MathHealth Seminar 

7th December

In person: 14:00-15:00, Seminar Room, Buiding VII, NOVA FCT

Online: ZOOM

https://videoconf-colibri.zoom.us/j/92057139184?pwd=dXRPZkNQV1hrMzdpZ0dPK0pVZlUzQT09

Speaker: Vanda Inácio, School of Mathematics, University of Edinburgh, Scotland

Title: The underlap coefficient as measure of a biomarker’s discriminatory ability in a multi-class disease setting

Abstract: The first step when evaluating a diagnostic test is to determine the variation in its values across different disease groups. In the three-class disease setting, the volume under the receiver characteristic surface (VUS) and the three-class Youden index (YI) are the commonly used summary measures of a test’s discriminatory ability. However, these measures are only appropriate under a stochastic ordering assumption for the distributions of test outcomes in the three groups. This assumption is stringent, not always plausible, and its violation can lead to incorrect conclusions about a test’s performance to distinguish between the three classes. To address this, we propose the underlap coefficient, study its properties, as well as its relationship with the VUS and YI when a stochastic order is enforced. We further propose Bayesian nonparametric estimators for both the unconditional underlap coefficient and for its covariate-specific version. A simulation study reveals a good performance of the proposed estimators across a range of conceivable scenarios. We have applied our methods to an Alzheimer’s disease (AD) dataset to assess how different potential AD biomarkers distinguish between individuals with normal cognition, mild impairment, and dementia, and how age and gender impact this discriminatory ability.

Short Bio: Vanda Inacio is Reader in Statistics at the School of Mathematics of the University of Edinburgh since 2016. Previously, she was an Assistant Professor at PUC Chile (2012–2016). Vanda received a PhD in Statistics from Universidade de Lisboa and a BSc in Applied Mathematics from Universidade Nova de Lisboa. Her main research interests are Bayesian (nonparametric) statistics, computational statistics, and biostatistics, with an emphasis on the statistical evaluation of medical tests. Vanda’s work has been published in some of the top-tier journals in the field, like Annals of Applied Statistics, Bayesian Analysis, Biostatistics, Biometrics, Statistics in Medicine, Statistical Science, and The American Statistician. Vanda is also a co-author of the R package ROCnReg, the only package integrating both frequentist and Bayesian methods for estimation of ROC curves (with and without covariates).