Understanding Potential Biases Inherent in Data

Описание: Dr. Gianfrancesco discusses the limitations of machine learning in population health research. She uses and example from a study of patients with rheumatoid arthritis to show how biases can affect study outcomes. She shows the limitations of commonly used correction methods, such as inverse probability weighting, in reducing these biases. She also discusses how biased machine learning algorithms can perpetuate health disparities.