Prediction Intervals Research Articles

Enhancing real‐time traffic volume prediction: A two‐step approach of object detection and time series modelling

AbstractA two‐step framework that integrates real‐time data collection with time series forecasting models for predicting traffic volume is proposed. In the first step, the framework utilizes live highway surveillance video data and YOLO‐v7 object detector to construct accurate traffic volume data. In the second step, an ARIMA–LSTM time series model is applied to forecast future traffic volumes. Experimental results show that YOLO‐v7 achieved a vehicle detection accuracy of over 93.30%, ensuring high precision in traffic volume data construction. The ARIMA–LSTM model demonstrated superior performance in traffic volume prediction, with a mean squared error of 87.97, root mean squared error of 10,388.57, and mean absolute error of 101.39. YOLO‐v7's detection speed of 7.8 ms per frame further validates the feasibility of real‐time data construction. The findings indicate that the combination of YOLO‐v7 for vehicle detection and ARIMA–LSTM for traffic prediction is highly effective, offering a significant reduction in training time compared to more complex deep learning models while maintaining high prediction accuracy. This research presents a unified solution for traffic data collection and prediction, enhancing transportation infrastructure planning and optimizing traffic flow. Future work will focus on extending the prediction intervals and further refining the models to improve performance.

Burden of rheumatic diseases among people with diabetes: A systematic review and meta-analysis

Diabetes mellitus type 2 (DM2) is a prevalent metabolic condition affecting over 500 million people globally and associated with serious comorbidities, including various rheumatologic conditions. Some studies have reported a significant association between rheumatological conditions and DM2. However, the global burden of rheumatological conditions among people with DM2 remains unknown. The aim of this study was to assess the cumulative prevalence of rheumatological conditions in DM2 patients. An extensive electronic search was conducted utilizing multiple databases of studies published until January 2024. The quality of the studies included in the review was evaluated using a modified version of the Newcastle-Ottawa Scale (NOS). The overall quality of the studies included was rated as moderate. The review included a total of 13 studies, with 830,649 DM2 patients reported to have rheumatological conditions. Eleven studies were used to determine the individual results of rheumatoid arthritis (RA), gout, and osteoarthritis (OA). The pooled prevalence of OA was 26% (95%CI: 19–32%) and the prediction interval ranged from 6% to 51%. The pooled prevalence of gout disease was 1% (95%CI: 0.0–5.0%), and the prediction interval ranged from 0% to 99%. For RA, the combined prevalence was 0.3% (95%CI: 0.2–0.5%) and the prediction interval ranged from 0% to 5%. In conclusion, this review suggests a considerable prevalence of OA among DM2 patients, while the prevalence of RA and gout was minimal. Early diagnosis and management of certain rheumatologic conditions in individuals with DM2 may help improve health outcomes and prevent premature mortality. Further research is warranted to explore and understand the mechanisms underlying the association between DM2 and rheumatological conditions.

Data-Driven State of Health Interval Prediction for Lithium-Ion Batteries

The accurate prediction of the state of health (SOH) for lithium-ion batteries is a key factor for improving the performance of battery management systems (BMS). However, traditional point prediction methods are difficult to effectively eliminate errors due to the uncertainty of variables and application environments. This paper presents a model for predicting the interval of lithium-ion batteries based on health indicators (HIs) during charging, which addresses the limitations of current point prediction in practical applications. First, twelve HIs are extracted from the current and voltage variables of the charging process. Secondly, feature selection is performed by random forest (RF) training, and the selected HIs are dimensioned using partial least squares (PLS). Finally, a long short-term memory network (LSTM) combined with quantile regression (QR) is used to derive the quantile values of the prediction points and each quantile is employed as input information for Gaussian kernel density estimation (KDE) to obtain the SOH probability density distribution. The experimental results based on the NASA PCOE Li-ion battery dataset and CALCE Li-ion battery dataset show that the SOH interval coverage is more than 90% and the average width of the interval is less than 0.294.

