Predicting Coronary Heart Disease Using an Improved LightGBM Model: Performance Analysis and Comparison

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 23366 - 23380

Published: Jan. 1, 2023

Coronary heart disease (CHD) is a dangerous condition that cannot be completely cured. Accurate detection of early coronary artery can assist physicians in treating patients. In this study, prediction model called HY_OptGBM was proposed for predicting CHD by using the optimized LightGBM classifier. To optimize classifier, hyperparameters were adjusted. addition, its loss function improved, and trained adjusted hyperparameters. applying most advanced hyperparameter optimization framework (OPTUNA). The improved referred to as focal (FL). evaluated data from Framingham Heart Institute. evaluate performance model, various metrics, including precision, recall, F score, accuracy, MCC, sensitivity, specificity, AUC, used. AUC value 97.9%, which better than other comparative models. results demonstrate rate identification among general population utilizing method. This, turn, could serve mitigate costs associated with medical treatment patients suffering CHD.

Language: Английский

---

A robust and interpretable ensemble machine learning model for predicting healthcare insurance fraud

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 2, 2025

Healthcare insurance fraud imposes a significant financial burden on healthcare systems worldwide, with annual losses reaching billions of dollars. This study aims to improve detection accuracy using machine learning techniques. Our approach consists three key stages: data preprocessing, model training and integration, result analysis feature interpretation. Initially, we examined the dataset's characteristics employed embedded permutation methods test performance runtime single models under different sets, selecting minimal number features that could still achieve high performance. We then applied ensemble techniques, including Voting, Weighted, Stacking methods, combine compare their performances. Feature interpretation was achieved through partial dependence plots (PDP), SHAP, LIME, allowing us understand each feature's impact predictions. Finally, benchmarked our against existing studies evaluate its advantages limitations. The findings demonstrate improved offer insights into interpretability in this context.

Language: Английский

---

Exploring the Molecular Interaction of PCOS and Endometrial Carcinoma through Novel Hyperparameter-Optimized Ensemble Clustering Approaches

Mathematics, Journal Year: 2024, Volume and Issue: 12(2), P. 295 - 295

Published: Jan. 16, 2024

Polycystic ovary syndrome (PCOS) and endometrial carcinoma (EC) are gynecological conditions that have attracted significant attention due to the higher prevalence of EC in patients with PCOS. Even this proven association, little is known about complex molecular pathways connect PCOS an increased risk EC. In order address this, our study presents two main innovations. To provide a solid basis for analysis, we first created dataset genes linked Second, start by building fixed-size ensembles, then refine configuration single clustering algorithm within ensemble at each step hyperparameter optimization process. This evaluates potential performance as whole, taking into consideration interactions between algorithm. All models individually optimized suitable method, which allows us tailor strategy model’s needs. Our approach aims improve ensemble’s performance, significantly enhancing accuracy robustness outcomes. Through approach, aim enhance understanding EC, potentially leading diagnostic treatment breakthroughs.

Language: Английский

---

Breaking through the limitation of carbon price forecasting: A novel hybrid model based on secondary decomposition and nonlinear integration

Journal of Environmental Management, Journal Year: 2024, Volume and Issue: 362, P. 121253 - 121253

Published: June 1, 2024

Language: Английский

---

A Decomposition-Integration Framework of Carbon Price Forecasting Based on Econometrics and Machine Learning Methods

Mathematics, Journal Year: 2025, Volume and Issue: 13(3), P. 464 - 464

Published: Jan. 30, 2025

Carbon price forecasting and pricing are critical for stabilizing carbon markets, mitigating investment risks, fostering economic development. This paper presents an advanced decomposition-integration framework which seamlessly integrates econometric models with machine learning techniques to enhance forecasting. First, the complete ensemble empirical mode decomposition adaptive noise (CEEMDAN) method is employed decompose data into distinct modal components, each defined by specific frequency characteristics. Then, Lempel–Ziv complexity dispersion entropy algorithms applied analyze these facilitating identification of their unique attributes. The subsequently employs GARCH predicting high-frequency components a gated recurrent unit (GRU) neural network optimized grey wolf algorithm low-frequency components. Finally, GRU model utilized integrate predictive outcomes nonlinearly, ensuring comprehensive precise forecast. Empirical evidence demonstrates that this not only accurately captures diverse characteristics different but also significantly outperforms traditional benchmark in accuracy. By optimizing optimizer (GWO) algorithm, enhances both prediction stability adaptability, while nonlinear integration approach effectively mitigates error accumulation. innovative offers scientifically rigorous efficient tool forecasting, providing valuable insights policymakers market participants trading.

