Chronic Diseases Prediction Using Machine Learning With Data Preprocessing Handling: A Critical Review

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 80698 - 80730

Published: Jan. 1, 2024

According to the World Health Organization (WHO), some chronic diseases such as diabetes mellitus, stroke, cancer, cardiac vascular, kidney failure, and hypertension are essential for early prevention. One of prevention that can be taken is predict using machine learning based on personal medical record or general checkup result. The common prediction objective minimize error low possible. most influencing factors quality data choice predictor methods. five main problems those lower outliers, missing values, feature selection, normalization, imbalance. After we ensure data, next task choose best factor consider when its performance evaluation (accuracy, recall, precision, f1-score). Thus, predicting disease aims produce increased solve in data. This paper presents a Systematic Literature Review (SLR) offers comprehensive discussion research preprocessing handling. covers methods supervised learning, ensemble deep reinforcement learning. handling discuss includes final discussions this open issues, potential future works improving

Language: Английский

---

Development of an efficient novel method for coronary artery disease prediction using machine learning and deep learning techniques

Technology and Health Care, Journal Year: 2024, Volume and Issue: 32(6), P. 4545 - 4569

Published: July 19, 2024

BACKGROUND: Heart disease is a severe health issue that results in high fatality rates worldwide. Identifying cardiovascular diseases such as coronary artery (CAD) and heart attacks through repetitive clinical data analysis significant task. Detecting its early stages can save lives. The most lethal condition CAD, which develops over time due to plaque buildup arteries, causing incomplete blood flow obstruction. Machine Learning (ML) progressively used the medical sector detect CAD disease. OBJECTIVE: primary aim of this work deliver state-of-the-art approach enhancing prediction accuracy by using DL algorithm classification context. METHODS: A unique ML technique proposed study predict accurately deep learning An ensemble voting classifier model developed based on various methods Naïve Bayes (NB), Logistic Regression (LR), Decision Tree (DT), XGBoost, Random Forest (RF), Convolutional Neural Network (CNN), Support Vector (SVM), K Nearest Neighbor (KNN), Bidirectional LSTM Long Short-Term Memory (LSTM). performance models novel are compared study. Alizadeh Sani dataset, consists random sample 216 cases with Synthetic Minority Over Sampling Technique (SMOTE) address imbalanced datasets, Chi-square test for feature selection optimization. Performance assessed assessment methodologies, confusion matrix, accuracy, recall, precision, f1-score, auc-roc. RESULTS: When achieves highest relative other algorithms, it demonstrates effectiveness several ways, including superior performance, robustness, generalization capability, efficiency, innovative approaches, benchmarking against baselines. These characteristics collectively contribute establishing promising solution addressing target problem machine related fields. CONCLUSION: Implementing significantly improved achieving rate 92% detection CAD. findings competitive par top outcomes among methods.

Language: Английский

---

Heart Disease Detection Using AI

International Journal of Innovative Science and Research Technology (IJISRT), Journal Year: 2024, Volume and Issue: unknown, P. 227 - 232

Published: March 12, 2024

Over the past few decades, cardiovascular disease has emerged as primary cause of death worldwide in both industrialized and developing nations. Early detection heart problems continued clinical monitoring can reduce rates. However, because it takes more time experience, is not possible to accurately detect disorders all cases have a specialist talk with patient for 24 hours. We demonstrate how machine learning be used estimate an individual's risk disease. This study presents data processing, which includes converting categorical columns working variables. outline three stages application: gathering datasets, running logistic regression, assessing properties dataset. The random forest classifier technique developed diagnose cardiac precisely. Data analysis needed this application since considered noteworthy. algorithm, improves accuracy research diagnosis, next covered, along experiments findings.

Language: Английский

---

A hybrid computational approach to process real-time streaming multi-sources data and improve classification for emergency patients triage services: moving forward to an efficient IoMT-based real-time telemedicine systems

Neural Computing and Applications, Journal Year: 2024, Volume and Issue: 36(17), P. 10109 - 10122

Published: Feb. 21, 2024

Language: Английский

---

Ensemble Meta-Learning using SVM for Improving Cardiovascular Disease Risk Prediction

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: May 19, 2024

Abstract Cardiovascular diseases (CVDs) remain a leading cause of mortality worldwide, posing significant public health challenge. Early identification individuals at high risk CVD is crucial for timely intervention and prevention strategies. Machine learning techniques are increasingly being applied in healthcare their ability to uncover complex patterns within large, multidimensional datasets. This study introduces novel ensemble meta-learning framework designed enhance cardiovascular disease (CVD) prediction. The strategically combines the predictive power diverse machine algorithms – logistic regression, K nearest neighbors, decision trees, gradient boosting, gaussian Naive Bayes XGBoost. Predicted probabilities from these base models integrated using support vector as meta-learner. Rigorous performance evaluation over publicly available dataset demonstrates improved this approach compared individual. research highlights potential improve modeling healthcare.

Language: Английский

---