Performance Of Prediction Models Research Articles

Reliability of Regression Based Hybrid Machine Learning Models for the Prediction of Solar Photovoltaics Power Generation

Solar Photovoltaics (PVs) have become widespread as micro-distributed electric power generators in urban residential and commercial areas due to their affordability and minimal maintenance requirements. Despite these advantages, PV generation is intermittent, necessitating the implementation of robust predictive algorithms to capture power generation trends effectively. Accurate PV generation prediction is crucial for balancing energy demand and supply. Precise predictions help utilities use other grid resources efficiently, enhance market participation, and make informed planning decisions. Therefore, ensuring the precise performance of predictive models is imperative in the power generation industry. However, the performance of the predictive models typically fluctuates over time, necessitating a reliability assessment of their performance. This paper develops a hybrid machine learning model by integrating the support vector machine (SVM), random forest (RF), and gradient boosting machine (GBM) with multivariate adaptive regression spline (MARS) to predict the hourly PV power generation for 3-day and 7-day periods in three urban areas of North Dakota. The performance of the models has been recorded over a year, and a probability distribution model is proposed to demonstrate the hybrid algorithm’s reliability and effectiveness. Results show the proposed prediction model is more reliable for shorter prediction periods. The hybrid-MARS models significantly improve prediction reliability compared to the stand-alone MARS model. Among these models, MARS-SVM exhibits the highest reliability and consistently better accuracy than other hybrid models.

Changes in prey-predator interactions in an Arctic food web under climate change

Global warming affects marine ecosystems by changing environmental conditions, ecosystem structure, and ecosystem functioning. In parts of the Arctic, increased sea temperature and decreased sea ice have led to a poleward expansion of boreal species and increased their interactions with native Arctic species. To investigate and quantify the changing interactions in an Arctic marine food web under new environmental conditions, we studied the interactions between key prey fish species in the seasonally ice-covered parts of the Barents Sea: adult polar cod (Boreogadus saida) and capelin (Mallotus villosus) and one of the major predators in the system: Atlantic cod (Gadus morhua). For this, we compared the predictive performance of threshold models predicting the abundance of adult polar cod as a function of Atlantic cod. Each model was associated with a hypothesis describing prey-predator interactions in different environmental conditions defined by threshold values of summer sea-ice or capelin stock biomass. The best predictive model showed that the predation effect of Atlantic cod on polar cod was strongest in years of low summer sea ice cover and low capelin stock biomass. Our results exemplified that Arctic species such as polar cod may experience increased predation pressure under climate change from boreal species such as Atlantic cod. These effects depend, however, not only on changes in abiotic drivers of species distributions, but also on food-web interactions involving mid-trophic level species such as capelin.

An effective surrogate-assisted rank method for evolutionary neural architecture search

Evolutionary neural architecture search (ENAS) is able to automatically design high-performed architectures of deep neural networks (DNNs) for specific tasks. In recent years, surrogate models have gained significant traction because they can estimate the performance of neural architectures, avoiding excessive computational costs for training. However, most existing surrogate models primarily predict the performance of architectures directly or predict pairwise comparison relationships, which makes it challenging to obtain the rank of a group of architectures when training samples are limited. To address this problem, we propose TCMR-ENAS, an effective triple-competition model-assisted rank method for ENAS. TCMR-ENAS employs a novel triple-competition surrogate model combined with a score-based fitness evaluation method to predict group performance rank. Moreover, a progressive online learning method is proposed to enhance the predictive performance of the triple-competition surrogate model in the framework of modified genetic search. To validate the effectiveness of TCMR-ENAS, we conducted a series of experiments on NAS-Bench-101, NAS-Bench-201, NATS-Bench and NAS-Bench-301, respectively. Experimental results show that TCMR-ENAS can achieve better performance with lower computational resources. The accuracies of searched architectures achieve the best results compared with those of the state-of-the-art methods with limited training samples. In addition, the factors that may influence the effectiveness of TCMR-ENAS are explored in the ablation studies.

