Primary Dataset Research Articles

Evaluating external generalizability of machine learning models for recycled aggregate concrete property prediction

Machine learning (ML) models have gained importance in predicting recycled aggregate concrete (RAC) properties, offering supposed benefits over conventional empirical and statistical techniques. However, whether ML models are externally generalizable for predicting RAC properties remains unanswered. This study addresses this gap by developing a systematic experimental framework for evaluating the external generalizability of ML models for predicting the compressive strength of RAC. Using a literature review, the authors created a primary dataset of 414 data points and sourced a secondary dataset comprising 330 data points from a previous paper. In Phase 1, prominent ML models like Random Forest (RF) and Extreme Gradient Boost (XGB) were tested for high-accuracy prediction on both primary and secondary datasets. A coefficient of determination (R2) as high as 0.76 for the primary dataset and 0.82 for the secondary dataset for testing sets was obtained for XGB. However, when the best-performing models of phase 1 were trained and tested with data sourced from different datasets in varying combinations, the ML model's performance significantly deteriorated (R2 < 0.25), demonstrating that ML models are not externally generalizable. The study's results highlight the trade-off between the ML model's prediction accuracy within the given dataset and its external generalizability. The study reveals that complex ML models, like XGB, may over-fit specific data, reducing their generalizability. These findings call for the intelligent usage of ML tools to identify nuanced hypotheses and promote rigorous science rather than unilaterally working on the accuracy of ML models. The study emphasizes the consideration of ML models' generalizability in the future.

A customised vision transformer for accurate detection and classification of Java Plum leaf disease

Vision Transformer (ViT) has recently attracted significant attention for its performance in image classification. However, studies have yet to explore its potential in detecting and classifying plant leaf disease. Most existing research on diseased plant leaf detection has focused on non-transformer convolutional neural networks (CNN). Moreover, the studies that applied ViT narrowly experimented using hyperparameters such as image size, patch size, learning rate, attention head, epoch, and batch size. However, these hyperparameters significantly contribute to the model performance. Recognising the gap, this study applied ViT to Java Plum disease detection using optimised hyperparameters. To harness the performance of ViT, this study presents an experiment on Java Plum leaf disease detection. Java Plum leaf diseases significantly threaten agricultural productivity by negatively impacting yield and quality. Timely detection and diagnosis are essential for successful crop management. The primary dataset collected in Bangladesh includes six classes, ‘Bacterial Spot’, ‘Brown Blight’, ‘Powdery Mildew’, and ‘Sooty Mold’, ‘healthy’, and ‘dry’. This experiment contributes to a thorough understanding of Java Plum leaf diseases. Following rigorous testing and refinement, our model demonstrated a significant accuracy rate of 97.51%. This achievement demonstrates the possibilities of using deep-learning tools in agriculture and inspires further research and application in this field. Our research offers a foundational model to ensure crop quality by precise detection, instilling confidence in the global Java Plum market.

THE IMPACT OF PRODUCT INNOVATION ON FEMALE ENTREPRENEURSHIP SUCCESS: A CASE OF SAUDI ARABIA

Recent reforms in Saudi Arabia granted empowerment to women, opening new entrepreneurial avenues for them. This study aims to assess the influence of product innovation on the success of female entrepreneurs in Saudi Arabia. Employing a quantitative approach, data was collected through a survey questionnaire. The primary data set included responses from 256 Saudi female entrepreneurs. SmartPLS was utilized for analysis, employing the PLS-SEM statistical technique. The results revealed a positive overall effect of product innovation on female entrepreneurs and their intentions. The study offers practical insights for female entrepreneurs and the Saudi government to address challenges faced by Saudi women. Recommendations include the formulation of policies that provide sustained support for women entrepreneurs. Additionally, the study suggests potential avenues for future research in this domain.

