidelines having oncologists’ consensus, ICL-MC and ICL-RAG respectively provided accurate responses for 83.3% and 79.2% of questions vs. 62.5% for GPT4-T. For responses using NCCN guidelines having oncologist consensus, ICL-RAG provided accurate responses for 83.3%, GPT4-T for 75.0% and ICL-MC for 33.3% of questions. When more flexibility was allowed in results interpretation (alignment either with ESMO or NCCN), GPT4-T provided accurate response for 87.5% vs. 70.8% with ICL-RAG and 58.3% with ICL-MC. No consensus around hallucinations was reported for ICL approaches, whereas GPT4-T hallucinated the response for 4.2% of questions with ESMO guidelines. Conclusions: ICL seems to improve the LLM performance for stricter tasks such as providing responses according to specific guidelines and reducing hallucinations. ICL outperformed GPT4-T in case of ESMO guidelines. This highlights the importance of taking local and latest guidelines into account when LLMs are used across different health systems and regulatory environments. In line with earlier studies, longer context for ICL makes models forget crucial information, which can be mitigated with the use of RAG to improve ICL performance and reduce costs when using the models.

Anthropogenically Driven Changes in the Carbon to Phosphorus Ratio of Marine Dissolved Organic Matter

AbstractMarine dissolved organic matter (DOM) cycles play a pivotal role in sustaining marine ecosystems and regulating the ocean's carbon sequestration from the atmosphere. However, the response of DOM cycles, including dissolved organic carbon (DOC) and dissolved organic phosphorus (DOP), to future climate change remains highly uncertain. Using the Community Earth System Model version 2 large ensemble simulations, we find that the C:P ratios in DOM are projected to increase by up to two‐fold in oligotrophic gyres by 2100. Increased upper ocean stratification reduces surface phosphate availability, thereby elevating phytoplankton C:P ratios and enhancing phytoplankton utilization of DOP, both acting to deprive DOM of P. Moreover, ocean stratification has a direct effect on exporting less DOC to the subsurface while accumulating more DOC at the sea surface. As a result of the strong sensitivity to ocean surface warming, the anthropogenically driven trends in upper ocean DOM concentration and its C:P ratios are estimated to emerge earlier from the simulated natural variability than upper ocean phosphate concentrations and net primary production—two key biogeochemical variables that are frequently monitored. This study suggests that changes in the C:P ratios of DOM could serve as a sensitive fingerprint of anthropogenic ocean warming, potentially exerting broad impacts on marine microbes. Our estimated 4% reduction in the globally integrated DOC export below 100 m is comparable to a 2% reduction in particulate organic carbon (POC) export by 2100, implying that global warming is likely to weaken the biological carbon pump through both DOC and POC.

Evaluation of the physical characteristics of reinforced concrete subject to corrosion using a poro-elastic acoustic model inversion technique applied to ultrasonic measurements

The use of reinforced concrete is foundational to modern infrastructure. Acknowledging this, it is imperative that health monitoring techniques be in place to study corrosion within these structures. By using a non-destructive method for detecting the early formation of cracks within reinforced concrete, the method presented in this paper seeks to improve upon traditional techniques of monitoring corrosion, within reinforced concrete structures. In this paper, the authors present a method to evaluate the physical characteristics of reinforced concrete subject to corrosion using a poro-elastic acoustic model inversion technique applied to a set of ultrasonic measurements, which constitutes a novel approach to the problem of observing the impact of corroding rebars and resulting concrete damage. A non-contact ultrasonic transducer is operated at a carrier frequency of 500 [kHz], with a layer of saltwater separating the sensor from the concrete surface. Following this non-contact measurement collection of the surface and rebar echo responses, a poro-elastic model is used to model the sound propagation, through an adapted version of the Biot-Stoll model. At first, a set of default parameters, obtained from the physical characteristics of the reinforced concrete, are used to match experimental and simulated acoustic signature of the sample. Performing statistical averaging along the corroding rebar within three samples over a period of nearly nine months, a small but monotonous increase in the distance between the concrete surface and the top of the rebar, indicating gradual corrosion of the rebar. Next, a non-linear optimization algorithm is used to optimize the match between measured and simulated echoes. Through the implementation of this model parameter optimization, the root mean square error between measured and simulated responses was reduced by 63.7% for the full signal, and 62.6% for the rebar echo.