Deep-learning and data-resampling: A novel approach to predict cyanobacterial alert levels in a reservoir

The proliferation of harmful algal blooms results in adverse impacts on aquatic ecosystems and public health. Early warning system monitors algal bloom occurrences and provides management strategies for promptly addressing high-concentration algal blooms following their occurrence. In this study, we aimed to develop a proactive prediction model for cyanobacterial alert levels to enable efficient decision-making in management practices. We utilized 11 years of water quality, hydrodynamic, and meteorological data from a reservoir that experiences frequent harmful cyanobacterial blooms in summer. We used these data to construct a deep-learning model, specifically a 1D convolution neural network (1D-CNN) model, to predict cyanobacterial alert levels one week in advance. However, the collected distribution of algal alert levels was imbalanced, leading to the biased training of data-driven models and performance degradation in model predictions. Therefore, an adaptive synthetic sampling method was applied to address the imbalance in the minority class data and improve the predictive performance of the 1D-CNN. The adaptive synthetic sampling method resolved the imbalance in the data during the training phase by incorporating an additional 156 and 196 data points for the caution and warning levels, respectively. The selected optimal 1D-CNN model with a filter size of 5 and comprising 16 filters achieved training and testing prediction accuracies of 97.3% and 85.0%, respectively. During the test phase, the prediction accuracies for each algal alert level (L-0, L-1, and L-2) were 89.9%, 79.2%, and 71.4%, respectively, indicating reasonably consistent predictive results for all three alert levels. Therefore, the use of synthetic data addressed data imbalances and enhanced the predictive performance of the data-driven model. The reliable forecasts produced by the improved model can support the development of management strategies to mitigate harmful algal blooms in reservoirs and can aid in building an early warning system to facilitate effective responses.

Predictive Modeling in IoT-Driven Agriculture: A Comparative Study of Regression and Classification with LIME Interpretability

Abstract. This study aims to compare the performance of regression and classification machine learning models in crop prediction by analyzing data from IoT-driven agriculture. Linear regression and random forest regression models were used to predict the percentage of root growth dry matter, while logistic regression and support vector machines were employed to classify crop production regions. To enhance model interpretability, the LIME tool was applied to analyze feature importance. The experimental results demonstrate that the models perform well in terms of prediction accuracy, and LIME provided clear feature explanations, helping identify the variables with the greatest impact on prediction outcomes. This research offers data-driven insights for optimizing resource management in smart agricultural systems.

Machine learning models for coagulation dysfunction risk in inpatients administered β-lactam antibiotics

The β-Lactam antibiotics represent a widely used class of antibiotics, yet the latent and often overlooked risk of coagulation dysfunction associated with their use underscores the need for proactive assessment. Machine learning methodologies can offer valuable insights into evaluating the risk of coagulation dysfunction associated with β-lactam antibiotics. This study aims to identify the risk factors associated with coagulation dysfunction related to β-lactam antibiotics and to develop machine learning models for estimating the risk of coagulation dysfunction with real-world data. A retrospective study was performed using machine learning modeling analysis on electronic health record data, employing five distinct machine learning methods. The study focused on adult inpatients discharged from 1 January 2018, to 31 December 2021, at the First Affiliated Hospital of Shandong First Medical University. The models were developed for estimating the risk of coagulation dysfunction associated with various β-lactam antibiotics based on electronic health record feasibility. The dataset was divided into training and test sets to assess model performance using metrics such as total accuracy and area under the curve. The study encompassed risk-factor analysis and machine learning model development for coagulation dysfunction in inpatients administered different β-lactam antibiotics. A total of 45,179 participants were included in the study. The incidence of coagulation disorders related to cefazolin sodium, cefoperazone/sulbactam sodium, cefminol sodium, amoxicillin/sulbactam sodium, and piperacillin/tazobactam sodium was 2.4%, 5.4%, 1.5%, 5.5%, and 4.8%, respectively. Machine learning models for estimating coagulation dysfunction associated with each β-lactam antibiotic underwent validation with 5-fold cross-validation and test sets. On the test set, the optimal models for cefazolin sodium, cefoperazone/sulbactam sodium, cefminol sodium, amoxicillin/sulbactam sodium, and piperacillin/tazobactam sodium yielded AUC values of 0.798, 0.768, 0.919, 0.783, and 0.867, respectively. The study findings suggest that machine learning classifiers can serve as valuable tools for identifying patients at risk of coagulation dysfunction associated with β-lactam antibiotics and intervening based on high-risk predictions. Enhanced access to administrative and clinical data could further enhance the predictive performance of machine learning models, thereby expanding pharmacovigilance efforts.

