Séminaire de Probabilités et Statistique :

Le 28 avril 2025 à 13:45 - UM - Bât 09 - Salle 109 (1er étage)


Présentée par Amico Maïlis - IDESP, Montpellier

Assessing cure status prediction from survival data using ROC curve



Survival analysis relies on the assumption that, if the follow-up would be long enough, the event of interest will eventually be observed for all observations. This assumption, however, is often not realistic because some subjects never experience the event of interest, that is, they are considered “cured”. A common approach to analysing this type of data consists in using cure models. Two types of information can be estimated from these models: the survival at a given time and the cure status, both possibly modelled as a function covariates. The cure status is often of interest to medical practitioners, and one is usually interested in predicting it based on markers. One way to evaluate the predicted performance of markers is Receiver Operating Characteristic (ROC) curves. In our case, however, the classical ROC curve method is not appropriate because the cure status is partially unobserved due to the presence of censoring in survival data. In this work, we propose a ROC curve estimator that aims to evaluate the cured/noncured status classification performance from cure survival data. This estimator, which handles the presence of censoring, decomposes sensitivity and specificity by means of the definition of conditional probability, and estimates these two quantities by means of weighted empirical distribution functions. Based on simulations, we demonstrate the good performance of the proposed method. Finally, we illustrate the methodology on a breast cancer dataset.

Séminaire en salle 109, également retransmis sur zoom : https://umontpellier-fr.zoom.us/j/7156708132



Retour