Forecasting changes in the earth’s climate system by instrumental measurements and paleodata in the phase-time region, consistent with changes in the barycentric motions of the sun. Part 1

The relevance of the research is due to the need to establish the root cause of climate change on Earth and predict changes in heliocosmic, climatic variables, natural disasters by intramental measurements and paleodata, as time series, for long intervals of the time horizon in the phase domain, taking into account their cyclicity and interdependence of changes. Purpose of research: application of the time series analysis method of the time series to establish the strength of the influence of the barycentric movements of the Sun on the variability of heliocosmic, climatic variables, natural disasters and fires, high-precision prediction of variables in the phase-time domain for long horizons of hundreds and thousands of years by instrumental measurements and paleodata. Objects of research: time series of changes in heliocosmic and climatic variables, natural disasters and fires, curves of changes in climatic changes obtained by analysis of ice cores in Antarctica and bottom sediments in Antarctica. Methods of research: wavelet phase-frequency and phase-time analysis of images of heliocosmic and climatic variables, natural disasters and fires; calculation of the consistency of changes in selected groups on a set of phase-time characteristics of variables in a sliding mode. Main results of research: In the images of wavelet phase-time functions of many variables constructed from observations in 1600–2010, Jose periods of ~178 years were found, which are contained in changes in the characteristics of barycentric movements of the Sun, solar activity, solar constant, CO2,N2OEl Nino, solar wind, the level of the Caspian Sea, temperature in Greenland, the speed of rotation of the Earth, snow accumulation rates in the Indian Ocean sector of Antarctica, global temperature; cyclicity of changes in heliocosmic and climatic variables are consistent with changes in Baricentre; there is a significant difference in the consistency of changes in the phase-time characteristics of variables obtained at the southern hot and northern cold latitudes of the planet by about 2.1 times; in the northern part of the planet, changes in variables are more chaotic, due to the different influence of changes in the Baricentre variable, the magnetic fields of the Sun and the Earth on the variability of variables at different latitudes of the planet; significant variability and resonances of the phase characteristics of heliocosmic and climatic variables are observed by changes in Baricentre on the graphs of coordinated changes in groups of variables in the observed and predicted time intervals. The studies reveal multiple oscillatory responses of the Earth’s climate system to the impact of the Baricentre variable due to the heterogeneity of its components and the Earth’s magnetosphere in space. Changes in the set of phase-time characteristics of variables on the same graph in the observed and predicted time intervals in the interval ± π are displayed as changes in autowaves characteristic of self-organizing systems, characterizing climatic changes in environments in combination with the influence of the variable Baricentre.

Crack propagation behavior and failure prediction of rocks with non-parallel conjugate flaws: Insights from the perspective of acoustic emission and DIC

Rock-like containing non-parallel conjugate flaws specimens (NPCFS) were prepared and uniaxial compression crack propagation tests were conducted employing acoustic emission (AE) and digital image correlation (DIC) techniques. The results show that: the angle of conjugate unilateral cracks increases, the average load-bearing capacity of the rock rises. Dominant frequencies from different conjugate defect angles exhibit distinct bimodal characteristics (low-frequency: 80–140 kHz, intermediate-frequency: 260–320 kHz). The concept of AE dominant frequency ratio β is proposed and introduced to quantify the strength of macro-scale failure of the specimens. The AE multiparameter is proposed to predict the failure load, and the mean value of failure load is predicted to be in the range of 86–93 % (failure load), the prediction interval of Ib value is in the range of 95–99 %, the prediction interval of critical slowing down variance value is in the range of 90–95 %, while the prediction interval of autocorrelation coefficient is in the range of 83–90 %. A favorable correlation was observed between AE events in NPCFS and surface deformations observed through DIC technology. Both ends of the conjugate flaw cannot propagate simultaneously; once one extends, the other ceases further expansion, forming a primary failure line that propagates towards the direction of the primary compressive stress.

Special applets for elucidating the difference between prediction intervals and confidence intervals—Why are arising ovals actually ellipses?