Development of a robust predictive model for neutropenia after esophageal cancer chemotherapy using GLMMLasso.

Neutropenia can easily progress to febrile neutropenia and is a risk factor for life-threatening infections. Predicting and preventing severe neutropenia can help avoid such infections. This study aimed to develop an optimal model using advanced statistical methods to predict neutropenia after 5-fluorouracil/cisplatin chemotherapy for esophageal cancer and to create a nomogram for clinical application. Patients who received 5-fluorouracil/cisplatin chemotherapy at Chiba University Hospital, Japan, between January 2011 and March 2021 were included. Clinical parameters were measured before the first, second, and third chemotherapy cycles and were randomly divided by patient into a training cohort (60%) and test cohort (40%). The predictive performance of Logistic, Stepwise, Lasso, and GLMMLasso models was evaluated by the area under the receiver-operating characteristic curve (AUC). A nomogram based on GLMMLasso was developed, and the accuracy of probabilistic predictions was evaluated by the Brier score. The AUC for the first cycle of chemotherapy was 0.781 for GLMMLasso, 0.751 for Lasso, 0.697 for Stepwise, and 0.669 for Logistic. The respective AUCs for GLMMLasso in the second and third cycles were 0.704 and 0.900. The variables selected by GLMMLasso were cisplatin dose, 5-fluorouracil dose, use of leucovorin, sex, cholinesterase, and platelets. A nomogram predicting neutropenia was created based on each regression coefficient. The Brier score for the nomogram was 0.139. We have developed a predictive model with high performance using GLMMLasso. Our nomogram can represent risk visually and may facilitate the assessment of the probability of chemotherapy-induced severe neutropenia in clinical practice.

Machine learning-based prediction of elevated N terminal pro brain natriuretic peptide among US general population.

Natriuretic peptide-based pre-heart failure screening has been proposed in recent guidelines. However, an effective strategy to identify screening targets from the general population, more than half of which are at risk for heart failure or pre-heart failure, has not been well established. This study evaluated the performance of machine learning prediction models for predicting elevated N terminal pro brain natriuretic peptide (NT-proBNP) levels in the US general population. Individuals aged 20-79years without cardiovascular disease from the nationally representative National Health and Nutrition Examination Survey 1999-2004 were included. Six prediction models (two conventional regression models and four machine learning models) were trained with the 1999-2002 cohort to predict elevated NT-proBNP levels (>125pg/mL) using demographic, lifestyle, and commonly measured biochemical data. The model performance was tested using the 2003-2004 cohort. Of the 10237 individuals, 1510 (14.8%) had NT-proBNP levels >125pg/mL. The highest area under the receiver operating characteristic curve (AUC) was observed in SuperLearner (AUC [95% CI]=0.862 [0.847-0.878], P<0.001 compared with the logistic regression model). The logistic regression model with splines showed a comparable performance (AUC [95% CI]=0.857 [0.841-0.874], P=0.08). Age, albumin level, haemoglobin level, sex, estimated glomerular filtration rate, and systolic blood pressure were the most important predictors. We found a similar prediction performance even after excluding socio-economic information (marital status, family income, and education status) from the prediction models. When we used different thresholds for elevated NT-proBNP, the AUC (95% CI) in the SuperLearner models 0.846 (0.830-0.861) for NT-proBNP>100pg/mL and 0.866 (0.849-0.884) for NT-proBNP>150pg/mL. Using nationally representative data from the United States, both logistic regression and machine learning models well predicted elevated NT-proBNP. The predictive performance remained consistent even when the models incorporated only commonly available variables in daily clinical practice. Prediction models using regularly measured information would serve as a potentially useful tools for clinicians to effectively identify targets of natriuretic-peptide screening.

