Median Absolute Percentage Error Research Articles

Automated calculation and reporting of vancomycin area under the concentration-time curve: a simplified single-trough concentration-based equation approach.

Vancomycin, a crucial antibiotic for Gram-positive bacterial infections, requires therapeutic drug monitoring (TDM). Contemporary guidelines advocate for AUC-based monitoring; however, using Bayesian programs for AUC estimation poses challenges. We aimed to develop and evaluate a simplified AUC estimation equation using a steady-state trough concentration (Ctrough) value. Utilizing 1,034 TDM records from 580 general hospitalized patients at a university-affiliated hospital in Ulsan, we created an equation named SSTA that calculates the AUC by applying Ctrough, body weight, and single dose as input variables. External validation included 326 records from 163 patients at a university-affiliated hospital in Seoul (EWUSH) and literature data from 20 patients at a university-affiliated hospital in Bangkok (MUSI). It was compared with other AUC estimation models based on the Ctrough, including a linear regression model (LR), a sophisticated model based on the first-order equation (VancoPK), and a Bayesian model (BSCt). Evaluation metrics, such as median absolute percentage error (MdAPE) and the percentage of observations within ±20% error (P20), were calculated. External validation using the EWUSH data set showed that SSTA, LR, VancoPK, and BSCt had MdAPE values of 6.4, 10.1, 6.6, and 7.5% and P20 values of 87.1, 82.5, 87.7, and 83.4%, respectively. External validation using the MUSI data set showed that SSTA, LR, and VancoPK had MdAPEs of 5.2, 9.4, and 7.2%, and P20 of 95, 90, and 95%, respectively. Owing to its decent AUC prediction performance, simplicity, and convenience for automated calculation and reporting, SSTA could be used as an adjunctive tool for the AUC-based TDM.

Accuracy of smartwatches for the remote assessment of exercise capacity

Exercise capacity is a strong independent predictor of cardiovascular and all-cause mortality. The utilization of well-established submaximal tests of exercise capacity such as the 6-min walk test (6MWT), 3-min step test (3MST) and 10-chair rise test (10CRT) in the community would improve patient care but requires remote monitoring technology. Consumer grade smartwatches provide such an opportunity, however, their accuracy in measuring physiological responses to these tests is unclear. The aim of this study was to determine the accuracy of consumer grade smartwatches in assessing exercise capacity to develop a framework for remote, unsupervised testing. 16 healthy adults (7 male (44%), age median 27 [interquartile range (IQR) 26,29] years) performed 6MWTs using two protocols: (1) standard—straight 30 m laps (6MWT-standard) and 2) continuous lap—circular 240 m laps around a park (6MWT-continuous lap), 3MSTs and 10CRTs. Each one of these four tests was performed three times across two clinic visits. Each participant was fitted with a Garmin Vivoactive4 and Fitbit Sense smartwatch to measure three parameters: distance, step counts and heart rate (HR) response. Reference measures were a meter-wheel, hand tally counter and ECG, respectively. Mean HR was measured at rest, peak exercise and recovery. Agreement was measured using Bland–Altman analysis for repeated measures and summarized as median absolute percentage errors (MAPE). Distance during 6MWT-continuous lap had better agreement than during 6MWT-standard for both Garmin (MAPE: 6.4% [3.0, 10.4%] versus 20.1% [13.9, 28.4%], p < 0.001) and Fitbit (8.0% [2.9, 10.1% versus 18.8% [15.2, 28.1%], p < 0.001). Garmin measured step count more accurately than Fitbit (MAPE: 1.8% [0.9, 2.9%] versus 8.0% [2.6, 12.3%], p < 0.001). Irrespective of test, both devices showed excellent accuracy in measuring HR at rest and recovery (≤ 3%), while accuracy decreased during peak exercise (Fitbit: ~ 12% and Garmin: ~ 7%). In young adults without mobility difficulties, exercise capacity can be measured remotely using standardized tests and consumer grade smartwatches.