The chemo-mechanics of initiation of crack advance in Niobium Tungsten Oxide single crystals due to delithiation

Niobium tungsten oxide (NWO) micron-sized single crystals show promise as the active particles in the anodes of lithium ion batteries due to their fast charge/discharge and high storage capabilities. Within the “block” phases such as Nb16W5O55, Nb14W3O44 and the HNb2O5 polymorph, Li ion diffusion occurs in a unidirectional manner along a single axis of the crystal and lithiation/delithiation results in significant length changes in this direction. A chemo-mechanics analysis is developed for delithiation of an NWO “block” phase single crystal that contains a pre-existing flaw in the form of a through-thickness edge crack. Galvanostatic delithiation occurs by the flux of Li ions through the free ends of the crystal and also through the faces of the crack, with concomitant shrinkage of the surface layer over a depth equal to that of the crack. The induced tensile stress in the surface layer can be of sufficient magnitude to advance the crack in the NWO crystal, consistent with recent experimental observations for Nb14W3O44. A 2D full-field finite element version of the chemo-mechanics model is used to explore the sensitivity of cracking to crack length, delithiation rate of the anode, diffusivity of Li ions within the NWO single crystal storage particle and to the kinetics of Li ion transfer across the particle surface-electrolyte interface. Additionally, a simplified, lumped parameter model is developed by assuming that the Li ion occupancy is spatially uniform in a surface layer that extends to a depth of the pre-existing crack, and has a spatially uniform but different value than that of the underlying core of the particle. This simplified model can be expressed as a set of first order differential equations in time and gives the main features of delithiation and the associated transient stress state in the single crystal. Both the full model and the lumped parameter version predict that crack advance from a pre-existing defect is triggered by a high delithiation rate, consistent with the experimental literature.

On Predicting Offshore Hub Height Wind Speed and Wind Power Density in the Northeast US Coast Using High-Resolution WRF Model Configurations during Anticyclones Coinciding with Wind Drought

We investigated the predictive capability of various configurations of the Weather Research and Forecasting (WRF) model version 4.4, to predict hub height offshore wind speed and wind power density in the Northeast US wind farm lease areas. The selected atmospheric conditions were high-pressure systems (anticyclones) coinciding with wind speed below the cut-in wind turbine threshold. There are many factors affecting the potential of offshore wind power generation, one of them being low winds, namely wind droughts, that have been present in future climate change scenarios. The efficiency of high-resolution hub height wind prediction for such events has not been extensively investigated, even though the anticipation of such events will be important in our increased reliance on wind and solar power resources in the near future. We used offshore wind observations from the Woods Hole Oceanographic Institution’s (WHOI) Air–Sea Interaction Tower (ASIT) located south of Martha’s Vineyard to assess the impact of the initial and boundary conditions, number of model vertical levels, and inclusion of high-resolution sea surface temperature (SST) fields. Our focus has been on the influence of the initial and boundary conditions (ICBCs), SST, and model vertical layers. Our findings showed that the ICBCs exhibited the strongest influence on hub height wind predictions above all other factors. The NAM/WRF and HRRR/WRF were able to capture the decreased wind speed, and there was no single configuration that systematically produced better results. However, when using the predicted wind speed to estimate the wind power density, the HRRR/WRF had statistically improved results, with lower errors than the NAM/WRF. Our work underscored that for predicting offshore wind resources, it is important to evaluate not only the WRF predictive wind speed, but also the connection of wind speed to wind power.

Совершенствование моделей инновационного процесса в нефтегазодобывающей отрасли

The evolution of innovation models in the oil and gas industry is crucial for ensuring competitiveness and sustainability in the rapidly changing global market. The article presents research results on advancing models of the innovation process as applied to enterprises in the Russian oil and gas industry. The research is based on the analysis and further development of ideas from various Russian and foreign authors regarding the use of the integrated network model of the fifth generation of the innovation process – 5G. The proposals formulated in the article are grounded in the potential opportunities of digitizing oil and gas enterprises and utilizing new digital tools. In this regard, taking into account the specifics of the innovation process in the oil and gas industry, the author’s version of the innovation process model is proposed, on the basis of which it is possible to reasonably plan, organize and digitalize the management of innovative processes of enterprises. The research results contribute to expanding the range of methodologically sound models of the innovation process and are aimed at enhancing the management of innovation processes in strategically important sectors of the Russian economy. By aligning with the principles of the fifth generation of innovation models and leveraging digital technologies, this model aims to enhance the competitiveness and resilience of the Russian oil and gas industry in the global market landscape.

