Robust Heart Disease Classification Model Based on Machine Learning

Torres J.; Ticona W. · Lecture Notes in Networks and Systems (2025) · DOI: 10.1007/978-3-031-96759-7_21

Resumen

Heart diseases are among the primary causes of morbidity and mortality worldwide. Over the years, the incidence of these diseases has been increasing due to various factors contributing to their global prevalence, mainly resulting from lifestyle choices, misdiagnoses, or incorrect treatments due to late identification of these diseases. Artificial intelligence technologies, particularly machine learning algorithms, have gained significant momentum in the medical field by enabling the creation of models that predict diseases to facilitate timely diagnoses and treatments. The proposed model predicts cardiac diseases using a robust machine learning-based classification approach. The methodology consists of four phases: Dataset acquisition; Preprocessing (Label Encoding, One-Hot Encoding, SMOTE, and Standardization); Model implementation (SVM, LR, NB, KNN, DNN, MLP, and SGD); Hyperparameter tuning (GridSearchCV); and Model Evaluation. The best results, obtained with an 80% training and 20% testing partition, were achieved using the MLP algorithm. Its metrics surpassed those of the other models, with an Accuracy of 94%, Precision of 93%, Recall of 93%, F1-Score of 93%, and an AUC value of 0.981 for the ROC curve. In conclusion, the results demonstrate that a well-applied machine learning algorithm is highly effective for predicting cardiac diseases using patient medical data, which can aid in facilitating timely diagnoses and treatments. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Instituciones: Universidad Tecnológica del Perú