Performance of the Machine Learning Algorithm in Stature Estimation Using Scapula Measurements from Post-Mortem Computed Tomography in a Thai Male Population

Objective: This study attempted to investigate the performance of stature prediction models from scapular dimensions based on post-mortem computed tomography (PMCT) using machine learning algorithms within the male population of Southern Thailand.Material and Methods: Linear Regression (LR), K-Nearest Neighbors (KNN), Random Forest tree (RF), and Support vector machine (SVM) algorithms were used to create the stature estimation model. Then its performance was compared by coefficient of determination (R2), mean absolute error (MAE), mean squared error (MSE), and root mean squared error (RMSE).Results: Machine learning is a valuable tool for estimating stature within this demographic. LR algorithm provided the best performance matrices, with the highest R2 being 0.316, and the lowest values for MAE, MSE, and RMSE being 4.379 cm, 29.530 cm, and 5.382 cm, respectively.Conclusion: The machine learning algorithm demonstrated valuable tools for estimation stature. However, it is essential to note that complex machine learning models do not always produce better performance measures than non-complex models.

A comparative analysis of prediction problems utilizing Interval type-2 fuzzy and machine learning models

This paper compares the effectiveness of Interval Type 2 Fuzzy Logic (IT2FL) and Machine Learning (ML) models in addressing real-world challenges. It employs four non-parametric ML algorithms (Support Vector Machine (SVM), K-Nearest Neighbor (KNN), (Random Forest (RF) and Classification and Regression Tree (CART)) and evaluates their performance against IT2FL using parameters generated through Gaussian membership functions. Initially, the IT2FL algorithm preprocesses the dataset by predicting missing values, leveraging IT2F membership to optimize decision-making and mitigate uncertainties. The study assesses the predictive performance, robustness, and interpretability of IT2F-ML models, analyzing datasets from cardiovascular disease patients to predict shock levels. After transforming the dataset using IT2FL, it is divided into 60% training and 40% testing sets to train and test four ML algorithms, aimed at predicting shock levels in patients. The models’ performance is evaluated using various metrics, demonstrating the superior learning enhancement and performance of the IT2F-SVR method compared to other predictive approaches on the same dataset. Moreso, it is observed that the integration of the IT2-FL algorithm with machine learning models offers a promising approach for addressing data incompleteness and improving prediction accuracy. Furthermore, the study provides valuable guidance for researchers and practitioners in selecting suitable methodologies for prediction tasks.

Circulating Micro-RNAs Predict the Risk of Recurrence in Triple-Negative Breast Cancer.

Triple-negative breast cancer (TNBC) is an aggressive subtype of breast cancer with a high tendency for developing a recurrent disease. Circulating micro-RNAs (cmiRNAs) obtained through liquid biopsy are potential prognostic biomarkers for the assessment of TNBC recurrence risk. In this study, we sequenced cmiRNAs from the serum samples of 14 recurrent and 19 non-recurrent TNBC cases and compared expression profiles in relation to recurrence status, survival data and miRNA expression in matched tumor samples. Differential expression analysis between recurrent and non-recurrent cases identified ten differentially expressed (DE) cmiRNAs, of which cmiRNAs miR-21-5p (p = 0.030, HR = 1.87, 95% CI 1.06-3.30), miR-16-5p (p = 0.032, HR = 0.53, 95% CI 0.30-0.95), and miR-26b-5p (p = 0.023, HR = 0.52, 95% CI 0.29-0.91) were associated with recurrence-free survival in multivariable survival analysis. Expression profiles of matched tumor and serum samples were shown to correlate with each other. DE cmiRNAs were associated with common cancer-related signaling pathways and improved the overall predictive performance of the logistic regression model assessing the recurrence risk. Our results indicate that recurrent and non-recurrent TNBC differ in their cmiRNA expression profiles, and that three specific cmiRNAs can be used to assess the risk of recurrence in TNBC.