Finite-size and finite-time scaling for kinetic rough interfaces.

We consider discrete models of kinetic rough interfaces that exhibit space-time scale invariance in height-height correlation. We use the generic scaling theory of Ramasco et al. [Phys. Rev. Lett. 84, 2199 (2000)0031-900710.1103/PhysRevLett.84.2199] to confirm that the dynamical structure factor of the height profile can uniquely characterize the underlying dynamics. We apply both finite-size and finite-time scaling methods that systematically allow an estimation of the critical exponents and the scaling functions, eventually establishing the universality class accurately. The finite-size scaling analysis offers an alternative way to characterize the anomalous rough interfaces. As an illustration, we investigate a class of self-organized interface models in random media with extremal dynamics. The isotropic version shows a faceted pattern and belongs to the same universality class (as shown numerically) as the Sneppen model (version A). We also examine an anisotropic version of the Sneppen model and suggest that the model belongs to the universality class of the tensionless Kardar-Parisi-Zhang (tKPZ) equationin one dimension.

A Simple, Robust, and Versatile MATLAB Formulation of the Dynamic Memdiode Model for Bipolar-Type Resistive Random Access Memory Devices

Modeling in an emerging technology like RRAM devices is one of the pivotal concerns for its development. In the current bibliography, most of the models face difficulties in implementing or simulating unconventional scenarios, particularly when dealing with complex input signals. In addition, circuit simulators like Spice require long running times for high-resolution results because of their internal mathematical implementation. In this work, a fast, simple, robust, and versatile model for RRAM devices built in MATLAB is presented. The proposed model is a recursive and discretized version of the dynamic memdiode model (DMM) for bipolar-type resistive switching devices originally implemented in LTspice. The DMM model basically consists of two coupled equations: one for the current (non-linear current generator) and a second one for the memory state of the device (time-dependent differential equation). This work presents an easy-to-use tool for researchers to reproduce the experimental behavior of their devices and predict the outcome from non-trivial experiments. Three study cases are reported, aimed at capturing different phenomenologies: a frequency effect study, a cycle-to-cycle variability fit, and a stochastic resonance impact analysis.

MGDrivE 3: A decoupled vector-human framework for epidemiological simulation of mosquito genetic control tools and their surveillance.

Novel mosquito genetic control tools, such as CRISPR-based gene drives, hold great promise in reducing the global burden of vector-borne diseases. As these technologies advance through the research and development pipeline, there is a growing need for modeling frameworks incorporating increasing levels of entomological and epidemiological detail in order to address questions regarding logistics and biosafety. Epidemiological predictions are becoming increasingly relevant to the development of target product profiles and the design of field trials and interventions, while entomological surveillance is becoming increasingly important to regulation and biosafety. We present MGDrivE 3 (Mosquito Gene Drive Explorer 3), a new version of a previously-developed framework, MGDrivE 2, that investigates the spatial population dynamics of mosquito genetic control systems and their epidemiological implications. The new framework incorporates three major developments: i) a decoupled sampling algorithm allowing the vector portion of the MGDrivE framework to be paired with a more detailed epidemiological framework, ii) a version of the Imperial College London malaria transmission model, which incorporates age structure, various forms of immunity, and human and vector interventions, and iii) a surveillance module that tracks mosquitoes captured by traps throughout the simulation. Example MGDrivE 3 simulations are presented demonstrating the application of the framework to a CRISPR-based homing gene drive linked to dual disease-refractory genes and their potential to interrupt local malaria transmission. Simulations are also presented demonstrating surveillance of such a system by a network of mosquito traps. MGDrivE 3 is freely available as an open-source R package on CRAN (https://cran.r-project.org/package=MGDrivE2) (version 2.1.0), and extensive examples and vignettes are provided. We intend the software to aid in understanding of human health impacts and biosafety of mosquito genetic control tools, and continue to iterate per feedback from the genetic control community.

