An Intelligent and Automated Machine Learning-Based Approach for Heart Disease Prediction and Personalized Care

The heart, as the most vital organ, plays a crucial role in supplying blood to the brain and all other organs in the human body. However, heart and blood vessel damage can lead to serious conditions such as heart attacks, strokes, and heart failure. Heart failure in particular occurs when the heart cannot pump enough blood to meet the needs of the body. Anticipating severe heart failure based on clinical and diagnostic information could be revolutionary as it has the potential to lower the risk of mortality for patients. To achieve this goal, a comprehensive research study was conducted involving dataset analysis and pre-processing (oversampling, normalization, and feature selection) and the application of various machine learning algorithms (decision tree, support vector machine, naive bayes, XGBoost, random forest, and logistic regression). The objective was to precisely evaluate the likelihood of heart failure development using computerized prediction. Such predictions hold significant value in the medical field, benefiting both doctors and patients alike. The results of this study were truly remarkable, with the logistic regression emerging as the best-performing model among all examined algorithms. With a precision of 0.93, recall of 0.93, F1-score of 0.93, and an impressive maximum accuracy of 93.33%, the logistic regression model demonstrated its effectiveness and reliability. This model proved to be a powerful and valuable tool in diagnosing heart failure. Its remarkable performance serves as a testament to its efficacy and potential to revolutionize the prediction and prevention of severe heart failure, ultimately improving patient outcomes in the medical field.