VashaNet: An automated system for recognizing handwritten Bangla basic characters using deep convolutional neural network

Automated character recognition is currently highly popular due to its wide range of applications. Bengali handwritten character recognition (BHCR) is an extremely difficult issue because of the nature of the script. Very few handwritten character recognition (HCR) models are capable of accurately classifying all different sorts of Bangla characters. Recently, image recognition, video analytics, and natural language processing have all found great success using convolutional neural network (CNN) due to its ability to extract and classify features in novel ways. In this paper, we introduce a VashaNet model for recognizing Bangla handwritten basic characters. The suggested VashaNet model employs a 26-layer deep convolutional neural network (DCNN) architecture consisting of nine convolutional layers, six max pooling layers, two dropout layers, five batch normalization layers, one flattening layer, two dense layers, and one output layer. The experiment was performed over 2 datasets consisting of a primary dataset of 5750 images, CMATERdb 3.1.2 for the purpose of training and evaluating the model. The suggested character recognition model worked very well, with test accuracy rates of 94.60% for the primary dataset, 94.43% for CMATERdb 3.1.2 dataset. These remarkable outcomes demonstrate that the proposed VashaNet outperforms other existing methods and offers improved suitability in different character recognition tasks. The proposed approach is a viable candidate for the high efficient practical automatic BHCR system. The proposed approach is a more powerful candidate for the development of an automatic BHCR system for use in practical settings.

Beyond meta-studies: Learnings from a large multi-site primary dataset on non-tangible benefits of nature-based solutions

Nature-Based Solutions (NBS) are growing in popularity as approaches for, among others, the reduction of hydro-meteorological risks. However, their uptake is still slow despite the recognition of their role in enabling a smarter, more systemic, and faster adaptation to climate change. Uncertainties regarding the valuation and subsequent economic feasibility of NBS are deemed to be a substantial barrier to their uptake. The monetary valuation of the non-tangible benefits of NBS (e.g., increased recreation and enhanced biodiversity) is seen as a plausible strategy to get closer to providing a holistic assessment. However, the quantification of these non-tangible benefits is often not integrated into economic assessments of NBS, partly as methods that can be replicated across sites have been lacking. This situation risks leading to inconsistent and/or biased valuations, which may negatively influence or delay the prioritization of NBS. Thus, our study aims to collect primary data on the non-tangible benefits of NBS across different European study sites and to identify patterns in how the different types and contexts of NBS influence people's valuation. We applied a contingent valuation survey to six different peri‑urban NBS study sites, where the focus is on the reduction of hydro-meteorological risks. The target of the survey was the general public, and we used willingness-to-pay questions to quantify the value given by the respondents to the NBS. Our results highlighted that people value the multiple benefits of NBS, and positively react to implementations improving nature across all sites. Moreover, similar variables seem to influence the value attributed to NBS across sites, with income and personal preferences ranking highly among them. These insights indicate that decision-makers and NBS-planners can learn more broadly from past experiences from a variety of (cultural) contexts, which can assist in delivering more targeted and multi-functional NBS.

A dynamic prediction model for preeclampsia using the sFlt-1/PLGF ratio combined with multiple factors