An improved and extended parameterization of the CO2 15 µm cooling in the middle and upper atmosphere (CO2_cool_fort-1.0)

Abstract. The radiative infrared cooling of CO2 in the middle atmosphere, where it emits under non-local thermodynamic equilibrium (non-LTE) conditions, is a crucial contribution to the energy balance of this region and hence to establishing its thermal structure. The non-LTE computation is too CPU time-consuming to be fully incorporated into climate models, and hence it is parameterized. The most used parameterization of the CO2 15 µm cooling for Earth's middle and upper atmosphere was developed by Fomichev et al. (1998). The valid range of this parameterization with respect to CO2 volume mixing ratios (VMRs) is, however, exceeded by the CO2 of several scenarios considered in the Coupled Climate Model Intercomparison Projects, in particular the abrupt-4×CO2 experiment. Therefore, an extension, as well as an update, of that parameterization is both needed and timely. In this work, we present an update of that parameterization that now covers CO2 volume mixing ratios in the lower atmosphere from ∼0.5 to over 10 times the CO2 pre-industrial value of 284 ppmv (i.e. 150 to 3000 ppmv). Furthermore, it is improved by using a more contemporary CO2 line list and the collisional rates that affect the CO2 cooling rates. Overall, its accuracy is improved when tested for the reference temperature profiles as well as for measured temperature fields covering all expected conditions (latitude and season) of the middle atmosphere. The errors obtained for the reference temperature profiles are below 0.5 K d−1 for the present-day and lower CO2 VMRs. Those errors increase to ∼1–2K d−1 at altitudes between 110 and 120 km for CO2 concentrations of 2 to 3 times the pre-industrial values. For very high CO2 concentrations (4 to 10 times the pre-industrial abundances), those errors are below ∼1 K d−1 for most regions and conditions, except at 107–135 km, where the parameterization overestimates them by ∼1.2 %. These errors are comparable to the deviation of the non-LTE cooling rates with respect to LTE at about 70 km and below, but they are negligible (several times smaller) above that altitude. When applied to a large dataset of global (pole to pole and four seasons) temperature profiles measured by MIPAS (Michelson Interferometer for Passive Atmospheric Spectroscopy) (middle- and upper-atmosphere mode), the errors of the parameterization for the mean cooling rate (bias) are generally below 0.5 K d−1, except between 5×10-3 and 3×10-4 hPa (∼85–98 km), where they can reach biases of 1–2 K d−1. For single-temperature profiles, the cooling rate error (estimated by the root mean square – rms – of a statistically significant sample) is about 1–2 K d−1 below 5×10-3 hPa (∼85 km) and above 2×10-4 hPa (∼102 km). In the intermediate region, however, it is between 2 and 7 K d−1. For elevated stratopause events, the parameterization underestimates the mean cooling rates by 3–7 K d−1 (∼10 %) at altitudes of 85–95 km and the individual cooling rates show a significant rms (5–15 K d−1). Further, we have also tested the parameterization for the temperature obtained by a high-resolution version of the Whole Atmosphere Community Climate Model (WACCM-X), which shows a large temperature variability and wave structure in the middle atmosphere. In this case, the mean (bias) error of the parameterization is very small, smaller than 0.5 K d−1 for most atmospheric layers, reaching only maximum values of 2 K d−1 near 5×10-4 hPa (∼ 96 km). The rms has values of 1–2 K d−1 (∼20 %) below ∼2×10-2 hPa (∼80 km) and values smaller than 4 K d−1 (∼2 %) above 10−4 hPa (∼105 km). In the intermediate region between ∼5×10-3 and ∼2×10-4 hPa (85–102 km), the rms is in the range of 5–12 K d−1. While these values are significant in percentage at ∼5×10-3–5×10-4 hPa, they are very small above ∼5×10-4 hPa (96 km). The routine is very fast, taking (1.5–7.5) ×10-5 s, depending on the extension of the atmospheric profile, the processor and the Fortran compiler.

