Biography

Rock Ethics Institute Faculty Fellowship Project

Building Ethical AI to Achieve Accuracy and Fairness in Genetic Risk Prediction of Human Diseases across Diverse Populations

 ABSTRACT

Predicting disease risk from an individual's DNA is a crucial step towards precision medicine. Artificial intelligence (AI) empowers such predictions but carries historical inequalities and algorithmic biases, thereby inducing moral concerns and exacerbating health disparities.

This project aims to develop innovative AI for effective and ethical prediction of genetic disease risk across diverse populations. To achieve accuracy and fairness, this project will address ethical risks in training data, modeling methodology, and benchmarking framework of machine learning (ML).

This project is expected to synergize ML, genomics, and ethics to advance responsible AI research in biomedicine, ultimately promoting health for all humanity. 

Xiang Zhu has been an assistant professor in the Department of Statistics and Huck Institutes of the Life Sciences since 2020 and a biostatistician (without compensation) at U.S. Department of Veterans Affairs Palo Alto Health Care System since 2018. He received his Ph.D. in Statistics from the University of Chicago in 2017, and he was a Stein Fellow at Stanford University in 2017–2020.

His research focuses on developing innovative statistical and computational methodology to gain novel biological and clinical insights into health and disease from large-scale and high-throughput genomic data collected in diverse human populations.