Detection of sedentary time and bouts using consumer-grade wrist-worn devices: a hidden semi-Markov model

BackgroundWrist-worn data from commercially available devices has potential to characterize sedentary time for research and for clinical and public health applications. We propose a model that utilizes heart rate in addition to step count data to estimate the proportion of time spent being sedentary and the usual length of sedentary bouts.MethodsWe developed and trained two Hidden semi-Markov models, STEPHEN (STEP and Heart ENcoder) and STEPCODE (STEP enCODEr; a steps-only based model) using consumer-grade Fitbit device data from participants under free living conditions, and validated model performance using two external datasets. We used the median absolute percentage error (MDAPE) to measure the accuracy of the proposed models against research-grade activPAL device data as the referent. Bland-Altman plots summarized the individual-level agreement with activPAL.ResultsIn OPTIMISE cohort, STEPHEN’s estimates of the proportion of time spent sedentary had significantly (p < 0.001) better accuracy (MDAPE [IQR] = 0.15 [0.06–0.25] vs. 0.23 [0.13–0.53)]) and agreement (Bias Mean [SD]=-0.03[0.11] vs. 0.14 [0.11]) than the proprietary software, estimated the usual sedentary bout duration more accurately (MDAPE[IQR] = 0.11[0.06–0.26] vs. 0.42[0.32–0.48]), and had better agreement (Bias Mean [SD] = 3.91[5.67] minutes vs. -11.93[5.07] minutes). With the ALLO-Active dataset, STEPHEN and STEPCODE did not improve the estimation of proportion of time spent sedentary, but STEPHEN estimated usual sedentary bout duration more accurately than the proprietary software (MDAPE[IQR] = 0.19[0.03–0.25] vs. 0.36[0.15–0.48]) and had smaller bias (Bias Mean[SD] = 0.70[8.89] minutes vs. -11.35[9.17] minutes).ConclusionsSTEPHEN can characterize the proportion of time spent being sedentary and usual sedentary bout length. The methodology is available as an open access R package available from https://github.com/limfuxing/stephen/. The package includes trained models, but users have the flexibility to train their own models.

Analyzing the impact of investor sentiment on S&amp;P 500 prices using deep learning models

Purpose This study aims to investigate to what extent the predictability of the standard and poor’s 500 (S&amp;P 500) price levels is enhanced by investors’ sentiments extracted from social media content, specifically platform X. Design/methodology/approach Two recurrent neural network (RNN) models are developed. The first RNN model is merely based on historical records and technical indicators. In addition to the variables included in the first RNN model, the second RNN model comprises the outputs of the sentiment analysis, performed using the TextBlob library. The study was conducted between December 28, 2011, and December 30, 2021, over 10 years, to obtain better results by feeding the RNN models with a significant quantity of data by extending the period and capturing an extensive timespan. Findings Comparing the performance of both models reveals that the second model, with sentiment analysis inputs, yields superior outcomes. The mean absolute error (MAE) of the second model registered 72.44, approximately 50% lower than the MAE of the technical model, its percentage value, the mean absolute percentage error, recorded 2.16%, and finally, the median absolute percentage error reached a value of 1.30%. This underscores the significant influence of digital platforms in influencing the behavior of certain assets like the S&amp;P 500, emphasizing the relevance of sentiment analysis from social media in financial forecasting. Originality/value This study contributes to the growing body of literature by highlighting the enhanced predictive power of deep learning models that incorporate investor sentiment from social media, thereby advancing the application of behavioral finance in financial forecasting.