ObjectivePreeclampsia (PE) is a pregnancy-related multi-organ disease and a significant cause of incidence rate and mortality of pregnant women and newborns worldwide. Delivery remains the only available treatment for PE. This study aims to establish a dynamic prediction model for PE.MethodsA total of 737 patients who visited our hospital from January 2021 to June 2022 were identified according to the inclusion and exclusion criteria, forming the primary dataset. Additionally, 176 singleton pregnant women who visited our hospital from July 2022 to November 2022 comprised the verification set. We investigated different gestational weeks of sFlt-1/PLGF (soluble FMS-like tyrosine kinase-1, placental growth factor) ratio combined with maternal characteristics and routine prenatal laboratory results in order to predict PE in each trimester. Multivariate logistic regression was used to establish the prediction model for PE at different gestational weeks. The discrimination, calibration, and clinical validity were utilized to evaluate predictive models as well as models in external validation queues.ResultsAt 20–24 weeks, the obtained prediction model for PE yielded an area under the curve of 0.568 (95% confidence interval, 0.479–0.657). At 25–29 weeks, the obtained prediction model for PE yielded an area under the curve of 0.773 (95% confidence interval, 0.703–0.842)and 0.731 (95% confidence interval, 0.653–0.809) at 30–34 weeks. After adding maternal factors, uterine artery pulsation index(Ut-IP), and other laboratory indicators to the sFlt-1/PLGF ratio, the predicted performance of PE improved. It found that the AUC improved to 0.826(95% confidence interval, 0.748 ∼ 0.904) at 20–24 weeks, 0.879 (95% confidence interval, 0.823 ∼ 0.935) at 25–29 weeks, and 0.862(95% confidence interval, 0.799 ∼ 0.925) at 30–34 weeks.The calibration plot of the prediction model indicates good predictive accuracy between the predicted probability of PE and the observed probability. Furthermore, decision-curve analysis showed an excellent clinical application value of the models.ConclusionUsing the sFlt-1/PLGF ratio combined with multiple factors at 25–29 weeks can effectively predict PE, but the significance of re-examination in late pregnancy is not significant.

Seismic net-to-gross estimation for a geologic model update: A case study from a turbidite lobe reservoir in the deepwater of the Niger Delta

Geologic model updates are routinely performed in mature fields to obtain improved descriptions of facies distributions and reservoir properties such as volume of shale, net to gross (NTG), and porosity, for better understanding of the static and dynamic behaviors of reservoirs for effective well placement, improved production, and monitoring. A simple but integrated seismic NTG estimation approach, using detuned seismic amplitudes is used in guiding NTG modeling during the geologic model update of a thin turbidite lobe reservoir in a mature oil field in the deep offshore Niger Delta. The objective is to address NTG overestimation and gross rock volume (GRV) uncertainty in a previous model arising from seismic amplitude tuning effects. A seismic NTG approach is chosen relative to sophisticated deterministic or stochastic inversion techniques to avoid tuning effects, which usually bias NTG estimates in thin turbidite reservoirs. The primary data set is a 2019 reprocessed prestack depth-migrated (PSDM) seismic data vintage, which had better resolution, higher signal-to-noise ratio, and more appropriate angle-stack apertures for amplitude variation with angle fidelity, than the older 2011 PSDM seismic data that are used in the previous geologic model. The methodology involved rock-physics analysis, seismic data quality checks (QCs), tuned area determination, detuning of composite seismic amplitudes of the top and base reservoir, and their direct calibration to NTG at wells. Good correlations are obtained between the detuned composite seismic amplitudes and NTG at wells. The seismic NTG map shows good calibrations at wells and provides a robust trend for net sand modeling in the oil pool and aquifer. Static model QCs and dynamic simulations prove that the seismic NTG attribute addressed the GRV uncertainty in the earlier model, thus giving confidence for using the updated model for the planning and geosteering of infill wells, sand completions, reservoir monitoring, and production.

Congruent patterns of cryptic cladogenesis revealed using RADseq and Sanger sequencing in a velvet worm species complex (Onychophora: Peripatopsidae: Peripatopsis sedgwicki)

