Lab News
Leading the AI Healthcare Revolution
Leading the AI Healthcare Revolution: Insights from California's Policy Discussion
Tina Hernandez-Boussard, PhD, was a featured AI expert at a recent policy discussion in Sacramento on shaping the AI revolution in healthcare care through innovated California policies. This discussion was led by Senator Josh Becker. Other experts included Arpit Davé of Amgen, and Joy Sacmar of Johnson & Johnson Robotics & Digital Solutions. The goal is to provide a framework to navigate challenges and seize opportunities to enhance patient care and drive medical advancements guided by AI.
Jan 2024
Paper on Bias Risks Cited by
White House AI Fact Sheet
A recently published paper, "Guiding Principles to Address the Impact of Algorithm Bias on Racial and Ethnic Disparities in Health and Health Care", of which Tina Hernandez-Boussard, PhD, was a co-author, has been cited by a White House Fact Sheet released Jan. 29, 2024. This sheet shares progress on President Biden's Executive Order to, "...ensure that America leads the way in seizing the promise and managing the risks of. artificial intelligence." Visit the below link for the original article.
Recent Publications
Sequence Modeling with Evo
In the ground-breaking preprint titled, "Sequence Modeling and Design from Molecular to Genome Scale with Evo,” we discuss how advances in machine learning combined with massive datasets of whole genomes (which encode DNA, RNA, and proteins) enable a biological foundation model that accelerates the mechanistic understanding and generative design of complex molecular interactions. Our team's contribution to this pioneering work encompasses the development of ethical considerations surrounding the application of Evo, a genomic foundation model boasting 7 Billion parameters. These ethical guidelines ensure that the advancements in multi-modal and multiscale learning facilitated by Evo are leveraged responsibly. Through our involvement, we highlight the importance of ethical practices in the cutting-edge domain of genomic modeling, ensuring that the potentials of Evo are realized with utmost consideration for ethical implications.
Feb 2024
Predictive models show promise in healthcare, but their successful deployment is challenging due to limited generalizability. Current external validation often focuses on model performance with restricted feature use from the original training data, lacking insights into their suitability at external sites. Our study introduces an innovative methodology for evaluating features during both the development phase and the validation, focusing on creating and validating predictive models for post-surgery patient outcomes with improved generalizability.