Deep learning-based automated bone age estimation for Saudi patients on hand radiograph images: a retrospective study

PurposeIn pediatric medicine, precise estimation of bone age is essential for skeletal maturity evaluation, growth disorder diagnosis, and therapeutic intervention planning. Conventional techniques for determining bone age depend on radiologists’ subjective judgments, which may lead to non-negligible differences in the estimated bone age. This study proposes a deep learning-based model utilizing a fully connected convolutional neural network(CNN) to predict bone age from left-hand radiographs.MethodsThe data set used in this study, consisting of 473 patients, was retrospectively retrieved from the PACS (Picture Achieving and Communication System) of a single institution. We developed a fully connected CNN consisting of four convolutional blocks, three fully connected layers, and a single neuron as output. The model was trained and validated on 80% of the data using the mean-squared error as a cost function to minimize the difference between the predicted and reference bone age values through the Adam optimization algorithm. Data augmentation was applied to the training and validation sets yielded in doubling the data samples. The performance of the trained model was evaluated on a test data set (20%) using various metrics including, the mean absolute error (MAE), median absolute error (MedAE), root-mean-squared error (RMSE), and mean absolute percentage error (MAPE). The code of the developed model for predicting the bone age in this study is available publicly on GitHub at https://github.com/afiosman/deep-learning-based-bone-age-estimation.ResultsExperimental results demonstrate the sound capabilities of our model in predicting the bone age on the left-hand radiographs as in the majority of the cases, the predicted bone ages and reference bone ages are nearly close to each other with a calculated MAE of 2.3 [1.9, 2.7; 0.95 confidence level] years, MedAE of 2.1 years, RMAE of 3.0 [1.5, 4.5; 0.95 confidence level] years, and MAPE of 0.29 (29%) on the test data set.ConclusionThese findings highlight the usability of estimating the bone age from left-hand radiographs, helping radiologists to verify their own results considering the margin of error on the model. The performance of our proposed model could be improved with additional refining and validation.

Groundwater quality assessment using machine learning models: a comprehensive study on the industrial corridor of a semi-arid region.

Water plays a significant role in sustaining the lives of humans and other living organisms. Groundwater quality analysis has become inevitable, because of increased contamination of water resources and global warming. This study used machine learning (ML) models to predict the water quality index (WQI) and water quality classification (WQC). Forty groundwater samples were collected near the Ranipet industrial corridor, and the hydrogeochemistry and heavy metal contamination were analyzed. WQC prediction employed random forest (RF), gradient boosting (GB), decision tree (DT), and K-nearest neighbor (KNN) models, and WQI prediction used extreme gradient boosting (XGBoost), support vector regressor (SVR), RF, and multi-layer perceptron (MLP) models. The grid search method is used to evaluate the ML model by F1 score, accuracy, recall, precision, and Matthews correlation coefficient (MCC) for WQC and the coefficient of determination (R2), mean absolute error (MAE), mean square error (MSE), and median absolute percentage error (MAPE) for WQI. The WQI results indicate that the groundwater quality of the study area is very poor and unsuitable for drinking or irrigation purposes. The performance metrics of the RF model excelled in predicting both WQC (accuracy = 97%) and WQI (R2 = 91.0%), outperforming other models and emphasizing ML's superiority in groundwater quality assessment. The findings suggest that ML models perform well and yield better accuracy than conventional techniques used in groundwater quality assessment studies.

Subgroup identification-based model selection to improve the predictive performance of individualized dosing.

Currently, model-informed precision dosing uses one population pharmacokinetic model that best fits the target population. We aimed to develop a subgroup identification-based model selection approach to improve the predictive performance of individualized dosing, using vancomycin in neonates/infants as a test case. Data from neonates/infants with at least one vancomycin concentration was randomly divided into training and test dataset. Population predictions from published vancomycin population pharmacokinetic models were calculated. The single best-performing model based on various performance metrics, including median absolute percentage error (APE) and percentage of predictions within 20% (P20) or 60% (P60) of measurement, were determined. Clustering based on median APEs or clinical and demographic characteristics and model selection by genetic algorithm was used to group neonates/infants according to their best-performing model. Subsequently, classification trees to predict the best-performing model using clinical and demographic characteristics were developed. A total of 208 vancomycin treatment episodes in training and 88 in test dataset was included. Of 30 identified models from the literature, the single best-performing model for training dataset had P20 26.2-42.6% in test dataset. The best-performing clustering approach based on median APEs or clinical and demographic characteristics and model selection by genetic algorithm had P20 44.1-45.5% in test dataset, whereas P60 was comparable. Our proof-of-concept study shows that the prediction of the best-performing model for each patient according to the proposed model selection approaches has the potential to improve the predictive performance of model-informed precision dosing compared with the single best-performing model approach.