Language: Английский

---

Molecular structure of NRG-1 protein and its impact on adult hypertension and heart failure: A new clinical Indicator diagnosis based on advanced machine learning

International Journal of Biological Macromolecules, Journal Year: 2025, Volume and Issue: unknown, P. 140955 - 140955

Published: Feb. 1, 2025

Language: Английский

---

EBHOA-EMobileNetV2: a hybrid system based on efficient feature selection and classification for cardiovascular disease diagnosis

Computer Methods in Biomechanics & Biomedical Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 23

Published: Feb. 19, 2025

The accurate prediction of cardiovascular disease (CVD) or heart is an essential and challenging task to treat a patient efficiently before occurring attack. Many deep learning machine frameworks have been developed recently predict in intelligent healthcare. However, lack data-recognized appropriate methodologies meant that most existing strategies failed improve accuracy. This paper presents healthcare framework based on model detect disease, motivated by present issues. Initially, the proposed system compiles data from multiple publicly accessible sources. To quality dataset, effective pre-processing techniques are used including (i) interquartile range (IQR) method identify eliminate outliers; (ii) standardization technique handle missing values; (iii) 'K-Means SMOTE' oversampling address issue class imbalance. Using Enhanced Binary Grasshopper Optimization Algorithm (EBHOA), dataset's features chosen. Finally, presence absence CVD predicted using MobileNetV2 (EMobileNetV2) model. Training evaluation approach were conducted UCI Heart Disease Framingham Study datasets. We obtained excellent results comparing with recent methods. beats current approaches concerning performance metrics, according experimental results. For research achieves higher accuracy 98.78%, precision 99%, recall 99% F1 score 99%. 99.39%, 99.50%, learning-based classification combined feature selection yielded best innovative has potential enhance consistency prediction, which would be advantageous for clinical practice care.

Language: Английский

---

Early prediction of intraventricular hemorrhage in very low birth weight infants using deep neural networks with attention in low-resource settings

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: March 24, 2025

Early prediction of intraventricular hemorrhage (IVH) in very low-birthweight infants (VLBWIs) remains challenging because multifactorial risk factors. IVH often occurs within a few hours after birth, yet its onset cannot be reliably predicted using clinical symptoms or vital signs such as blood pressure heart rate. Accurate early is critical for timely intervention but due to the limited feasibility current methods resource-constrained settings. Traditional require advanced equipment and specialized expertise. Therefore, developing predictive model based on set available factors crucial accurate reliable VLBWIs. We propose deep neural network-based with an attention mechanism (DNN-A) that utilizes readily The was trained tested data from 387 infants, incorporating eight variables, including maternal age, delivery mode, endotracheal intubation, birth weight, gestational age at delivery, APGAR scores 1 5 min, sex. Our DNN-A achieved 90% 87% accuracies training testing sets, respectively, area under curve 87, outperformed various state-of-the-art machine-learning-based models prediction. These results underscore effectiveness predicting VLBWIs low-resource

Language: Английский

---

Ensuring privacy in e-healthcare: secured wireless sensor networks for enhanced data protection

Journal of the Chinese Institute of Engineers, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 16

Published: April 12, 2025

Language: Английский

---

Machine Learning-based Electric Load Forecasting for Peak Demand Control in Smart Grid

Computers, materials & continua/Computers, materials & continua (Print), Journal Year: 2022, Volume and Issue: 74(3), P. 4785 - 4799

Published: Nov. 15, 2022

Increasing energy demands due to factors such as population, globalization, and industrialization has led increased challenges for existing infrastructure. Efficient ways of generation consumption like smart grids homes are implemented face these with reliable, cheap, easily available sources energy. Grid integration renewable other clean distributed is increasing continuously reduce carbon air pollutants emissions. But the increase in electric demand enhance instability grid. Short-term electrical load forecasting reduces grid fluctuation enhances robustness power quality Electrical advance on basic historical data modelling plays a crucial role peak control, reinforcement demand, balancing cost reduction. accurate very challenging task nonstationary nonlinearly nature data. Machine learning artificial intelligence have recognized more reliable methods based The purpose this study model Jajpur, Orissa using regression type machine algorithms Gaussian process (GPR). whether Jajpur taken simulation decided way that will be considered learn connection among past, current, future dependent variables, factors, relationship Based network able forecast one day ahead. helpful stability control management.

Language: Английский

---