AbstractIn the last years special “ovals” appear increasingly often in diagrams and applets for discussing crucial items of statistical inference (when dealing with confidence intervals for an unknown probability p; approximation of the binomial distribution by the normal distribution; especially in German literature, see e.g. [Meyer, Stochastik in der Schule 2011; 31:18–22] or [Riemer, Statistik unterrichten. Eine handlungsorientierte Didaktik der Stochastik; 2023, p. 69ff]). These applets concerning confidence intervals aim at illustrating vividly the crucial differences (as we know, often vague for students) and connections between confidence intervals and prediction intervals. In English literature such applets seem to be not so well known, thus at the beginning, we describe how they work and highlight their high potential for teaching and learning. The fact that the mentioned ovals are, actually, ellipses is not important for stochastic purposes which is probably the reason why one can hardly find related explanations in the literature. Another aim of this paper is to explain the ellipse shape in a preferably elementary manner involving geometry, even spatial geometry.

Inference in mixed models with a mixture of distributions and controlled heteroscedasticity

. In the realm of mixed models, this article explores the estimation method for incorporating a mixture of distributions and addressing controlled heteroscedasticity. By relaxing the assumptions of normality and homoscedasticity, we introduce a more flexible approach to analyzing data. The article presents estimation techniques for variance components, estimable vectors, and cumulants, while also developing prediction intervals and prediction ellipsoids for future observations. A numerical example is employed to illustrate the method and compare it with traditional ANOVA and Bayesian estimation methods. The results demonstrate the superior flexibility and broader applicability of the proposed methods in diverse contexts. By extending the analysis beyond conventional assumptions, these approaches enhance the accuracy and robustness of statistical inference in mixed models.

How uncertainty in calibration data affects the modeling of non-point source pollutant loads in baseflow

Baseflow is a major transport pathway for non-point source (NPS) pollutants. Watershed water quality (WWQ) models calibrated by low-quality data may produce misleading predictions of baseflow NPS pollutant loads, resulting in poor management decisions. We evaluated how models of the baseflow nitrate loads in the Huron River basin, southwest of Lake Erie, were affected by uncertainty in the calibration data. Based on a five-year time series of daily streamflow, nitrate concentration, and specific conductance, two sets of “observed” baseflow nitrate load data that include uncertainty were estimated using various tracer-based and non-tracer-based hydrograph separation methods, in conjunction with assumptions regarding baseflow nitrate concentrations. We calibrated the Soil and Water Assessment Tool plus (SWAT+) model with the two “observed” data sets and used the Generalized Likelihood Uncertainty Estimation (GLUE) approach to quantify parameter and predictive uncertainties. The results showed that baseflow accounted for 26 %–34 % of the mean annual total streamflow (11.8 m3/s) and 8 %–37 % of the mean annual total nitrate load (14.3 kg·ha−1·year−1) in the Huron River basin. The baseflow and nitrate load estimates from the non-tracer-based methods resembled those from the tracer-based method but had greater uncertainty. The posterior parameter distributions, as well as the weighted means and 90 % prediction intervals of the simulated baseflow nitrate loads, exhibited minimal variation when different calibration data sets for SWAT+ and different threshold likelihood values for GLUE were used. Our analysis emphasizes the necessity of calibrating WWQ models with baseflow pollutant loads/concentrations when addressing water quality issues related to baseflow. It also demonstrates the feasibility of utilizing multiple non-tracer-based hydrograph separation methods to estimate baseflow NPS pollutant loads. These non-tracer-based methods offer a simplicity and broader applicability compared to tracer-based methods. This study has provided insights into how calibration data uncertainty impacts the modeling of NPS pollution in baseflow and highlights the practical value of non-tracer-based hydrograph separation methods.