In the present study, first generation DNA sequencing (mitochondrial cytochrome c oxidase subunit one, COI) and reduced-representative genomic RADseq data were used to understand the patterns and processes of diversification of the velvet worm, Peripatopsis sedgwicki species complex across its distribution range in South Africa. For the RADseq data, three datasets (two primary and one supplementary) were generated corresponding to 1,259–11,468 SNPs, in order to assess the diversity and phylogeography of the species complex. Tree topologies for the two primary datasets were inferred using maximum likelihood and Bayesian inferences methods. Phylogenetic analyses using the COI datasets retrieved four distinct, well-supported clades within the species complex. Five species delimitation methods applied to the COI data (ASAP, bPTP, bGMYC, STACEY and iBPP) all showed support for the distinction of the Fort Fordyce Nature Reserve specimens. In the main P. sedgwicki species complex, the species delimitation methods revealed a variable number of operational taxonomic units and overestimated the number of putative taxa. Divergence time estimates coupled with the geographic exclusivity of species and phylogeographic results suggest recent cladogenesis during the Plio/Pleistocene. The RADseq data were subjected to a principal components analysis and a discriminant analysis of principal components, under a maximum-likelihood framework. The latter results corroborate the four main clades observed using the COI data, however, applying additional filtering revealed additional diversity. The high overall congruence observed between the RADseq data and COI data suggest that first generation sequence data remain a cheap and effective method for evolutionary studies, although RADseq does provide a far greater resolution of contemporary temporo-spatial patterns.

Semantic Representation of Low‐Cycle‐Fatigue Testing Data Using a Fatigue Test Ontology and ckan.kupferdigital Data Management System

Addressing a strategy for publishing open and digital research data, this article presents the approach for streamlining and automating the process of storage and conversion of research data to those of semantically queryable data on the web. As the use case for demonstrating and evaluating the digitalization process, the primary datasets from low‐cycle‐fatigue testing of several copper alloys are prepared. The fatigue test ontology (FTO) and ckan.kupferdigital data management system are developed as two main prerequisites of the data digitalization process. FTO has been modeled according to the content of the fatigue testing standard and by reusing the basic formal ontology, industrial ontology foundry core ontology, and material science and engineering ontology. The ckan.kupferdigital data management system is also constructed in such a way that enables the users to prepare the protocols for mapping the datasets into the knowledge graph and automatically convert all the primary datasets to those machine‐readable data which are represented by the web ontology language. The retrievability of the converted digital data is also evaluated by querying the example competency questions, confirming that ckan.kupferdigital enables publishing open data that can be highly reused in the semantic web.

USE OF STATISTICAL TOOLS TO STUDY THE CONSUMER PREFERENCE OVER GREEN PRODUCTS

Now a days, people are more concerned with the environment change. Hence, they prefer environment friendly products. The present study is based on the behavior of consumers towards green products. To study this, a primary data set is taken. The data is collected through a questionnaire and the response of consumers of green product is collected to know their preference over the green products. The data is analysed in detail to describe the various statistical methods. The techniques used here are explained in detail. It is required to know the basics of the calculations if one wants to use these tools. From the data it is found that the consumers’ preference does depend on many parameters. The tabular and graphical presentation of the data is employed to give the clear picture of the responses. The relationship between two parameters responsible for the preferences of people over green marketing is shown by contingency tables and appropriate hypothesis tests are used to study the relationship between variables. A detailed discussion is done in the paper.

AI Detection of Glottic Neoplasm Using Voice Signals, Demographics, and Structured Medical Records.

This study investigated whether artificial intelligence (AI) models combining voice signals, demographics, and structured medical records can detect glottic neoplasm from benign voice disorders. We used a primary dataset containing 2-3 s of vowel "ah", demographics, and 26 items of structured medical records (e.g., symptoms, comorbidity, smoking and alcohol consumption, vocal demand) from 60 patients with pathology-proved glottic neoplasm (i.e., squamous cell carcinoma, carcinoma in situ, and dysplasia) and 1940 patients with benign voice disorders. The validation dataset comprised data from 23 patients with glottic neoplasm and 1331 patients with benign disorders. The AI model combined convolutional neural networks, gated recurrent units, and attention layers. We used 10-fold cross-validation (training-validation-testing: 8-1-1) and preserved the percentage between neoplasm and benign disorders in each fold. Results from the AI model using voice signals reached an area under the ROC curve (AUC) value of 0.631, and additional demographics increased this to 0.807. The highest AUC of 0.878 was achieved when combining voice, demographics, and medical records (sensitivity: 0.783, specificity: 0.816, accuracy: 0.815). External validation yielded an AUC value of 0.785 (voice plus demographics; sensitivity: 0.739, specificity: 0.745, accuracy: 0.745). Subanalysis showed that AI had higher sensitivity but lower specificity than human assessment (p < 0.01). The accuracy of AI detection with additional medical records was comparable with human assessment (82% vs. 83%, p = 0.78). Voice signal alone was insufficient for AI differentiation between glottic neoplasm and benign voice disorders, but additional demographics and medical records notably improved AI performance and approximated the prediction accuracy of humans. NA Laryngoscope, 134:4585-4592, 2024.