A novel hybrid modelling framework for GPP estimation: Integrating a multispectral surface reflectance based [formula omitted] simulator into the process-based model

Terrestrial gross primary productivity (GPP) is the key element in the carbon cycle process. Accurate GPP estimation hinges on the maximum carboxylation rate (Vcmax,025). The high uncertainty in deriving ecosystem-level Vcmax,025 has long hampered efforts toward the performance of the GPP model. Recently studies suggest the strong relationship between spectral reflectance and Vcmax,025. We proposed the multispectral surface reflectance-driven Vcmax,025 simulator using the fully connected deep neural network and built the hybrid modelling framework for GPP estimation by integrating the data-driven Vcmax,025 simulator in the process-based model. The performance of hybrid GPP model was evaluated at 95 flux sites. The result shows that the multispectral surface reflectance-driven Vcmax,025 simulator acquires the satisfactory estimation, with correlation coefficient (R), root mean square error (RMSE) and median absolute percentage error (MdAPE) ranging from 0.34 to 0.80, 14 to 43 μmol m−2 s−1 and 21 % to 66 % across different land cover types, respectively. The hybrid framework generates good GPP estimates with R, RMSE and MdAPE varying from 0.76 to 0.89, 1.79 to 6.16 μmol m−2 s−1 and 27 % to 90 %, respectively. Compared with EVI-driven method, the multispectral surface reflectance significantly improves the Vcmax,025 and GPP estimates, with MdAPE declining by 0.6 %–18 % and 1 % to 21 %, respectively. The Shapley value analysis reveals that red (620–670 nm), near-infrared (841–876 nm) and shortwave infrared (1628–1652 nm and 2105–2155 nm) are the key bands for Vcmax,025 estimation. This study highlights the potential of multispectral surface reflectance for quantifying ecosystem-level Vcmax,025. The new hybrid framework fully extracts the information of all available spectral bands using deep learning to reduce parameter uncertainty while maintains the description of photosynthetic process to ensure its physical reasonability. It can serve as a powerful tool for accurate global GPP estimation.

Forecasting daily COVID-19 cases with gradient boosted regression trees and other methods: evidence from U.S. cities.

There is a vast literature on the performance of different short-term forecasting models for country specific COVID-19 cases, but much less research with respect to city level cases. This paper employs daily case counts for 25 Metropolitan Statistical Areas (MSAs) in the U.S. to evaluate the efficacy of a variety of statistical forecasting models with respect to 7 and 28-day ahead predictions. This study employed Gradient Boosted Regression Trees (GBRT), Linear Mixed Effects (LME), Susceptible, Infectious, or Recovered (SIR), and Seasonal Autoregressive Integrated Moving Average (SARIMA) models to generate daily forecasts of COVID-19 cases from November 2020 to March 2021. Consistent with other research that have employed Machine Learning (ML) based methods, we find that Median Absolute Percentage Error (MAPE) values for both 7-day ahead and 28-day ahead predictions from GBRTs are lower than corresponding values from SIR, Linear Mixed Effects (LME), and Seasonal Autoregressive Integrated Moving Average (SARIMA) specifications for the majority of MSAs during November-December 2020 and January 2021. GBRT and SARIMA models do not offer high-quality predictions for February 2021. However, SARIMA generated MAPE values for 28-day ahead predictions are slightly lower than corresponding GBRT estimates for March 2021. The results of this research demonstrate that basic ML models can lead to relatively accurate forecasts at the local level, which is important for resource allocation decisions and epidemiological surveillance by policymakers.