A machine learning-based probabilistic wind speed prediction model with multi-resolution data, quantile regression and bound estimation

Wind speed prediction can be important for management of long-span cross-sea bridges, and probabilistic prediction models are needed to take uncertainty of future wind speed into consideration. However, there are two shortcomings of existing models. One is the lack of supplementary input of detailed fluctuations of wind speed, and the other is that fixed quantiles are usually selected when the predicted probability distribution is converted into a confidence interval. To this end, we propose a novel machine learning-based probabilistic model with multi-resolution data, quantile regression and bound estimation. The proposed model is composed of the distribution prediction part and the interval prediction part. In the distribution prediction part, we employ multi-resolution data as input, which can complement details of fluctuations of wind speed to reduce uncertainty. The interval prediction part employs the lower upper bound estimation method to postprocess predicted quantiles and output a superior interval than selecting fixed quantiles. The proposed model is validated on the monitoring wind speed data of two long-span cross-sea bridges, where the superiority of the distribution part and the effectiveness of the interval part are demonstrated.

Prediction intervals for future Pareto record claims

AbstractStochastic models and methods for quantifying extreme events are of interest in numerous disciplines. Within this paper, statistical prediction of extreme claims or losses in insurance industry is considered based on upper record values, which describe successively largest observations in a sequence of data over time. The problem of predicting a future record value (here in particular, a future record claim) based on a sequence of previously observed record values (here, past record claims) is addressed by means of prediction intervals. For an underlying Pareto distribution, respective exact and approximate intervals from the literature are summarized and modified and new ones are developed. In a simulation study, these prediction intervals are evaluated and compared regarding coverage frequency and length. The impact of the number of observed record values as well as the choice of the Pareto distribution is discussed. In the case of a small number of record values, the use of k-th record values is considered as an option for statistical analyses to predict, e.g., second largest record claims. Selected prediction methods are applied to several real data sets, which turn out to perform well and to be able to capture the magnitude of future record claims, even for fairly small numbers of record observations. For comparison, generalized Pareto distributions are fitted to real data sets and a corresponding point predictor as well as a respective upper prediction interval for the next record to appear are derived and evaluated.

Forecasting school enrollments in the Australian Capital Territory

Abstract School enrollment forecasts are vital for effective planning. This study introduces a probabilistic multiregional population projection model, which accounts for different components, including preschool entries, migration, grade progression, and graduations. Using different distributions with a cohort component projection and Monte Carlo simulation, this paper forecasts student enrollments for each school and academic year level in the Australian Capital Territory based on the annual record-level administrative data. In the in-sample validation tests, the model’s overall performance is robust, and the probabilistic design offers reliable prediction intervals reflecting the variation in observed values. The paper ends with a discussion on the importance of prediction intervals for informing school planning.

Uncertainty-Aware Seismic Signal Discrimination using Bayesian Convolutional Neural Networks

Seismic signal classification plays a crucial role in mitigating the impact of seismic events on human lives and infrastructure. Traditional methods in seismic hazard assessment often overlook the inherent uncertainties associated with the prediction of this complex geological phenomenon. This work introduces a probabilistic framework that leverages Bayesian principles to model and quantify uncertainty in seismic signal classification by applying a Bayesian Convolutional Neural Network (BCNN). The BCNN was trained on a dataset that comprises waveforms detected in the Southern California region and achieved an accuracy of 99.1%. Monte Carlo Sampling subsequently creates a 95% prediction interval for probabilities that considers epistemic and aleatoric uncertainties. The ability to visualize both aleatoric and epistemic uncertainties provides decision-makers with information to determine the reliability of seismic signal classifications. Further, the use of Bayesian CNN for seismic signal classification provides a more robust foundation for decision-making and risk assessment in earthquake-prone regions.