Reflection of Daily, Seasonal and Interannual Variations in Run-Off of a Small River in the Water Isotopic Composition (δ2H, δ18O): A Case of the Ala-Archa Mountain River Basin with Glaciation (Kyrgyzstan, Central Asia)

Small intermountain river basins are most suitable for developing new methods to estimate water balance due to their well-defined catchment boundaries, relatively rapid runoff processes, and accessible landscapes for study. In general terms, dissecting the hydrograph of a small mountain river requires calibration of the flow model against multi-year data sets, including (a) glacier mass balance and snow water content, (b) radiation balance calculation, (c) estimation of the groundwater contribution, and (d) water discharge measurements. The minimum primary data set is limited to the precipitation and temperature distributions at the catchment. This approach postulates that the conditions for the formation of all components of river flow are known in advance. It is reduced to calculating the dynamic balance between precipitation (input part) and runoff, ablation, and evaporation (output part). In practice, accurately accounting for the inflow and outflow components of the balance, as well as the impact of regulating reservoirs, can be a challenging task that requires significant effort and expense, even for the extensively researched catchments. Our studies indicate the potential benefits of an approach based on one-time, but detailed, observations of stable isotope composition, temperature, and water chemistry, in addition to standard datasets. This paper presents the results of the 2022–2023 work conducted in the basin of the small mountain river Ala-Archa, located on the northern slope of the Kyrgyz Range in Tien-Shan, which was chosen as an example due to its well-studied nature. Our approach could identify previously unknown factors of flow formation and assess the time and effectiveness of work in similar conditions.

Systematic druggable genome-wide Mendelian randomization identifies therapeutic targets for lung cancer

BackgroundDrug repurposing provides a cost-effective approach to address the need for lung cancer prevention and therapeutics. We aimed to identify actionable druggable targets using Mendelian randomization (MR).MethodsSummary-level data of gene expression quantitative trait loci (eQTLs) were sourced from the eQTLGen resource. We procured genetic associations with lung cancer and its subtypes from the TRICL, ILCCO studies (discovery) and the FinnGen study (replication). We implemented Summary-data-based Mendelian Randomization analysis to identify potential therapeutic targets for lung cancer. Colocalization analysis was further conducted to assess whether the identified signal pairs shared a causal genetic variant.FindingsIn the main analysis dataset, we identified 55 genes that demonstrate a causal relationship with lung cancer and its subtypes. However, in the replication cohort, only three genes were found to have such a causal association with lung cancer and its subtypes, and of these, HYKK (also known as AGPHD1) was consistently present in both the primary analysis dataset and the replication cohort. Following HEIDI tests and colocalization analyses, it was revealed that HYKK (AGPHD1) is associated with an increased risk of squamous cell carcinoma of the lung, with an odds ratio and confidence interval of OR = 1.28,95%CI = 1.24 to 1.33.InterpretationWe have found that the HYKK (AGPHD1) gene is associated with an increased risk of squamous cell carcinoma of the lung, suggesting that this gene may represent a potential therapeutic target for both the prevention and treatment of lung squamous cell carcinoma.