QTc interval prolongation impact on in-hospital mortality in acute coronary syndromes patients using artificial intelligence and machine learning

BackgroundPrediction of mortality in hospitalized patients is a crucial and important problem. Several severity scoring systems over the past few decades and machine learning models for mortality prediction have been developed to predict in-hospital mortality. Our aim in this study was to apply machine learning (ML) algorithms using QTc interval to predict in-hospital mortality in ACS patients and compare them to the validated conventional risk scores.ResultsThis study was retrospective, using supervised learning, and data mining. Out of a cohort of 500 patients admitted to a tertiary care hospital from September 2018 to August 2020, who presented with ACS. Prediction models for in-hospital mortality in ACS patients were developed using 3 ML algorithms. We employed the ensemble learning random forest (RF) model, the Naive Bayes (NB) model and the rule-based projective adaptive resonance theory (PART) model. These models were compared to one another and to two conventional validated risk scores; the Global Registry of Acute Coronary Events (GRACE) risk score and Thrombolysis in Myocardial Infarction (TIMI) risk score. Out of the 500 patients included in our study, 164 (32.8%) patients presented with unstable angina, 148 (29.6%) patients with non-ST-elevation myocardial infarction (NSTEMI) and 188 (37.6%) patients were having ST-elevation myocardial infarction (STEMI). 64 (12.8%) patients died in-hospital and the rest survived. Performance of prediction models was measured in an area under the receiver operating characteristic curve (AUC) ranged from 0.83 to 0.93 using all available variables compared to the GRACE score (0.9 SD 0.05) and the TIMI score (0.75 SD 0.02). Using QTc as a stand-alone variable yielded (0.67 SD 0.02) with a cutoff value 450 using Bazett’s formula, whereas using QTc in addition to other variables of personal and clinical data and other ECG variables, the result was 0.8 SD 0.04. Results of RF and NB models were almost the same, but PART model yielded the least results. There was no significant difference of AUC values after replacing the missing values and applying class balancer.ConclusionsThe proposed method can effectively predict patients at high risk of in-hospital mortality early in the setting of ACS using only clinical and ECG data. Prolonged QTc interval can be used as a risk predictor of in-hospital mortality in ACS patients.

Adaptive hybrid prediction model for adapting to data distribution shifts in machining quality prediction

Abstract The accuracy of data-driven intelligent prediction for machining quality relies on the training samples. However, in actual applications, the continuous operation of machining equipment leads to gradual distribution shifts between the process data and the training samples for modeling. The shifts result in a degradation in the performance of predictive model, previous studies have often overlooked this issue. To tackle with the intricate problem, this research proposes a real-time model optimization approach. Firstly, a method for detecting machining data distribution shifts based on the two-sample Kolmogorov–Smirnov test is proposed. Then, an adaptive hybrid prediction model (AHPM) capable of real-time optimization is developed. This model consists of a deep neural network (DNN) and a broad learning system (BLS). DNN plays a primary role in prediction within the hybrid model with excellent generalization capability. BLS quickly completes optimization prior to DNN with its unique parameter update mechanism to compensate for prediction loss. Experimental results indicate that AHPM achieves the shortest optimization time while maintaining high accuracy, with post-optimization error reduction rates for mean squared error, mean absolute error, and mean absolute percentage error all exceeding 10%. In the test of application to actual machining cases, accuracy improved by 8.88% compared to traditional methods without optimization.