Frequency of red blood cell allo-immunization in transfused patients with sickle cell disease in Africa: a systematic review with meta-analysis

Background and Objectives: Blood transfusion is an effective and proven treatment for some severe complications of sickle cell disease. Recurrent transfusions have put patients with sickle cell disease at risk of developing antibodies against the various antigens they were exposed to. This study aims to investigate the frequency of red blood cell alloimmunization in patients with sickle disease in Africa. Materials and Methods: This is a systematic review of peer-reviewed and published literature. The review was conducted consistent with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses checklist.Data sources for the review include MEDLINE, PubMed, AJOL, CINAHL, Psych-Info and Academic Search Complete. Included in this review are articles that reported the frequency/prevalence of red blood cell alloimmunization in sickle cell disease patients in Africa. Eligible studies were subjected to independent full-text screening and data extraction. Risk of bias assessment was conducted with the aid of the mixed method appraisal tool. We employed a random-effects model of meta-analysis to estimate the pooled prevalence. We computed Cochrane’s Q statistics and I2 and prediction interval to quantify heterogeneity in effect size. Results: The prevalence estimates range from 2.6% to 29%. Pooled prevalence was estimated to be 12.1% (95% CI 8.1 to 17.7), with significant heterogeneity (I2= 91.83; PI = 2 to 54%). Conclusion: The frequency of red cell alloantibody varies considerably in Africa. Keywords: Systematic review; red blood cell; alloimmunization; sickle cell disease; Africa.

7166 EFFECT OF ANTI-NEOPLASTIC THERAPIES ON CARCINOMA AND AGGRESSIVE PITUITARY TUMORS : A SYSTEMATIC REVIEW AND META-ANALYSIS

Abstract Disclosure: V.D. Nunes-Nogueira: None. A.R. Cardoso: None. A.C. Zimmermann: None. Introduction: Approximately 10% of the pituitary adenomas can have an aggressive behavior and 0.2% can present metastasis, being classified as pituitary carcinomas. Objective: Evaluate the tumor control obtained by anti-neoplastic therapies that are still little explored in the context of carcinoma and aggressive pituitary tumors. Methods: A systematic review according to the Joanna Briggs Institute has been conducted. We have included studies that evaluated patients with carcinoma or aggressive pituitary tumors who underwent treatment with at least one of the following therapies: Temozolomide, VEGF Inhibitors, Peptide Receptor Radionuclide Therapy (PRRT), Everolimus, Immune Checkpoint Inhibitors, EGFR Inhibitors, Capecitabine and temozolomide (CAPTEM). The main outcome assessed by the study was the frequency of tumor response according to the RECIST protocol classification. Embase, Medline, LILACS, and CENTRAL were our source databases. Proportional meta-analyses were performed to calculate the overall frequencies of complete remission (CR), partial remission (PR; minimum shrinkage of 30%), stable disease (SD) and progression disease (PD). We used Stata Software 18, the updated command “metaprop_one” and fit the logistic normal random effect model. Results: None clinical trial was found, and sixty case report/case series were included. The most frequent therapy used was temozolomide (101 patients), followed by PRRT, pembrolizumab, lapatinib, CAPTEM, bevacizumab, ipilimumab/ nivolumab and everolimus, with respectively 9, 6 ,6, 6, 4, 3 and 1 individuals. The majority of patient in temozolomide studies had prolactinomas (40/101), followed by Cushing’s disease (25/101), the overall frequency of CR with this drug was 1% (ch^2=2.7; p=0.049, Predictive Interval (PI) 1% to 93%, 7 events), PR was 45% ( ch^2=1.8, p=0.091, PI= 8-89%, 46 events), SD 14% (CI 95%=5 a 33%, 20 events), and PD was 25% (95%CI: 14-40%, 28 events). Two patients receiving CAPTEM presented CR, no other studies presented this outcome. In the PRRT studies, the overall frequency of PR was 22% (95%CI 6 to 58%) and PD was 44% (95%CI 18 to 75).Conclusion: Temozolomide has been the therapy most frequent used in the management of carcinoma and aggressive pituitary tumor, being PR the most frequent outcome, followed by PD and SD, with low percentage of CR. The other drugs approached by this review showed similar poor response although they were less tested in the context of this neoplasms. Presentation: 6/2/2024