Active convolutional neural networks sign language (ActiveCNN-SL) framework: a paradigm shift in deaf-mute communication

Real-time speech-to-text and text-to-speech technologies have significantly influenced the accessibility of communication for individuals who are deaf or mute. This research aims to assess the efficacy of these technologies in facilitating communication between deaf or mute individuals and those who are neither deaf nor mute. A mixed-method approach will incorporate qualitative and quantitative data collection and analysis techniques. The study will involve participants from deaf or mute and non-deaf or non-mute communities. The research will scrutinize the precision and efficiency of communication using these technologies and evaluate user experience and satisfaction. Furthermore, the study intends to pinpoint potential obstacles and limitations of these technologies and offer suggestions for enhancing their effectiveness in fostering inclusivity. The study proposes an active learning framework for sign language gesture recognition, termed Active Convolutional Neural Networks—Sign Language (ActiveCNN-SL). ActiveCNN-SL aims to minimize the labeled data required for training and augment the accuracy of sign language gesture recognition through iterative human feedback. This proposed framework holds the potential to enhance communication accessibility for deaf and mute individuals and encourage inclusivity across various environments. The proposed framework is trained using two primary datasets: (i) the Sign Language Gesture Images Dataset and (ii) the American Sign Language Letters (ASL)—v1. The framework employs Resnet50 and YoloV.8 to train the datasets. It has demonstrated high performance in terms of precision and accuracy. The ResNet model achieved a remarkable accuracy rate of 99.98% during training, and it also exhibited a validation accuracy of 100%, surpassing the baseline CNN and RNN models. The YOLOv8 model outperformed previous methods on the ASL alphabet dataset, achieving an overall mean average accuracy for all classes of 97.8%.

Advanced machine learning approaches for predicting permeability in reservoir pay zones based on core analyses

Permeability is the most important petrophysical characteristic for determining how fluids pass through reservoir rocks. This study aims to develop and assess intelligent computer-based models for predicting permeability. The research focuses on three novel models—Decision Tree, Bagging Tree, and Extra Trees—while also investigating previously applied techniques such as random forest, support vector regressor (SVR), and multiple variable regression (MVR). The primary dataset consists of 197 data points from a heterogeneous petroleum reservoir in the Jeanne d'Arc Basin, including laboratory-derived permeability (K), oil saturation (SO), water saturation (SW), grain density (ρgr), porosity (φ), and depth. The most effective machine learning models are identified by a thorough analysis that makes use of a variety of statistical metrics, such as the coefficient of the determinant (R2), mean squared error (MSE), mean absolute error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE), maximum error (maxE), and minimum error (minE). Additionally, core features are ranked based on their importance in permeability modeling. This study deviates from conventional approaches by proposing an efficient means of forecasting permeability, reducing reliance on labor-intensive and time-consuming laboratory work. The findings reveal that MVR is unsuitable for permeability prediction, with all developed models outperforming it. Extra Trees emerges as the most accurate model, with an R2 of 0.976, while random forest and bagging tree exhibit slightly lower R2 values of 0.961 and 0.964, respectively. The ranking of these algorithms based on performance criteria is as follows: extra trees, bagging tree, random forest, SVR, decision tree, and MVR. The study also presents a detailed analysis of the impact of input parameters, highlighting porosity (φ) and water saturation (SW) as the most influential, while grain density (ρgr), oil saturation (SO), and depth are considered less important. This study contributes to the petroleum industry's knowledge by showcasing the inadequacy of MVR and highlighting the superior performance of machine learning models, particularly Extra Trees. The proposed models employed in this study can help engineers and researchers determine reservoir permeability quickly and accurately by using a few core attributes, reducing the dependency on resource-intensive and time-consuming laboratory work.

