Predicting the Classification of Heart Failure Patients Using Optimized Machine Learning Algorithms

Project Code :TCMAPY2229

Objective

The objective of this project is to develop a robust machine learning framework capable of predicting the survival outcomes of heart failure patients, thereby assisting healthcare professionals in making timely and informed decisions. By utilizing clinical data that includes patient attributes such as age, ejection fraction, serum creatinine levels, and other relevant health metrics, the project seeks to build an accurate prediction model that can be deployed in real-world clinical settings. The goal is to evaluate and compare the performance of three widely used machine learning algorithms—Random Forest Regressor, XGBoost, and Support Vector Regression—against one another to identify the most effective approach. Furthermore, the project emphasizes the optimization of model hyperparameters using Adaptive Inertia Weight Particle Swarm Optimization (AIW-PSO), a technique designed to enhance predictive performance by adjusting the exploration and exploitation phases in the search space. This will ensure that the models are optimized to their full potential and can deliver highly accurate predictions with minimal computational effort.

Abstract

Heart failure (HF) remains a major cause of morbidity and mortality, making the prediction of patient survival outcomes essential for timely clinical interventions. This study proposes an optimized machine learning approach for predicting heart failure survival using Random Forest Regressor (RFR), XGBoost, and Support Vector Regression (SVR). The dataset, sourced from Kaggle, includes clinical features such as age, ejection fraction, and serum creatinine levels for 299 heart failure patients. To address data imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was employed. Hyperparameters for each model were optimized using the Adaptive Inertia Weight Particle Swarm Optimization (AIW-PSO) algorithm, which effectively balanced exploration and exploitation in parameter tuning. The performance of the optimized models was evaluated, with the XGBoost model achieving the highest predictive accuracy. The study highlights the significance of hyperparameter optimization using AIW-PSO in improving the predictive power of machine learning models for clinical decision-making. The results demonstrate the potential of machine learning techniques in enhancing heart failure management, offering clinicians a reliable and interpretable tool for predicting patient outcomes.

Key words: Heart Failure Prediction, Machine Learning Algorithms, Clinical Prediction Models, Survival Prediction, Predictive Modeling,Medical Decision Support, Heart Failure Management.

NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.