8699 ​​Efficacy of Glucagon-like Peptide-1 Analogs In Women With Polycystic Ovary Syndrome: A Systematic Review and Meta-Analysis of Randomized Clinical Trials

Abstract Disclosure: F.A. Kelly: None. A.D. Lôbo: None. I.P. da Silva: None. J.C. Cardoso: None. M.S. Barros: None. F.D. Pessôa: None. M.G. Leite: None. R.Y. Ura Sudo: None. A.M. De Almeida: None. F.A. Moraes: None. V. Morbach: None. M.C. Souza: None. P.G. Lima: None. M.P. Lima: None. I.B. Andrade: None. M.B. Jardine: None. L.M. Lopes: None. Polycystic ovary syndrome (PCOS) affects one in ten women of reproductive age, painting a complex picture of metabolic dysfunction. Its hallmark triad includes irregular menstrual cycles, androgen excess, and polycystic ovaries. Beyond infertility concerns, PCOS carries a heavy burden of metabolic imbalances, often characterized by insulin resistance, dyslipidemia, and chronic low-grade inflammation. These intertwined factors place women with PCOS at increased risk for type 2 diabetes and cardiovascular disease, posing a significant long-term health challenge. Current first-line treatments like metformin focus mainly on managing insulin resistance and regulating menstrual cycles. However, unmet needs remain in addressing the broader metabolic derangements and preventing associated health complications. In light of this, glucagon-like peptide-1 (GLP-1) analogs have emerged as potential game-changers in the PCOS treatment landscape. Their multifaceted effects on metabolism beyond blood sugar regulation offer a promising avenue for optimizing care and improving long-term health outcomes in women with PCOS. This meta-analysis aims to evaluate the use of GLP-1 analogs on patients with PCOS to reduce body mass index (BMI), glycated hemoglobin (HbA1C), fasting blood glucose (FBG), high density lipid (HDL), low density lipid (LDL) and triglyceride. PubMed, Embase, and Cochrane databases were systematically searched for randomized controlled trials (RCT) comparing the use of GLP-1 analogs to placebo or other therapies in patients with PCOS. Treatment effects for continuous endpoints were pooled through mean differences (MD) with 95% confidence intervals. Heterogeneity was evaluated with the Cochran Q test, the prediction interval and I² statistics. Statistical analysis was performed in Revman 5.4. The results were reported following the Preferred Reporting Items for Systematic Review (PRISMA) guidelines.A total of 4 RCTs with 228 patients were included, of whom 137 were randomized to GLP-1 therapy, 58 to placebo, and 33 to other treatments. Over a mean follow-up time of 24 weeks, patients submitted to GLP-1 analogs presented significant decreases in HbA1C (MD -1.40%; 95% CI -1.64 to -1.16; p<000001; I²=0% ), FBG (MD -4.40 mg/dL; 95% CI -7.63 to -1.17; p=0.008; I²=0%) and HDL-c (MD -0.82 mg/dL; 95% CI -1.33 to -0.31; p=0.002; I²=3%); and a nonsignificant increase in LDL-c (MD 0.37 mg/dL; 95% CI -1.30 to 2.03; p=0.67; I²=0%); and nonsignificant decrease in BMI (MD -1.28 Kg/m²; 95% CI -3.18 to 0.63; p=0.19; I²=98%), and triglyceride (MD -10.50 mg/dL; 95% CI -22.81 to 1.81; p=0.09; I²=0%).In this meta-analysis of RCTs of patients with PCOS, the use of GLP-1 analogs showed a significant reduction in HbA1C, FBG, and HDL-c. A larger sample size of people are needed for further statistical analyses. Presentation: 6/3/2024

An LSTM-based predictive monitoring method for data with time-varying variability

The recurrent neural network and its variants have shown great success in processing sequences in recent years. However, these neural networks have not aroused much attention in the process monitoring literature. Whereas the flexibility and outstanding performance of these models enable them to be solutions to many problem settings within process monitoring. One such problem setting is process data with inherent time-varying variability, an aspect of data that is often ignored. Therefore, this paper proposes a prospective monitoring method for detecting structural changes in data with inherent time-varying variability. We integrate long short-term memory (LSTM) prediction intervals with a neural network (NN) for variability prediction to detect structural changes in data online. We compare our proposed method to benchmarks in a simulation study, showing that the proposed model outperforms other NN-based predictive monitoring methods for mean shift detection. The proposed method is also applied in two case studies to sensor data, which confirms that the proposed method is an effective technique for detecting shifts.