Comparison of the Karney Polygon Method and the Shoelace Method for Calculating Area

In calculating the area of an area, latitude and longitude coordinates are based on data from Global Administrative Region Database and Google Earth can be used. The aim of this research is to calculate the area. This research uses the Karney and Shoelace method to determine its accuracy based on Median Absolute Percentage Error in calculating the area of an area. Median Absolute Percentage Error results use data based on Global Administration The Regional Database by applying the polygon method proposed by Karney is 18.73%, and the percentage is 18.19% by applying the Shoelace method. Based on Google Earth data, implementation the method proposed by Karney obtained a percentage of 19.14%, and the application of shoelaces method obtained a percentage of 19.72%. In this case, Karney polygons and the Shoelace method has good accuracy because the value is below 20%. The proposed Shoelace method is easier to perform understand compared to the Karney method for calculating land area because it uses the Universal Transverse Mercator coordinate system, which projects points on the Earth's surface onto a flat plane.

A novel bidirectional LSTM deep learning approach for COVID-19 forecasting

COVID-19 has resulted in significant morbidity and mortality globally. We develop a model that uses data from thirty days before a fixed time point to forecast the daily number of new COVID-19 cases fourteen days later in the early stages of the pandemic. Various time-dependent factors including the number of daily confirmed cases, reproduction number, policy measures, mobility and flight numbers were collected. A deep-learning model using Bidirectional Long-Short Term Memory (Bi-LSTM) architecture was trained on data from 22nd Jan 2020 to 8 Jan 2021 to forecast the new daily number of COVID-19 cases 14 days in advance across 190 countries, from 9 to 31 Jan 2021. A second model with fewer variables but similar architecture was developed. Results were summarised by mean absolute error (MAE), root mean squared error (RMSE), mean absolute percentage error (MAPE), and total absolute percentage error and compared against results from a classical ARIMA model. Median MAE was 157 daily cases (IQR: 26–666) under the first model, and 150 (IQR: 26–716) under the second. Countries with more accurate forecasts had more daily cases and experienced more waves of COVID-19 infections. Among countries with over 10,000 cases over the prediction period, median total absolute percentage error was 33% (IQR: 18–59%) and 34% (IQR: 16–66%) for the first and second models respectively. Both models had comparable median total absolute percentage errors but lower maximum total absolute percentage errors as compared to the classical ARIMA model. A deep-learning approach using Bi-LSTM architecture and open-source data was validated on 190 countries to forecast the daily number of cases in the early stages of the COVID-19 outbreak. Fewer variables could potentially be used without impacting prediction accuracy.

Estimating the concentration of total suspended solids in inland and coastal waters from Sentinel-2 MSI: A semi-analytical approach

Inland and coastal waters provide key ecosystem services and are closely linked to human well-being. In this study, we propose a semi-analytical method, which can be applied to Sentinel-2 MultiSpectral Instrument (MSI) images to retrieve high spatial-resolution total suspended solids (TSS) concentration in a broad spectrum of aquatic ecosystems ranging from clear to extremely turbid waters. The presented approach has four main steps. First, the remote sensing reflectance (Rrs) at a band lacking in MSI (620 nm) is estimated through an empirical relationship from Rrs at 665 nm. Second, waters are classified into four types (clear, moderately turbid, highly turbid, and extremely turbid). Third, semi-analytical algorithms are used to estimate the particulate backscattering coefficient (bbp) at a reference band depending on the water types. Last, TSS is estimated from bbp at the reference band. Validation and comparison of the proposed method with three existing methods are performed using a simulated dataset (N = 1000), an in situ dataset collected from global inland and coastal waters (N = 1265) and satellite matchups (N = 40). Results indicate that the proposed method can improve TSS estimation and provide accurate retrievals of TSS from all three datasets, with a median absolute percentage error (MAPE) of 14.88 %, 31.50 % and 41.69 % respectively. We also present comparisons of TSS mapping between the Sentinel-3 Ocean and Land Colour Instrument (OLCI) and MSI in Lake Kasumigaura, Japan and the Tagus Estuary, Portugal. Results clearly demonstrate the advantages of using MSI for TSS monitoring in small water bodies such as rivers, river mouths and other nearshore waters. MSI can provide more detailed and realistic TSS estimates than OLCI in these water bodies. The proposed TSS estimation method was applied to MSI images to produce TSS time-series in Lake Kasumigaura, which showed good agreements with in situ and OLCI-derived TSS time-series.

