@inproceedings{a3d373aa5d624897ac16591adb9e9775,
title = "Generalizability of AI Survival Models in the Context of Preterm Birth Prediction",
abstract = "This study investigates whether integrating a Variational Information Bottleneck (VIB) into deep survival models improves generalization across heterogeneous populations. We evaluate performance on preterm birth prediction using three distinct datasets, training on one and testing on the others. Compared to classical and non-VIB deep models, the VIB-enhanced models achieved higher time-dependent AUC in several cases, suggesting potential benefits despite limited sample sizes.",
keywords = "distributional shift, generalizability, survival analysis",
author = "Anne Fischer and Isabelle Dehaene and Mark Hoogendoorn",
note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.; 23rd International Conference on Artificial Intelligence in Medicine, AIME 2025 ; Conference date: 23-06-2025 Through 26-06-2025",
year = "2025",
doi = "10.1007/978-3-031-95841-0\_29",
language = "English",
isbn = "9783031958403",
volume = "15735 LNAI",
series = "Lecture Notes in Computer Science",
publisher = "Springer Science and Business Media Deutschland GmbH",
pages = "154--159",
editor = "Riccardo Bellazzi and \{Juarez Herrero\}, \{Jos{\'e} Manuel\} and Lucia Sacchi and Bla{\v z} Zupan",
booktitle = "Artificial Intelligence in Medicine - 23rd International Conference, AIME 2025, Proceedings",
address = "Germany",
}