Preoperative Computed Tomography Radiomics-Based Models for Predicting Microvascular Invasion of Intrahepatic Mass-Forming Cholangiocarcinoma

Objectives The aim of the study is to investigate the ability of preoperative CT (Computed Tomography)-based radiomics signature to predict microvascular invasion (MVI) of intrahepatic mass-forming cholangiocarcinoma (IMCC) and develop radiomics-based prediction models. Materials and Methods Preoperative clinical data, basic CT features, and radiomics features of 121 IMCC patients (44 with MVI and 77 without MVI) were retrospectively reviewed. The loading and display of CT images, delineation of the volume of interest, and feature extraction were performed using 3D Slicer. Radiomics features were selected by the LASSO logistic regression model. Multivariate logistic regression analysis was used to establish the radiomics model, radiologic model, and combined model in the training set (n = 85) to predict the MVI of IMCC, and then verified in the validation set (n = 36). Results Among the 3948 radiomics features extracted from multiphase dynamic enhanced CT imaging, 16 most stable features were selected. The AUC of the radiomics model for predicting MVI in the training set and validation set were 0.935 and 0.749, respectively. The AUC of the radiologic model for predicting MVI in the training set and validation set were 0.827 and 0.796, respectively. When radiomics and radiologic models are combined, the predictive performance of the combined model (constructed with shape, intratumoral vessels, portal venous phase tumor-liver CT ratio, and radscore) is optimal, with an AUC of 0.958 in the training set and 0.829 in the test set for predicting MVI. Conclusions CT radiomics signature is a powerful predictor for predicting MVI. The preoperative combined model (constructed with shape, intratumoral vessels, portal venous phase tumor-liver CT ratio, and radscore) performed well in predicting the MVI.

Analysing crash severity on expressways in India using statistical and ML approaches

The high injury severity of traffic crashes on Indian expressways is a significant concern for road safety experts, although studies dedicated to this critical issue are limited. The present study investigates the factors influencing crash severity using multinomial logit (MNL), decision tree (DT) and random forest (RF) models on a dataset of 2,747 crashes on three selected expressways. The dependent variable, crash severity, had four injury severity categories: fatal, severe, minor, and property damage only (PDO). Various explanatory variables, included traffic and speed characteristics, temporal and geometric characteristics, primary contributing factors, crash type, and the vehicles involved. Synthetic minority oversampling (SMOTE) and randomise class balancing (RCB) techniques were also employed to tackle the class imbalance issue in the dataset. The predictive performance of models was evaluated using classification accuracy and Kappa value. The RF model showed the highest predictive accuracy on the RCB dataset. The key findings highlight the critical need for enforcement of speed limits and entry restrictions, lane discipline, and improvement of design deficiencies such as provisions of truck laybys and bus bays. These measures can help in policy-making and engineering improvements to enhance road safety on expressways in India.

Evaluation of Drug-Drug Interactions Between Clarithromycin and Direct Oral Anticoagulants Using Physiologically Based Pharmacokinetic Models.

Objective: This study assessed the pharmacokinetic (PK) interactions between clarithromycin (a P-glycoprotein [P-gp] inhibitor) and four direct oral anticoagulants (DOACs) (P-gp substrates) using physiologically based PK (PBPK) models to elucidate the influence of P-gp in the interaction between them. Methods: PBPK models for clarithromycin, DABE-dabigatran (DAB), rivaroxaban, apixaban, and edoxaban were constructed using GastroPlus™ (version 9.9), based on physicochemical data and PK parameters from the literature. The models were optimized and validated in healthy subjects. We evaluated the predictive performance of the established model and further assessed the impact of P-gp on the PK of the four DOACs. Successfully validated models were then used to evaluate potential drug-drug interactions (DDIs) between clarithromycin and the DOACs. Results: The established PBPK models accurately described the PK of clarithromycin, DABE-DAB, rivaroxaban, apixaban, and edoxaban. The predicted PK parameters (Cmax, Tmax, AUC0-t) were within 0.5-2 times the observed values. A sensitivity analysis of P-gp parameters indicated that an increase in P-gp expression was reduced by in vivo exposure to DOACs. The models demonstrated good predictive ability for DDIs between clarithromycin and the anticoagulants, and the ratio of the predicted values to the observed values of Cmax and the area under the curve (AUC) in the DDI state was within the range of 0.5-2. Conclusions: Comprehensive PBPK models for clarithromycin, DABE-DAB, rivaroxaban, apixaban, and edoxaban were developed, which can effectively predict DDIs mediated by P-gp's function. These models provide theoretical support for clinical dose adjustments and serve as a foundation for future PBPK model development for DOACs under specific pathological conditions.