Intrusion detection in cloud computing based on time series anomalies utilizing machine learning

The growth of cloud computing is hindered by concerns about privacy and security. Despite the widespread use of network intrusion detection systems (NIDS), the issue of false positives remains prevalent. Furthermore, few studies have approached the intrusion detection problem as a time series issue, requiring time series modeling. In this study, we propose a novel technique for the early detection of intrusions in cloud computing using time series data. Our approach involves a method for Feature Selection (FS) and a prediction model based on the Facebook Prophet model to assess its efficiency. The FS method we propose is a collaborative feature selection model that integrates time series analysis techniques with anomaly detection, stationary, and causality tests. This approach specifically addresses the challenge of misleading connections between time series anomalies and attacks. Our results demonstrate a significant reduction in predictors employed in our prediction model, from 70 to 10 predictors, while improving performance metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), Median Absolute Percentage Error (MdAPE), and Dynamic Time Warping (DTW). Furthermore, our approach has resulted in reduced training, prediction, and cross-validation times of approximately 85%, 15%, and 97%, respectively. Although memory consumption remains similar, the utilization time has been significantly reduced, resulting in substantial resource usage reduction. Overall, our study presents a comprehensive methodology for effective early detection of intrusions in cloud computing based on time series anomalies, employing a collaborative feature selection model and the Facebook Prophet prediction model. Our findings highlight the efficiency and performance improvements achieved through our approach, contributing to the advancement of intrusion detection techniques in the context of cloud computing security.

Pelvic PET/MR attenuation correction in the image space using deep learning.

The five-class Dixon-based PET/MR attenuation correction (AC) model, which adds bone information to the four-class model by registering major bones from a bone atlas, has been shown to be error-prone. In this study, we introduce a novel method of accounting for bone in pelvic PET/MR AC by directly predicting the errors in the PET image space caused by the lack of bone in four-class Dixon-based attenuation correction. A convolutional neural network was trained to predict the four-class AC error map relative to CT-based attenuation correction. Dixon MR images and the four-class attenuation correction µ-map were used as input to the models. CT and PET/MR examinations for 22 patients ([18F]FDG) were used for training and validation, and 17 patients were used for testing (6 [18F]PSMA-1007 and 11 [68Ga]Ga-PSMA-11). A quantitative analysis of PSMA uptake using voxel- and lesion-based error metrics was used to assess performance. In the voxel-based analysis, the proposed model reduced the median root mean squared percentage error from 12.1% and 8.6% for the four- and five-class Dixon-based AC methods, respectively, to 6.2%. The median absolute percentage error in the maximum standardized uptake value (SUVmax) in bone lesions improved from 20.0% and 7.0% for four- and five-class Dixon-based AC methods to 3.8%. The proposed method reduces the voxel-based error and SUVmax errors in bone lesions when compared to the four- and five-class Dixon-based AC models.

Demand Side Control for Energy Saving in Renewable Energy Resources Using Deep Learning Optimization

Convolutional neural networks a type of deep learning technology, are used for forecasting future power usage. The mean absolute error, mean square error, root mean square error, and mean constant percentage error are used to evaluate the performance of the models. These metrics are used to rank the models. Among the models tested, the CNN stack demonstrates the highest precision in estimating energy consumption and solar power output, with a mean absolute error of 0.015% points, a root mean square error of 0.23% points, and an average absolute percentage deviation of 1.71% points. However, for wind turbine (WT) energy generation, the recurrent neural network proves for most accurate model, achieving 0.070 as the median absolute percentage error is 2.65, Both the root of the mean square error and the average absolute error are 0.38. The training and validation data utilized in this study comprise the International Renewable Energy Agency’s data on solar power production, WT power generation, and the “ensue” dataset, which includes hourly power consumption information from the Pennsylvania, New Jersey, and Maryland interconnection.

