CO10.3 - Personalizing renal replacement therapy initiation in the intensive care unit: a statistical reinforcement learning-based dynamic strategy with external validation on the AKIKI randomized controlled trials

Autor:	Grolleau, F., Petit, F., Gaudry, S., Diard, E., Quenot, J., Dreyfuss, D., Viet-Thi, T., Porcher, R.
Zdroj:	Revue d Epidemiologie et de Sante Publique; May 2023, Vol. 71 Issue: 1, Number 1 Supplement 2
Abstrakt:	There has been a growing demand for the development of validated data-driven decision support tools for the personalization of renal replacement therapy (RRT) initiation in intensive care unit (ICU) patients with acute kidney injury. Trials sequentially randomizing patients each day have never been conducted for RRT initiation. We used clinical data from routine care and trials to learn and validate optimal dynamic strategies for RRT initiation in the ICU.
Databáze:	Supplemental Index
Externí odkaz:	Zobrazit plný text záznamu Full Text from ScienceDirect