7941 Comparison of Metformin with Inositol versus Metformin Alone in Women with Polycystic Ovarian Syndrome (PCOS): A Systematic Review and Meta Analysis of Randomized Controlled Trials

Abstract Disclosure: F.A. Kelly: None. A.D. Lôbo: None. I.B. Andrade: None. P.G. Lima: None. I.P. da Silva: None. M.C. Souza: None. V. Morbach: None. M.S. Barros: None. F.A. Moraes: None. M.B. Jardine: None. L.M. Lopes: None. M.P. Lima: None. M.G. Leite: None. R.Y. Ura Sudo: None. J.C. Cardoso: None. A.M. De Almeida: None. F.D. Pessôa: None. Metformin was the first medication targeting insulin resistance in PCOS, and it has been extensively studied as a metabolic treatment option. It addresses the chronic inflammatory state of PCOS and improves ovulation, metabolic parameters, and fertility.In recent years, inositols, specifically myo-inositol, and D-chiro-inositol, have emerged as potential treatment options for PCOS. These compounds act as a messenger for hormones, including insulin and follicle-stimulating hormones. Inositols effectively address insulin resistance and modulate the insulin signaling pathways.Although inositols have recently been included in the recommendations of the European Journal of Endocrinology's 2023 International evidence-based guideline for improvement in metabolic measures and considered with benefits in ovulation, hirsutism, and weight in patients with PCOS, the confidence in the available evidence supporting their use is considered to be limited. We aimed to perform a meta-analysis to understand the role of inositol, on glycemic control, LH/FSH ratio, hirsutism, HOMA-IR, and body mass index (BMI) in individuals with polycystic ovary syndrome (PCOS).PubMed, Embase, and Cochrane databases were systematically searched for RCTs comparing the use of metformin alone versus metformin and inositol in patients with polycystic ovary syndrome. Statistical analysis was performed in R software 4.3.1. A random-effects model was used to calculate the risk ratios (RRs) and mean differences (MDs) with 95% confidence intervals (CIs). A p-value of < 0.05 was considered statistically significant. Heterogeneity was examined with the Cochran Q test, prediction interval, and I² statistics. A total of 4 RCTs and 277 patients were included, of whom 138 were randomized to metformin alone and 139 to metformin plus myo-inositol. Compared with metformin alone, Myo-inositol plus Metformin significantly reduced hirsutism (MD -0.97; 95% CI -1.53; -0.40; p=0.0008) and LH/FSH ratio (MD -0.15; 95% CI -0.26, -0.03; p=0.02). There was no significant difference between groups for BMI (MD: 0.97; 95% CI -0.37, 2.32; p=0.16) and HOMA-IR (MD -0.17; 95% CI -0.35, 0.01; p=0.07).In this meta-analysis of RCTs of patients with polycystic ovary syndrome, inositol was associated with a reduction in hirsutism and LH/FSH ratio, compared with metformin monotherapy. Further studies are needed to clarify whether there are benefits for the use of inositol in PCOS treatment. Presentation: 6/3/2024

Efficient prediction uncertainty quantification in dam behavior monitoring with attention-based sequence-to-sequence learning