Validity and reliability of a commercial wearable sensor system for measuring spatiotemporal gait parameters in a post-stroke population: the effects of walking speed and asymmetry

Objective. Commercial wearable sensor systems are a promising alternative to costly laboratory equipment for clinical gait evaluation, but their accuracy for individuals with gait impairments is not well established. Therefore, we investigated the validity and reliability of the APDM Opal wearable sensor system to measure spatiotemporal gait parameters for healthy controls and individuals with chronic stroke. Approach. Participants completed the 10 m walk test over an instrumented mat three times in different speed conditions. We compared performance of Opal sensors to the mat across different walking speeds and levels of step length asymmetry in the two populations. Main results. Gait speed and stride length measures achieved excellent reliability, though they were systematically underestimated by 0.11 m s−1 and 0.12 m, respectively. The stride and step time measures also achieved excellent reliability, with no significant errors (median absolute percentage error <6.00%, p> 0.05). Gait phase duration measures achieved moderate-to-excellent reliability, with relative errors ranging from 4.13%–21.59%. Across gait parameters, the relative error decreased by 0.57%–9.66% when walking faster than 1.30 m s−1; similar reductions occurred for step length symmetry indices lower than 0.10. Significance. This study supports the general use of Opal wearable sensors to obtain quantitative measures of post-stroke gait impairment. These measures should be interpreted cautiously for individuals with moderate-severe asymmetry or walking speeds slower than 0.80 m s−1.

Application of Double Seasonal Autoregressive Integrated Moving Average (DSARIMA) for Stock Forecasting

Background: Stock price forecasting assists investors to anticipate risks and opportunities in making prudent investments and maximizing returns. Objective: This study aims to identify the most accurate model for stock forecasting. Methods: This paper utilized the daily closing stock price of Unilever Indonesia, Tbk (UNVR) from January 1, 2018 to July 31, 202. Double Seasonal Autoregressive Integrated Moving Average (DSARIMA), was utilized in this study. Mean Absolute Scaled Error (MASE) and Median Absolute Percentage Error (MdAPE) are used to compare forecasting accuracy. Results: Following conducting each model, we assessed that the best models are DSARIMAX (0,1,[4]) ([3],1,1)5(1,1,0)253, regarding MASE and MdAPE corresponding to approximately 1.423 and 0.111. The scope of this study has limitations to a test set for one-month forecast periods. Conclusion: As stock prices rise, investors require precise forecasts. Models of forecasting must perform well. This analysis shows how the DSARIMA generate forecasts stock prices more accurately. This investigation evaluated the closing stock price of UNVR. Both MASE and MdAPE assess prediction. After analyzing each model, DSARIMAX (0,1,[4])([3],1,1)5(1,1,0)253 has the lowest MASE and MdAPE values, 1.423 and 0.111, respectively. The procedure lasted one month. Research may combine forecasts and improve their accuracy.

944-P: Continuous Glucose Monitoring (CGM) Sampling Duration to Estimate Long-Term Glycemic Control in Adults with T2D Not Using Insulin

