Machine Learning Methods for Precision Medicine Research Designed to Reduce Health Disparities: A Structured Tutorial

Abstract

Precision medicine research designed to reduce health disparities often involves studying multi-level datasets to understand how diseases manifest disproportionately in one group over another, and how scarce health care resources can be directed precisely to those most at risk for disease. In this article, we provide a structured tutorial for medical and public health research­ers on the application of machine learning methods to conduct precision medicine research designed to reduce health dispari­ties. We review key terms and concepts for understanding machine learning papers, including supervised and unsupervised learning, regularization, cross-validation, bagging, and boosting. Metrics are reviewed for evaluating machine learners and major families of learning approaches, including tree-based learning, deep learning, and ensemble learning. We highlight the advan­tages and disadvantages of different learning approaches, describe strategies for interpret­ing “black box” models, and demonstrate the application of common methods in an example dataset with open-source statistical code in R.Ethn Dis. 2020;30(Suppl 1):217-228; doi:10.18865/ed.30.S1.217

Machine learning for the prediction of preclinical airway management in injured patients: a registry-based trial

“…Therefore, the RF is robust to outliers, works well with nonlinear data, and has a lower risk of overfitting than single decision trees. As a result, the RF could handle even the relatively small prevalence of airway management cases in the test sets, achieved a good PRC area, and had a robust performance [ 15 , 25 ]. Given the prevalence between the different test sets, the RFs differ, and a final model cannot be given.…”

“…As a result, the RF could handle even the relatively small prevalence of airway management cases in the test sets, achieved a good PRC area, and had a robust performance. 15,25 Given the prevalence between the different test sets, the RFs differ, and a final model cannot be given.…”

Phenotyping of ABCA4 Retinopathy by Machine Learning Analysis of Full-Field Electroretinography

“…Ensemble learning is a meta-learning approach that combines decisions from several base learning models to improve the final prediction performance. 16 , 17 This effectively leverages differences in the way the base models are learning predictive features, by allowing each base model to contribute output class probabilities in a vote. The ensemble in this study comprised three base models, including the support vector machine (SVM), adaboost with decision trees, and logistic regression, extended hierarchically in two steps.…”

Accelerating health disparities research with artificial intelligence

“…Traditional approaches to address health disparities often fail to capture the issue's complexity and leave many gaps unresolved ( 11 ). By analyzing large datasets for patterns, correlations, or predictive markers that are difficult to identify through traditional analyses, AI can help us uncover hidden mechanisms and underpinnings of health disparities ( 12 , 13 ). AI has many advantages compared with traditional strategies for addressing health disparities, notably in its ability to uncover unexpected correlations and relationships that have remained unidentified in human-driven analyses, offering new insights.…”

“…There is a need to establish stringent ethical guidelines and frameworks for designing AI algorithms. Prioritization of transparency, fairness, and accountability throughout the AI development and implementation process is key ( 8 , 13 , 18 , 31 , 32 ). We must leverage tools like the AI Fairness Project to mitigate biases and discrimination in AI algorithms used within healthcare.…”