Prediction for post-ERCP pancreatitis in non-elderly patients with common bile duct stones: a cross-sectional study at a major Chinese tertiary hospital (2015–2023)

BackgroundPost-ERCP pancreatitis is one of the most common adverse events in ERCP-related procedures. The purpose of this study is to construct an online model to predict the risk of post-ERCP pancreatitis in non-elderly patients with common bile duct stones through screening of relevant clinical parameters.MethodsA total of 919 cases were selected from 7154 cases from a major Chinese tertiary hospital. Multivariable logistic regression model was fitted using the variables selected by the LASSO regression from 28 potential predictor variables. The internal and external validation was assessed by evaluating the receiver operating characteristic curve and the area under curve. Restricted cubic spline modelling was used to explore non-linear associations. The interactive Web application developed for risk prediction was built using the R “shiny” package.ResultsThe incidence of post-ERCP pancreatitis was 5.22% (48/919) and significantly higher in non-elderly patients with female, high blood pressure, the history of pancreatitis, difficult intubation, endoscopic sphincterotomy, lower alkaline phosphatase and smaller diameter of common bile duct. The predictive performance in the test and external validation set was 0.915 (95% CI, 0.858–0.972) and 0.838 (95% CI, 0.689–0.986), respectively. The multivariate restricted cubic spline results showed that the incidence of pancreatitis was increased at 33–50 years old, neutrophil percentage > 58.90%, hemoglobin > 131 g/L, platelet < 203.04 or > 241.40 × 109/L, total bilirubin > 18.39 umol / L, aspartate amino transferase < 36.56 IU / L, alkaline phosphatase < 124.92 IU / L, Albumin < 42.21 g / L and common bile duct diameter between 7.25 and 10.02 mm. In addition, a web server was developed that supports query for immediate PEP risk.ConclusionThe visualized networked version of the above model is able to most accurately predict the risk of PEP in non-elderly patients with choledocholithiasis and allows clinicians to assess the risk of PEP in real time and provide preventive treatment measures as early as possible.

Efficient Sentence Representation Learning via Knowledge Distillation with Maximum Coding Rate Reduction

Addressing the demand for effective sentence representation in natural language inference problems, this paper explores the utility of pre-trained large language models in computing such representations. Although these models generate high-dimensional sentence embeddings, a noticeable performance disparity arises when they are compared to smaller models. The hardware limitations concerning space and time necessitate the use of smaller, distilled versions of large language models. In this study, we investigate the knowledge distillation of Sentence-BERT, a sentence representation model, by introducing an additional projection layer trained on the novel Maximum Coding Rate Reduction (MCR2) objective designed for general-purpose manifold clustering. Our experiments demonstrate that the distilled language model, with reduced complexity and sentence embedding size, can achieve comparable results on semantic retrieval benchmarks, providing a promising solution for practical applications.

Modelling future coastal water pollution: impacts of point sources, socio-economic developments & multiple pollutants

Urbanisation is happening worldwide. In 2100, over 70% of the population is projected to live in highly urbanised areas. As a result, urban wastewater discharge may increase. This may add multiple pollutants to rivers and coastal waters. However, current knowledge on how urbanisation-related socio-economic developments affect coastal water pollution is limited. In this study, we analysed individual and combined impacts of wastewater treatment improvements, economic growth and city expansion on future coastal water pollution from point sources (sewage and open defecation) by sub-basin taking a multi-pollutant approach. We improved the existing MARINA-Multi model (version Global-1.0) by integrating hydrology and pollutant retentions in order to quantify river exports of total dissolved nitrogen, total dissolved phosphorus, microplastics and triclosan to coastal waters for 2010 and 2100 using scenario analysis. Globally, river exports from point sources are projected to more than double by 2100 for all pollutants, especially in Africa and Asia. Wastewater treatment improvements, economic growth and city expansion can have a positive (less pollution) or negative (more pollution) impact on future coastal water pollution. These impacts differ among pollutants and sub-basins. Wastewater treatment improvements may globally reduce multi-pollutant issues (−30% to −38% change on average) compared to the reference scenario (positive impact). Economic growth and city expansion may globally enhance multi-pollutant issues (+15% to +25% and +28% to +33% change on average, respectively) compared to the reference scenario (negative impact). A combined scenario, accounting for all three socio-economic developments simultaneously, may globally reduce or enhance pollutant issues (−21% to +50% change on average) compared to the reference scenario. In the combined scenario, the reinforcements of positive and negative impacts are pollutant- and region-dependent. Our study gives insights into future coastal water pollution, which aids in identifying management strategies for urban areas, hence contributing to reaching Sustainable Development Goal 14.

Systematic review: The use of large language models as medical chatbots in digestive diseases.