Displacement is an intuitive monitoring indicator of dam structural behavior. Conventional deterministic modeling methods frequently disregard the inherent uncertainty associated with monitoring data. This study aims to quantify such uncertainties within dam displacement predictions by augmenting a sequence-to-sequence structure with the novel recurrent unit, termed as tiny gated unit (TGU), attention mechanism, as well as quantile regression, resulting in the Att-S2STQ model. Specifically, TGU adopts only one gate to identify nonlinearity in data sequences, while Bahdanau attention mechanism dynamically assigns weights to different parts of the input sequence. The inclusion of quantile regression allows Att-S2STQ to produce prediction intervals (PIs), from which the probability density functions (PDFs) and cumulative distribution functions (CDFs) are further derived via kernel density estimation. PIs, PDFs, and CDFs jointly serve to reveal the prediction uncertainty. The effectiveness of the model is illustrated through comparative experiments using real-world dam monitoring datasets. Results indicate that the proposed model outperforms traditional methods in point, interval, and probability predictions, while also having a simpler structure and faster training. The superiority of both accuracy and efficiency makes it a valuable tool for dam management, aiding in data-driven decision-making and enhancing operational safety.

Research on methane Hazard interval prediction method based on hybrid “model-data”driven strategy

Safe mining of coal has an important impact on energy security, while effective control of methane hazard is the key to ensuring safe coal mining. Methane concentration is the main factor determining the hazard of methane in coal mines, and in order to limit the impact of methane on coal mine safety, this study proposes a methane concentration interval prediction method based on a hybrid “model-data” driven idea. Firstly, by analyzing the data and constructing a methane concentration prediction method based on model-driven, which reduces the influence of multicollinearity in the methane concentration series on the prediction effect, and then, in combination with the deep learning technique, a method based on the Wasserstein distance to improve the Informer model is proposed, and finally a hybrid-driven methane concentration interval prediction model is established by introducing the IOWGA operator and the statistical method. After an example analysis of a coal mine in Guizhou Province, China, the hybrid-driven model proposed in this study has better applicability and prediction accuracy in the methane concentration prediction task, which can effectively prevent the occurrence of coal mine accidents and is more in line with the needs of coal mine safety production.

Effectiveness of osteopathic craniosacral techniques: a meta-analysis.

Craniosacral osteopathic manipulative medicine-also known as craniosacral therapy (CST)-is a widely taught and used component of osteopathic medicine. This paper seeks to systematically review and conduct a meta-analysis of randomized controlled trials assessing the clinical effectiveness of CST compared to standard care, sham treatment, or no treatment in adults and children. A search of Embase, PubMed, and Scopus was conducted on 10/29/2023 and updated on 5/8/2024. There was no restriction placed on the date of publication. A Google Scholar search was conducted to capture grey literature. Backward citation searching was also implemented. All randomized controlled trials employing CST for any clinical outcome were included. Studies not available in English as well as studies that did not report adequate data were excluded. Multiple reviewers were used to assess for inclusions, disagreements were settled by consensus. PRISMA guidelines were followed in the reporting of this meta-analysis. Cochrane's Risk of Bias 2 tool was used to assess for risk of bias. All data were extracted by multiple independent observers. Effect sizes were calculated using a Hedge's G value (standardized mean difference) and aggregated using random effects models. The GRADE system was used to assess quality of evidence. The primary study outcome was the effectiveness of CST for selected outcomes as applied to non-healthy adults or children and measured by standardized mean difference effect size. Twenty-four RCTs were included in the final meta-analysis with a total of 1,613 participants. When subgroup analyses were performed by primary outcome only, no significant effects were found. When secondary outcomes were included in subgroup analyses, results showed that only Neonate health, structure (g = 0.66, 95% CI [0.30; 1.02], Prediction Interval [-0.73; 2.05]) and Pain, chronic somatic (g = 0.34, 95% CI [0.18; 0.50], Prediction Interval [-0.41; 1.09]) show reliable, statistically significant effect. However, these should not be interpreted as positive results as wide prediction intervals, high bias, and statistical limitations temper the real-world implications of this finding. CST demonstrated no significant effects in this meta-analysis, indicating a lack of usefulness in patient care for any of the studied indications.Pre-registration available at https://doi.org/10.17605/OSF.IO/54K6G. https://osf.io/57t4w.