Comparative Analysis of Machine Learning Techniques for Enhanced Predictive Modeling in Healthcare

Recently, there has been a surge in predictive modeling in healthcare because it can save lives through early diagnostics and treatments. Advances in algorithms and computing power can be harnessed to develop predictive models using Python-based programs and automated ML platforms. This study compares the performance of neural networks and automated ML models in predicting coronary heart disease using data from the UCI repository. We utilized Python-based neural networks and IBM Watson's Snap Logistic Regression, Extra Trees Classifier, and Logistic Regression. Our results show that IBM Watson's models achieved an accuracy of 86%, outperforming the neural network's 80%. This comparative analysis highlights the potential of automated ML platforms in enhancing predictive modeling in healthcare.