Abstract 4136459: Machine Learning in Time-to-Event Prediction of Ventricular Arrhythmias among Older Adults with Type 2 Diabetes and Coronary Artery Disease

Introduction: Patients with Type 2 diabetes mellitus (T2DM) have an increased risk for coronary artery disease (CAD) compared to patients without T2DM. Ventricular arrhythmias (VA), such as ventricular fibrillation and ventricular tachycardia, are the major causes of mortality among patients with CAD. Thus, T2DM patients with CAD, especially older adults, have a higher risk of VA compared to patients without T2DM. The time-to-event prediction can contribute to decreasing mortality due to VA in older adults with T2DM and CAD by providing an estimated time and probability of diagnosis at specific time points. However, the limited number of older T2DM patients with CAD, who are diagnosed with ventricular arrhythmias, lowers the performance of a traditional time-to-event analysis model. Hypothesis: We hypothesize that machine learning can improve performance in time-to-event prediction. Methods: This study includes 3975 participants aged &gt;65 years, diagnosed with T2DM and CAD in the All of US database. The baseline time was the earliest date when the patients were older than 65 years and diagnosed with T2DM and CAD. Of all participants, 379 were diagnosed with VA within 5 years from baseline. We compared the machine learning–based time-to-event analysis models to the Cox Proportional Hazards model to explore the potential of machine learning in time-to-event analysis. Additionally, the Synthetic Minority Oversampling Technique was applied to resolve the imbalanced data issue. The model performance was evaluated based on the concordance index, which calculates the correlation between predicted risk scores and actual observations, after 5-fold cross-validation. Results: The average concordance index of the Gradient Boosting Machines model was the highest among the five models, including the Cox Proportional Hazards model, with balanced data (Table). Conclusion: We identified that machine learning could improve the performance of prediction models in time-to-event analysis. The Synthetic Minority Oversampling Technique could mitigate the issue of imbalanced data. In addition, machine learning-based survival models showed higher concordance index values compared to the Cox Proportional Hazards model.

Explainable machine learning for early prediction of sepsis in traumatic brain injury: A discovery and validation study.

People with traumatic brain injury (TBI) are at high risk for infection and sepsis. The aim of the study was to develop and validate an explainable machine learning(ML) model based on clinical features for early prediction of the risk of sepsis in TBI patients. We enrolled all patients with TBI in the Medical Information Mart for Intensive Care IV database from 2008 to 2019. All patients were randomly divided into a training set (70%) and a test set (30%). The univariate and multivariate regression analyses were used for feature selection. Six ML methods were applied to develop the model. The predictive performance of different models were determined based on the area under the curve (AUC) and calibration curves in the test cohort. In addition, we selected the eICU Collaborative Research Database version 1.2 as the external validation dataset. Finally, we used the Shapley additive interpretation to account for the effects of features attributed to the model. Of the 1555 patients enrolled in the final cohort, 834 (53.6%) patients developed sepsis after TBI. Six variables were associated with concomitant sepsis and were used to develop ML models. Of the 6 models constructed, the Extreme Gradient Boosting (XGB) model achieved the best performance with an AUC of 0.807 and an accuracy of 74.5% in the internal validation cohort, and an AUC of 0.762 for the external validation. Feature importance analysis revealed that use mechanical ventilation, SAPSII score, use intravenous pressors, blood transfusion on admission, history of diabetes, and presence of post-stroke sequelae were the top six most influential features of the XGB model. As shown in the study, the ML model could be used to predict the occurrence of sepsis in patients with TBI in the intensive care unit.