Objective: Current consensus is that 10-14 days of continuous glucose monitoring (CGM) sampling time is adequate to assess long-term glycemic control in T1D and insulin-using T2D populations. Optimal CGM sampling duration has not been widely studied in people with T2D not using insulin. We studied the relationship between true and estimated long-term (90-day) glycemic control in people with T2D not using insulin. Methods: The study population (N = 829) was composed of adults with T2D enrolled in Level2, a virtual specialty clinic, with CGM data on 75% (68) of the 90 days preceding an A1c lab result and no insulin fills during this period. The mean age and A1c were 56 years and 6.7% respectively. We computed the median absolute percentage error (MAPE) and Pearson correlations (R) between glycemic metrics from sampling periods of increasing duration and the full 90-day period. Because this population was at low risk for hypoglycemia glucose management indicator (GMI) and time above range (&amp;gt;180 mg/dL, TAR) were the primary outcomes of interest. We also assessed the correlations between GMI and A1c. Results: GMI and TAR from 14-day measurements were highly correlated (R = 0.92) with 90-day values, increasing to R = 0.94 with 30-day measurements. Correlations of GMI and A1c were high with 14 and 30 days (R = 0.79, R = 0.82) but did not exceed R = 0.85 with longer sampling periods. MAPE was 2.2 (SD = 2.6) and 1.5 (SD = 2.1) for GMI with 14- and 30-day measurements, respectively, and 29.2 (SD = 47.4) and 21.2 (SD = 27.1) for TAR. Conclusion: In the T2D population not using insulin, 14 days of CGM data is sufficient to estimate long-term mean glucose and the discordance with A1c is consistent with previous studies of people with T1D and insulin-using T2D. TAR is subject to greater error than in previous studies of people with T1D and insulin-using T2D and more work is needed to understand this discrepancy. Disclosure S.Bacon: Employee; Optum Labs, Research Support; Level2. N.Thompson: Employee; Level2, Stock/Shareholder; UnitedHealth Group. D.Ferrell: Stock/Shareholder; Level2, Medexpress Urgent Care. J.Fung: Employee; Level2.

A hybrid model to predict the pressure gradient for the liquid-liquid flow in both horizontal and inclined pipes for unknown flow patterns

Accurate prediction of the pressure gradient (PG) for the oil-water flow requires identification of the flow pattern (FP), which is usually achieved by using either an expensive measurement system or time-consuming manual observations. This study proposes a hybrid scheme where two machine-learning (ML) models are coupled in a series to predict the PG value without any conclusive FP information. The first model (M1) determines the oil-water FP, whereas the second model (M2) predicts the oil-water PG. 1637 experimental data points for the oil-water flow in both horizontal and inclined pipes are used to develop the models. The important feature subset is identified using the modified Binary Grey Wolf Optimization Particle Swarm Optimization (BGWOPSO) algorithm. The MLs' performance is evaluated using metrics including accuracy, sensitivity, specificity, and F1-score for the M1, and coefficient of variation of root mean squared error, mean absolute percentage error (MAPE), and median absolute percentage error for the M2. The evaluation metrics are cross-validated using a repeated train-test split strategy. The results showed that the overall FP classification accuracy is greater than 91%, with 90.61% sensitivity and 98.53% specificity using the weighted majority voting for M1. With the Gaussian Process regression for M2, the evaluation metrics for the PG prediction were found to be 10.65%, 86.26 Pa/m, and 0.96 for MAPE, root mean square error, and adjusted coefficient of determination, respectively. Statistical analysis showed that the selected features for liquids' and pipe's properties using the BGWOPSO algorithm were adequate to attain superior performance for both models. The achieved MAPE using the proposed hybrid model is superior to existing mechanistic or correlation-based models reported in the literature (between 26 and 69%). The proposed hybrid scheme can significantly reduce the costs associated with identifying the oil-water flow profile and be critical in designing energy-efficient transportation of liquid-liquid flow.

An effective ensemble method for missing data imputation

The presence of missing data in a dataset plays a vital role in the design of classification, clustering, or regression methods. An efficient missing data imputation can enhance the overall performance of a machine learning method. This paper ensembles k-nearest neighbour imputation, local least square imputation, miss forest imputation, and k-means clustering imputation using the bagging approach to handle missing values over a wide range of datasets. The method has been tested with eight different datasets in terms of root mean square error, median absolute percentage error, mean absolute percentage error, and standard deviation. Experimental results show that our method gives a low error rate compared to its closed competitors.