Hi, I'm Sunjae Bae.
I am an assistant professor of surgery and population health at NYU Grossman School of Medicine.
I mainly work on 3 research areas:
I analyze clinical data to improve decision-making in chronic kidney disease and kidney transplantation.
I develop methods to find the best-fitting drug based on the individual patient's characteristics (heterogeneous treatment effect) and to distill the "truth" from the data in hand (causal inference).
I use artificial intelligence to better solve these problems.
I received my PhD, MHS, and MPH at Johns Hopkins Bloomberg School of Public Health. Before Hopkins, I received a KMD at Kyung Hee University and saw patients, mostly at Mokpo National Hospital [Korean].
About
Education
PhD, Johns Hopkins Bloomberg School of Public Health, 2020.
MHS, Johns Hopkins Bloomberg School of Public Health, 2020.
MPH, Johns Hopkins Bloomberg School of Public Health, 2014.
KMD, Kyung Hee University, 2009.
Select Grants & Awards
Pre-Doctoral Fellowship Award. 2018-2019 (American Society of Nephrology) [interview with ASN]
Charlotte Silverman Fund. 2018 (Johns Hopkins Bloomberg School of Public Health)
Young Investigator Award. 2018 (American Transplant Congress)
Three Minute Thesis Competition, Finalist. 2018 (Johns Hopkins University)
Odd Gigs
I hosted a radio show on a national network. Check them out here [Korean].
I used to take landscape photos. This hobby turned into portraits when my kid was born.
Select Publications
Full list available on PubMed and Google Scholar.
[Am J Transplant, 2018] Who Can Tolerate a Marginal Kidney? Predicting Survival After Deceased-Donor Kidney Transplantation by Donor-Recipient Combination.
In kidney transplantation (KT), both donor quality and recipient health condition are key determinants of survival after KT. Understanding the interaction between the two factors can improve donor-recipient matching. We estimated post-KT survival using random forests, a machine learning algorithm that can reflect complex interactions between predictors with minimal assumptions. Also check out our online tool!
[Am J Transplant, 2016] Changes in discard rate after the introduction of the Kidney Donor Profile Index (KDPI).
We assessed how providing a numeric risk score of donor health conditions influenced clinical practice in kidney transplantation using a national dataset (n=57657). Our findings suggest an adverse “labeling effect” of the score: in other words, the improved risk index might have prompted clinicians to be more risk-averse.
