Thursday, May 1st
11:45am - 1:00pm, ARB Hess Commons
Qing Pan, PhD
Professor, Department of Biostatistics and Bioinformatics
George Washington University, Milken Institute School of Public Health
Predictions of Advanced Adenoma and High-Risk Pregnancies in Longitudinal Screening Studies
Panel count data is common in cancer screening. In the context of colorectal cancer screening, our work focuses on the prediction of the probability of advanced adenoma conditional on patient-level risk factors and/or event history. We implement the joint frailty model proposed by Huang et al. (2006), which involves a non‐stationary Poisson process for recurrent adenoma events and informative screening time using semi‐parametric Cox models correlated by a latent frailty variable. Coefficients and baseline intensity functions are estimated through estimating equations. The subject-specific frailty value is estimated by the borrow‐strength method (Huang and Wang 2004). In addition, marginal models for the adenoma and screening events are also applicable when average covariate effects on the population level are of interest. Predictions of individual risks based on the marginal model and predictions based on the frailty models for patients with or without screening history are compared. When a patient’s screening history is available and sufficient adenoma events are observed, the predictions based on the frailty model with estimated subject‐specific frailty are superior. However, in the cases of early censoring when adenoma events are not observed for most patients or screening history is not available, the prediction based on the marginal model has better performance. For future screening, the individualized screening intervals based on the dynamic predictions of advanced adenoma risks will detect adenomas earlier with shorter lag times between adenoma occurrences compared to the current practice of fixed screening intervals for all.
In a separate project, machine learning and deep learning models to identify pregnancies with elevated risks of adverse outcomes are compared. A novel GRU model that accommodates both static and time-varying information and allows interactions between these two kinds of covariates through additional attention layers provides better performance. Contributions of various types of covariates (questionnaires, blood tests, and ultrasound) to the prediction accuracy are compared for clinical practice in low- and middle-income countries.
Biostatistics Departmental Seminars & Lectures
Lectures are in-person only unless marked otherwise.
For all Zoom inquiries to attend seminars with a listed virtual option, please send an email to Erin Elliott, Programs Coordinator (ee2548@cumc.columbia.edu).
During the Fall and Spring semesters, the Department of Biostatistics holds regular seminars on Thursdays, called the Levin Lecture Series, on a wide variety of topics which are of interest to both students and faculty. Over each semester, there are also often guest lectures outside the regular Thursday Levin Lecture Series, to provide a robust schedule the covers the wide range of topics in Biostatistics. The speakers are invited guests who spend the day of their seminar discussing their research with Biostatistics faculty and students.
