aiqbench analysis
This research demonstrates how binarizing survival outcomes in clinical machine learning discards critical temporal information and censoring details, leading to suboptimal feature selection compared to time-to-event methods like the Cox partial log-likelihood. By integrating these advanced statistical formulations directly into Bayesian network architectures, practitioners can recover prognostic signals that standard binary approaches miss across various cancer cohorts. The study provides a methodological argument for adopting survival-aware modeling as the default practice in medical AI development to preserve data integrity and predictive accuracy.
The Cost of Binarizing Survival Outcomes in Clinical Prognostic Modeling
arXiv:2608.04046v1 Announce Type: cross Abstract: Survival analysis is an established framework for analyzing time-to-event data, yet many clinical machine learning studies still binarize the outcome before model training. This practice excludes censored patients, collapses temporal information into a single threshold, and can affect which features are selected as prognostically relevant. We examine the cost of this binarization in the context of Bayesian network (BN) feature selection, using two recent publications as case studies: one that applies BN-based feature selection to a head-and-neck cancer cohort and a second surgical cohort study that, while not BN-based, likewise binarizes its survival endpoint. We replace the binary scoring function with the Cox partial log-likelihood for feature-to-outcome edges, a modification we call the Survival-Aware Bayesian network, and recover prognostic features that binarization misses. Our ablation experiment confirms that the improvement is driven by the time-to-event scoring formulation rather than by retaining more patients. The results generalize across five endpoint-cohort combinations in head-and-neck cancer and extend to three further cancer types (breast, colorectal, and kidney). We propose that clinical studies with survival outcomes should use time-to-event methods by default, as binarization discards the prognostic signal retained by survival analysis.
Abstract:Survival analysis is an established framework for analyzing time-to-event data, yet many clinical machine learning studies still binarize the outcome before model training. This practice excludes censored patients, collapses temporal information into a single threshold, and can affect which features are selected as prognostically relevant. We examine the cost of this binarization in the context of Bayesian network (BN) feature selection, using two recent publications as case studies: one that applies BN-based feature selection to a head-and-neck cancer cohort and a second surgical cohort study that, while not BN-based, likewise binarizes its survival endpoint. We replace the binary scoring function with the Cox partial log-likelihood for feature-to-outcome edges, a modification we call the Survival-Aware Bayesian network, and recover prognostic features that binarization misses. Our ablation experiment confirms that the improvement is driven by the time-to-event scoring formulation rather than by retaining more patients. The results generalize across five endpoint-cohort combinations in head-and-neck cancer and extend to three further cancer types (breast, colorectal, and kidney). We propose that clinical studies with survival outcomes should use time-to-event methods by default, as binarization discards the prognostic signal retained by survival analysis.
Quantitative Methods (q-bio.QM); Machine Learning (cs.LG)
arXiv:2608.04046 [q-bio.QM]
(or arXiv:2608.04046v1 [q-bio.QM] for this version)
https://doi.org/10.48550/arXiv.2608.04046 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 283 words
Related coverage
Large Language Models are Approximate Survival Estimators
arXiv · Oct 1, 2026
How Guardoc transforms medical document processing with Amazon Nova models
AWS · Jul 27, 2026
A Multimodal Foundation Model for Longitudinal Patient Representation and Scalable Insight Generation in Oncology
arXiv · Aug 26, 2026
Models Can't Remember Their Training. Neither Can You.
LessWrong · Jul 19, 2026