Cardiovascular disease detection using a novel stack-based ensemble classifier with aggregation layer, DOWA operator, and feature transformation

Computers in Biology and Medicine, Год журнала: 2024, Номер 173, С. 108345 - 108345

Опубликована: Март 27, 2024

Язык: Английский

---

A Survey on Machine Learning Techniques for Heart Disease Prediction

SN Computer Science, Год журнала: 2025, Номер 6(4)

Опубликована: Апрель 1, 2025

Язык: Английский

---

Hybrid Time-Frequency Domain Analysis for Cardiovascular Disease Forecasting Over ECG Data

Communications in computer and information science, Год журнала: 2025, Номер unknown, С. 316 - 327

Опубликована: Янв. 1, 2025

Язык: Английский

---

Identification and diagnosis of chronic heart disease: A deep learning-based hybrid approach

Alexandria Engineering Journal, Год журнала: 2025, Номер 124, С. 470 - 483

Опубликована: Апрель 11, 2025

Язык: Английский

---

Exploring Predictive Methods for Cardiovascular Disease: A Survey of Methods and Applications

IEEE Access, Год журнала: 2024, Номер 12, С. 101497 - 101505

Опубликована: Янв. 1, 2024

Because cardiovascular disease (CVD) is still one of the world's leading causes death, sophisticated predictive models are required for early detection and prevention. This study examined how to make compare different CVD prediction using a large dataset that included biochemical, clinical, demographic information about each person. During preprocessing stage, we took great care ensure data's accuracy quality. We have utilized variety machine learning algorithms such as random forest, logistic regression, support vector machines, deep neural networks. assessed performance these accuracy, sensitivity, specificity, area under receiver operating characteristic curve (AUC-ROC). Our findings show while more algorithms—especially models—perform better at spotting possible instances CVD, conventional models—such regression—offer significant power. also investigated role feature selection has in improving interpretability efficiency model. highlights potential transform emphasizes importance many forms data provide thorough risk evaluation. research adds continuing efforts personalized medicine by providing on creating precise effective tools health management.

Язык: Английский

---

DASMcC: Data Augmented SMOTE Multi-Class Classifier for Prediction of Cardiovascular Diseases Using Time Series Features

IEEE Access, Год журнала: 2023, Номер 11, С. 117643 - 117655

Опубликована: Янв. 1, 2023

One of the leading causes mortality worldwide is cardiovascular disease (CVD). Electrocardiography (ECG) a noninvasive and cost-effective tool to diagnose heart's health. This study presents multi-class classifier for prediction four different types Cardiovascular Diseases, i.e., Myocardial Infarction, Hypertrophy, Conduction Disturbances, ST-T abnormality using 12-lead ECG. There are key steps involved in presented work: data preprocessing, feature extraction, preparation, augmentation, modelling CVD classification. The sixteen-time domain augmented features used train classifier. work can be divided into three parts: extracting from raw ECG signals, preparation training, testing, validating A comparative performance five classifiers (i.e., Random Forest (RF), K Nearest Neighbors (KNN), Gradient Boost, Adda XG Boost has also been presented. Accuracy, precision, recall, F1 scores evaluation. Further, Receiver Operating Curve (ROC) traced, Area Under (AUC) calculated ensure unbiased application proposed Smart Healthcare framework discussed.

Язык: Английский

---

Hybrid Feature Selection with Chaotic Rat Swarm Optimization-Based Convolutional Neural

Data & Metadata, Год журнала: 2024, Номер 3, С. 262 - 262

Опубликована: Янв. 1, 2024

Introduction: Early diagnosis of Cardiovascular Disease (CVD) is vital in reducing mortality rates. Artificial intelligence and machine learning algorithms have increased the CVD prediction capability clinical decision support systems. However, shallow feature incompetent selection methods still pose a greater challenge. Consequently, deep are needed to improvise frameworks. Methods: This paper proposes an advanced CDSS for detection using hybrid DL method. Initially, Improved Hierarchical Density-based Spatial Clustering Applications with Noise (IHDBSCAN), Adaptive Class Median-based Missing Value Imputation (ACMMVI) Using Representatives-Adaptive Synthetic Sampling (CURE-ADASYN) approaches introduced pre-processing stage enhancing input quality by solving problems outliers, missing values class imbalance, respectively. Then, features extracted, optimal subsets selected model Information gain Owl Optimization algorithm (IG-IOOA), where OOA improved search functions local process. These fed proposed Chaotic Rat Swarm Optimization-based Convolutional Neural Networks (CRSO-CNN) classifier detecting heart disease. Results: Four UCI datasets used validate framework, results showed that OOA-DLSO-ELM-based approach provides better disease high accuracy 97,57 %, 97,32 96,254 % 97,37 four datasets. Conclusions: Therefore, this CRSO-CNN improves classification reduced time complexity all

Язык: Английский

---

Disease Prognosis Using Artificial Intelligence Neural Networks

Journal of Artificial Intelligence and Capsule Networks, Год журнала: 2024, Номер 6(1), С. 1 - 14

Опубликована: Фев. 13, 2024

The proactive anticipation of disease occurrence stands as a pivotal facet within healthcare and medical research endeavors, dedicated to forecasting the probability an individual manifesting particular condition or ailment in future. This fundamental pursuit integrates diverse data reservoirs, encompassing history, genetic profiles, lifestyle determinants, emerging technological advancements, construct predictive frameworks capable furnishing early indications insights pertaining potential health vulnerabilities. overarching aim prediction resides practitioners individuals alike with requisite knowledge resources undertake pre-emptive measures, render informed choices, ultimately enhance holistic well-being. Neural Network algorithm emerges dependable approach for prognostication, offering heightened precision several advantages compared conventional methodologies, including its capacity discern intricate features from images adaptability across computing platforms. proposed study offers comprehensive review methods, comparing approaches machine learning interventions provide swift reliable results. Further suggests model that utilizes neural network algorithms overcome shortcomings methods.

Язык: Английский

---

Future prediction for precautionary measures associated with heart-related issues based on IoT prototype

Multimedia Tools and Applications, Год журнала: 2024, Номер 83(23), С. 63723 - 63753

Опубликована: Апрель 16, 2024

Язык: Английский

---

AN INTEGRATED MACHINE LEARNING FRAMEWORK FOR ACCURATE CARDIOVASCULAR DISEASE PREDICTION

International Journal Of Trendy Research In Engineering And Technology, Год журнала: 2024, Номер 08(01), С. 31 - 39

Опубликована: Янв. 1, 2024

This paper introduces an advanced real-time system designed to predict cardiovascular diseases with integrated machine learning. Cardiovascular diseases, the highest global mortality rate, have become increasingly prevalent, straining healthcare systems worldwide. These driven by factors such as high blood pressure, stress, age, gender, and cholesterol levels, prompted numerous early diagnosis approaches, but their accuracy requires refinement due critical nature of diseases. presents DLCDD (Deep Learning based Disease Diagnosis) framework, specifically addressing data-related challenges missing values imbalances. The mean replacement technique is employed for handling values, while Synthetic Minority Over-sampling Technique (SMOTE) utilized address imbalances in dataset. In essence, represents a significant advance precise disease prediction, uniting deep learning cutting-edge data processing feature selection methods, diagnostic challenges.

Язык: Английский

---