On the Generalizability of Machine Learning Classification Algorithms and Their Application to the Framingham Heart Study

The use of machine learning algorithms in healthcare can amplify social injustices and health inequities. While the exacerbation of biases can occur and be compounded during problem selection, data collection, and outcome definition, this research pertains to the generalizability impediments that occur during the development and post-deployment of machine learning classification algorithms. Using the Framingham coronary heart disease data as a case study, we show how to effectively select a probability cutoff to convert a regression model for a dichotomous variable into a classifier. We then compare the sampling distribution of the predictive performance of eight machine learning classification algorithms under four stratified training/testing scenarios to test their generalizability and their potential to perpetuate biases. We show that both extreme gradient boosting and support vector machine are flawed when trained on an unbalanced dataset. We then show that the double discriminant scoring of type 1 and 2 is the most generalizable with respect to the true positive and negative rates, respectively, as it consistently outperforms the other classification algorithms, regardless of the training/testing scenario. Finally, we introduce a methodology to extract an optimal variable hierarchy for a classification algorithm and illustrate it on the overall, male and female Framingham coronary heart disease data.