An Experimental Study on Deep Learning Technique Implemented on Low Specification OpenMV Cam H7 Device

This research aims to identify and recognize the OpenMV Camera H7. In this research, all tests were carried out using Deep Machine Learning and applied to several functions, including Face Recognition, Facial Expression Recognition, Detection and Calculation of the Number of Objects, and Object Depth Estimation. Face Expression Recognition was used in the Convolutional Neural Network to recognize five facial expressions: angry, happy, neutral, sad, and surprised. This allowed the use of a primary dataset with a 48MP resolution camera. Some scenarios are prepared to meet environment variability in the implementation, such as indoor and outdoor environments, with different lighting and distance. Most pre-trained models in each identification or recognition used mobileNetV2 since this model allows low computation cost and matches with low hardware specifications. The object detection and counting module compared two methods: the conventional Haar Cascade and the Deep Learning MobileNetV2 model. The training and validation process is not recommended to be carried out on OpenMV devices but on computers with high specifications. This research was trained and validated using selected primary and secondary data, with 1500 image data. The computing time required is around 5 minutes for ten epochs. On average, recognition results on OpenMV devices take around 0.3 - 2 seconds for each frame. The accuracy of the recognition results varies depending on the pre-trained model and the dataset used, but overall, the accuracy levels achieved tend to be very high, exceeding 96.6%.

Generalization of a Deep Learning Model for Continuous Glucose Monitoring-Based Hypoglycemia Prediction: Algorithm Development and Validation Study.

Predicting hypoglycemia while maintaining a low false alarm rate is a challenge for the wide adoption of continuous glucose monitoring (CGM) devices in diabetes management. One small study suggested that a deep learning model based on the long short-term memory (LSTM) network had better performance in hypoglycemia prediction than traditional machine learning algorithms in European patients with type 1 diabetes. However, given that many well-recognized deep learning models perform poorly outside the training setting, it remains unclear whether the LSTM model could be generalized to different populations or patients with other diabetes subtypes. The aim of this study was to validate LSTM hypoglycemia prediction models in more diverse populations and across a wide spectrum of patients with different subtypes of diabetes. We assembled two large data sets of patients with type 1 and type 2 diabetes. The primary data set including CGM data from 192 Chinese patients with diabetes was used to develop the LSTM, support vector machine (SVM), and random forest (RF) models for hypoglycemia prediction with a prediction horizon of 30 minutes. Hypoglycemia was categorized into mild (glucose=54-70 mg/dL) and severe (glucose<54 mg/dL) levels. The validation data set of 427 patients of European-American ancestry in the United States was used to validate the models and examine their generalizations. The predictive performance of the models was evaluated according to the sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). For the difficult-to-predict mild hypoglycemia events, the LSTM model consistently achieved AUC values greater than 97% in the primary data set, with a less than 3% AUC reduction in the validation data set, indicating that the model was robust and generalizable across populations. AUC values above 93% were also achieved when the LSTM model was applied to both type 1 and type 2 diabetes in the validation data set, further strengthening the generalizability of the model. Under different satisfactory levels of sensitivity for mild and severe hypoglycemia prediction, the LSTM model achieved higher specificity than the SVM and RF models, thereby reducing false alarms. Our results demonstrate that the LSTM model is robust for hypoglycemia prediction and is generalizable across populations or diabetes subtypes. Given its additional advantage of false-alarm reduction, the LSTM model is a strong candidate to be widely implemented in future CGM devices for hypoglycemia prediction.

#645 Inferring trends in renal function based on a single serum creatinine with a spot urine creatinine-to-osmolality ratio in outpatient settings