Interest in large language models (LLMs), such as OpenAI's ChatGPT, across multiple specialties has grown as a source of patient-facing medical advice and provider-facing clinical decision support. The accuracy of LLM responses for gastroenterology and hepatology-related questions is unknown. To evaluate the accuracy and potential safety implications for LLMs for the diagnosis, management and treatment of questions related to gastroenterology and hepatology. We conducted a systematic literature search including Cochrane Library, Google Scholar, Ovid Embase, Ovid MEDLINE, PubMed, Scopus and the Web of Science Core Collection to identify relevant articles published from inception until January 28, 2024, using a combination of keywords and controlled vocabulary for LLMs and gastroenterology or hepatology. Accuracy was defined as the percentage of entirely correct answers. Among the 1671 reports screened, we identified 33 full-text articles on using LLMs in gastroenterology and hepatology and included 18 in the final analysis. The accuracy of question-responding varied across different model versions. For example, accuracy ranged from 6.4% to 45.5% with ChatGPT-3.5 and was between 40% and 91.4% with ChatGPT-4. In addition, the absence of standardised methodology and reporting metrics for studies involving LLMs places all the studies at a high risk of bias and does not allow for the generalisation of single-study results. Current general-purpose LLMs have unacceptably low accuracy on clinical gastroenterology and hepatology tasks, which may lead to adverse patient safety events through incorrect information or triage recommendations, which might overburden healthcare systems or delay necessary care.

Developing a competency model for Chinese general practitioners: a mixed-methods study.

The Chinese government has formulated a series of policies and strengthened training of general practitioners (GPs) to support their role as "gatekeepers" of residents' health. This study aimed to explore the core competencies of Chinese GPs and develop a competency framework in line with China's actual conditions, which can provide a more scientific basis for the education, training, and evaluation of GPs. Literature analysis and behaviour event interviews were conducted to build the competency dictionary and the initial version of the competency model. Two rounds of Delphi were performed to gain consensus on the final model. The questionnaire survey was carried out in 10 provinces (municipalities, autonomous regions) of China, and GPs were invited to score the importance of each competency item. The total sample was randomly divided into two groups. One group was for exploratory factor analysis (EFA), and the other was for confirmatory factor analysis (CFA) to examine the scale's reliability and validity. The dictionary of general practitioners' competency including 107 competency items was constructed. After two rounds of Delphi, a consensus was reached on 60 competencies in 6 domains. A total of 1917 valid questionnaires were obtained in the nationwide survey. The average importance score of all second-level indicators is 4.53 ± 0.45. The Cronbach's α coefficient is 0.984. The results of the five factors extracted by EFA showing the 68.16% cumulative explained variance variation is considered to be consistent with the six dimensions obtained by Delphi after thorough discussion. The model fitness indexes obtained by CFA were acceptable (χ2/df = 4.909, CFI = 0.869, NFI = 0.841, RMSEA = 0.065). The values of the composite reliability (CR) of the six dimensions were all greater than 0.7 (0.943, 0.927, 0.937, 0.927, 0.943, 0.950), and the average of variance extracted (AVE) were all greater than 0.5 (0.562, 0.613, 0.649, 0.563, 0.626, 0.635). The results showed that the model has good reliability and validity. A competency model for GPs suited to China has been developed, which may offer guidance for future training and medical licensing examinations of GPs.

Exploring the Relative Importance of the MJO and ENSO to North Pacific Subseasonal Predictability

AbstractHere we explore the relative contribution of the Madden‐Julian Oscillation (MJO) and El Niño Southern Oscillation (ENSO) to midlatitude subseasonal predictive skill of upper atmospheric circulation over the North Pacific, using an inherently interpretable neural network applied to pre‐industrial control runs of the Community Earth System Model version 2. We find that this interpretable network generally favors the state of ENSO, rather than the MJO, to make correct predictions on a range of subseasonal lead times and predictand averaging windows. Moreover, the predictability of positive circulation anomalies over the North Pacific is comparatively lower than that of their negative counterparts, especially evident when the ENSO state is important. However, when ENSO is in a neutral state, our findings indicate that the MJO provides some predictive information, particularly for positive anomalies. We identify three distinct evolutions of these MJO states, offering fresh insights into opportune forecasting windows for MJO teleconnections.