Pseudo-Labeling and Time-Series Data Analysis Model for Device Status Diagnostics in Smart Agriculture

This study proposes an automated data-labeling model that combines a pseudo-labeling algorithm with waveform segmentation based on Long Short-Term Memory (LSTM) to effectively label time-series data in smart agriculture. This model aims to address the inefficiency of manual labeling for large-scale data generated by agricultural systems, enhancing the performance and scalability of predictive models. Our proposed method leverages key features of time-series data to automatically generate labels for new data, thereby improving model accuracy and streamlining data processing. By automating the labeling process, we reduce dependence on manual labeling, which is often labor-intensive and prone to errors in large datasets. This approach enables the efficient preparation of labeled data for applications such as anomaly detection, pattern recognition, and predictive modeling in smart agriculture. Experimental results demonstrate that the automated labeling model achieves 89% accuracy in agricultural environments and reduces data processing time by 30%. Future research will focus on extending the model’s applicability to diverse agricultural settings, enhancing generalization performance, and improving real-time processing capabilities, thereby advancing intelligent and sustainable smart agriculture systems.

Height prediction of individuals with osteogenesis imperfecta by machine learning

BackgroundOsteogenesis imperfecta (OI) is a genetic disorder characterized by low bone mass, bone fragility and short stature. There is a significant gap in knowledge regarding the growth patterns across different types of OI, and the prediction of height in individuals with OI was not adequately addressed. In this study, we described the growth patterns and predicted the height of individuals with OI employing multiple machine learning (ML) models. Accurate height prediction enables effective monitoring and facilitates the development of personalized intervention plans for managing OI.MethodThis study included cross-sectional data for 323 participants with OI, and the median height Z-score for OI types I, III and IV were − 0.62 (-5.93 ~ 3.24), -3.97 (-10.44 ~ -0.02) and − 1.64 (-6.67 ~ 2.44), respectively. Based on the cross-sectional data of participants, the height curves across different gender and OI types were plotted and compared. Subsequently, feature selection techniques, specifically the filter and wrapper methods, were employed to identify predictive factors for the height of participants. Finally, multiple machine learning (ML) models were constructed for height prediction, and the performance of each model was systematically evaluated.ResultsThe analysis of height curves revealed that male with OI are significantly taller than female with OI from the age of 14 (p = 0.045), individuals with OI type III are statistically shorter than those with OI types I and IV starting from 3 years old (p = 0.006), and those with OI type IV are statistically shorter than those with OI type I from the age of 10 (p = 0.028). The application of filter and wrapper methods identified gender (p = 0.001), age (p < 0.001), Sillence types (p = 0.007), weight Z-score (p < 0.001) and aBMD Z-score (p = 0.021) as significant predictive factors for height. The optimal performance of predictive models was registered by gradient boosting classifier (GB) (bias = 5.783, accuracy = 92.59%, R2 = 0.828), random forest (RF) (bias = 6.155, accuracy = 90.12%, R2 = 0.788), ensemble machine learning (EML) (bias = 6.250, accuracy = 91.36%, R2 = 0.825) and deep neuron networks (DNNs) (bias = 6.223, accuracy = 90.12%, R2 = 0.821).ConclusionThis study analyzed a large cohort of individuals with OI and provided detailed height patterns across different gender and OI types that are crucial for assessing overall growth. Gender, age, Sillence types, weight Z-score and aBMD Z-score were identified as predictive factors for height. The predictive models of GB, RF, EML and DNNs had higher accuracy to evaluate the height of individuals with OI. This study allows guardians and physicians to timely monitor the height parameters, and facilitate the creation of personalized intervention schedules tailored to the needs of individuals with OI.