CLIP Colloquium: Jaan Altosaar (Columbia)
Zoom link: https://umd.zoom.us/j/98806584197?pwd=SXBWOHE1cU9adFFKUmN2UVlwUEJXdz09 (passcode if needed: clip)
Scaling health care delivery with machine learning
Abstract: American health care customers spend more for less. In this talk, we investigate the incentives that underpin our health care system and place patients at odds with both providers and payors who might need to increase revenue. For example, electronic health records sometimes serve billing purposes better than they serve care delivery purposes. We seek to contrast the flows of patient data through the American health care system with the transparent flow of data enabled by open source electronic health records. Further, we show that open source health care software can accelerate the development of clinical decision support built using machine learning. Such decision support can empower our health care heroes and enable better outcomes at lower cost. We propose an open source nonprofit foundation to build such technology equitably and sustainably for health care delivery worldwide.
Bio: Jaan Altosaar is a Postdoctoral Research Scientist at Columbia University Irving Medical Center focused on machine learning methods that improve the decisions people make in health and science. He completed his Ph.D. at Princeton University advised by David Blei and Shivaji Sondhi. During his Ph.D., he worked at the Center for Data Science at New York University, Google Brain, and DeepMind, and his work was supported by fellowships from the Natural Sciences and Engineering Research Council of Canada. Prior to Princeton, Jaan earned his B.Sc. in Mathematics and Physics from McGill University and founded usefulscience.org, a nonprofit that makes accessible the science that can improve daily life. He has built probabilistic machine learning methods for hospital readmission prediction, recommender systems, and statistical physics simulations, and is now building methods that address issues in mental health, health disparity, and women's health.