The impact of internal climate variability on OH trends between 2005 and 2014

The hydroxyl radical (OH) lies at the nexus of climate and air quality as the primary oxidant for both reactive greenhouse gases and many hazardous air pollutants. To better understand the role of climate variability on spatiotemporal patterns of OH, we utilize a 13-member ensemble of the Community Earth System Model version 2-Whole Atmosphere Community Climate Model version 6 (CESM2-WACCM6), a fully coupled chemistry-climate model, spanning the years 1950–2014. Ensemble members vary only in their initial conditions of the climate state in 1950. We focus on the final decade of the simulation, 2005–2014, when prior studies disagree on the signs of the global OH trends. The ensemble mean global airmass-weighted mean tropospheric column OH ( ΩTOH ), which is an estimate of the forced signal, increases by 0.06%/year between 2005 and 2014 while regional ΩTOH trends range from −0.56%/year over Southern Europe to +0.64%/year over South America. We show that ten-year ΩTOH trends are strongly affected by internal climate variability, as the spread of ΩTOH trends across the ensemble varies between 0.23%/year in Asia and 1.53%/year in South America. We train a fully connected neural network to emulate the ΩTOH simulated by the CESM2-WACCM6 model and combine it with satellite observations to interpret the role of OH chemical proxies. While the OH chemical proxies are subject to internal variability, the impact of internal variability on ΩTOH trends is primarily due to the meteorological parameters except for South America. Forced trends in global mean ΩTOH do not unambiguously emerge from trends driven by internal variability over the 2005–2014 period. The observation-constrained ΩTOH presents opposite trends due to climate variability, resulting in varying conclusions on the attribution of OH to CH4 trends.

Calibration and development of safety performance functions for two-way stop-control intersections on rural two-lane highways in Louisiana

The first edition of the Highway Safety Manual (HSM) contains a simplistic version of the crash prediction model for two-way stop-controlled intersections (TWSC) on rural two-lane highways. This model considers AADT on major and minor roads, with the base conditions defined as no intersection skewness, no turning lanes, and no lighting. A crash modification factor (CMF) will be applied if an intersection has conditions different from the base condition. However, the HSM model does not take account of curvature. It is well known that curved TWSC intersections are less safe than non-curved ones, particularly on rural two-lane roadways. This paper presents the development of crash prediction models incorporating intersection geometrics for TWSC intersections on rural two-lane highways in Louisiana. Then, it compares the results from the developed model with the calibrated HSM model. The negative binomial model was used with 5126 TWSC intersections verified one by one, including both three- and four-leg intersections from all parishes (counties). The estimation results indicate that AADT, curve radius, and intersection skewness angle significantly impact expected crash frequency for both three- and four-leg intersections. This research utilizes cumulative residual plots, mean absolute error, and root mean square error for comparative analysis of HSM models, HSM models with calibration, and Louisiana-specific models. The results show that Louisiana-specific SPFs outperformed the calibrated SPFs with superior reliability. Calibration factors of 0.58 for three-leg intersections and 0.46 for four-leg intersections are estimated, suggesting that the original HSM model overpredicts crashes in Louisiana.

Exploring walking behaviour and perceived walkability of older adults in London

IntroductionThere is an emerging imperative to design walkable communities catering to older adults, particularly given the context of an increasingly aging population. However, studies of how older citizens perceive walkability remain scarce. Therefore, this study aims to identify the factors that influence walking behaviours and perceived walkability among senior citizens, whilst also illuminating the associations between them. MethodsA quantitative analysis was conducted on 200 older individuals living in Ealing, London, an area with a large proportion of senior citizens. Utilising the self-developed Neighbourhood Environment Walkability Scale for Older Adults (NEWS-OA), based on the abbreviated version of NEWS, both linear and logistic regression models were employed to decode the data. ResultsOur results show that NEWS-OA is appropriate for assessing perceived walkability among senior citizens and reveal concerns regarding traffic safety. Communities with high walkability are likely to encourage an extended distance, duration, and frequency of walks. However, socioeconomic status (SES) was found to have a less pronounced effect on walking frequency and distance. The study further demonstrates that perceptions of safety were lower among older adults with a high SES. We found that positive perceptions of all the variables tested, except Aesthetics, encouraged walking. Furthermore, the study reveals the crucial influence of these perceptions on older adults’ inclination to walk for various purposes, such as the role of perceived safety in terms of social interaction. ConclusionsThis study offers insights that can be used to make London a more age-friendly and walkable city. It can also serve as a reference for other cities to enhance inclusiveness.