Abstract Background and Aims Assessing trends in renal function can be challenging when paired serum creatinine (SCr) levels are not available, especially at the initial visit for evaluating renal function. The aim of this study was to validate the use of a spot urine creatinine-to-osmolality ratio (sUCr/Osm) as a surrogate indicator of urinary creatinine excretion rate for inferring renal function trends.1 The objective is to create a flowchart that assists in determining appropriate outpatient management based on a single SCr value with sUCr/Osm. Method The current study conducted a secondary analysis using the primary datasets provided by The Assessment, Serial Evaluation, and Subsequent Sequelae of Acute Kidney Injury (ASSESS-AKI) Study.2 The term “stable renal function” refers to the absence of an interval increase in SCr of greater than or equal to 26.5 μmol/L (0.3 mg/dL) from a visit three months earlier to the index visit. The definition of “unstable renal function” is the presence of an interval increase in SCr of 26.5 μmol/L (0.3 mg/dL) or more within the same duration. The cut-off value for sUCr/Osm was set at 7.07 when Cr is measured in μmol/L (0.08 when Cr is measured in mg/dL), which represents the 95% lower reference limit.1 A positive designation was given if the value of sUCr/Osm was not less than the 95% lower reference limit, and negative otherwise. The comparative test employed a tubular injury biomarker, urinary neutrophil gelatinase-associated lipocalin (UNGAL). The cut-off value for UNGAL was established at 100 ng/mL based on the maximum Youden's Index value. A positive result was defined as a UNGAL level less than or equal to 100 ng/mL, while a negative result was defined as a UNGAL level greater than 100 ng/mL. Results Of all the participants (N = 1, 570), 596 (38.0%) were female sex. The age (mean ± standard deviation) was 64.6 ± 13.0 years, the body mass index (BMI) was 30.1 ± 7.7 kg/m2, and the SCr level was 102.5 ± 51.4 μmol/L (1.16 ± 0.58 mg/dL). The sensitivity of sUCr/Osm in identifying “stable renal function” for all individuals was 0.926 (95% confidence interval: 0.912-0.939), with a positive predictive value of 0.906 (0.891-0.921) and an accuracy of 0.845 (0.827-0.863). These values of sUCr/Osm were either better or comparable to those of UNGAL- 0.867 (0.849-0.884), 0.917 (0.902-0.932), and 0.808 (0.788-0.827), respectively. In individuals (N = 521) with an estimated glomerular filtration rate (eGFR) of less than 60 cc/min/1.73m², sUCr/Osm also performed better or similarly to UNGAL. Table 1 provides direct comparisons between sUCr/Osm and UNGAL in identifying “stable renal function”. Conclusion To assess renal function, sUCr/Osm may be effectively used to infer the trend. Fig. 1 presents a flowchart for the corresponding management based on a single SCr-based eGFR and sUCr/Osm.

Outcomes of patients with COPD switching from multiple-inhaler to once-daily single-inhaler triple therapy in a real-world primary care setting in England: a retrospective pre-post cohort study

BackgroundCompared with multiple-inhaler triple therapy (MITT), single-inhaler triple therapy (SITT) with fluticasone furoate/umeclidinium/vilanterol (FF/UMEC/VI) demonstrated improved lung function and meaningful improvements in chronic obstructive pulmonary disease (COPD) Assessment Test score....

Environmental, social and governance assets and diversity scores: exploring their relationship to carbon emissions in global companies

PurposeThe interconnected challenges of climate change and social inclusivity have placed unprecedented pressure on businesses to adopt responsible practices. While previous research has explored the individual impacts of environmental, social, and governance (ESG) performance and diversity initiatives, there remains a dearth of comprehensive investigations into how these factors collectively influence carbon emission scores. Drawing on the legitimacy theory, we explore whether ESG and diversity scores predict global companies' carbon emission scores. As concerns about the environmental impact of businesses grow, understanding the relationships between ESG performance, diversity management, and carbon emissions becomes imperative for sustainable corporate practices.Design/methodology/approachThe primary dataset for this study includes 1,268 worldwide firm-year data for 2021. The sample is subjected to missing data examination as a component of the filtration process. Data preprocessing is performed before machine learning analysis, including verifying missing data. Our research resulted in the final sample, which includes 627 worldwide firm data from 2021. Data regarding all publicly traded companies was obtained from Refinitiv Eikon.FindingsOur findings showed that corporate carbon emission performance in global corporations is influenced by ESG performance and total diversity score.Originality/valueFirms involve in ESG as well as diversity practices to be able to achieve sustainable success. Yet, the forecasting of carbon emissions based on ESG scores and diversity scores remains inadequately established due to conflicting findings and enigmas prevalent in the literature.