Personalised dynamic super learning: an application in predicting hemodiafiltration convection volumes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chatton, Arthur, Bally, Michèle, Lévesque, Renée, Malenica, Ivana, Platt, Robert W., Schnitzer, Mireille E.
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917867699568640
author Chatton, Arthur
Bally, Michèle
Lévesque, Renée
Malenica, Ivana
Platt, Robert W.
Schnitzer, Mireille E.
author_facet Chatton, Arthur
Bally, Michèle
Lévesque, Renée
Malenica, Ivana
Platt, Robert W.
Schnitzer, Mireille E.
contents Obtaining continuously updated predictions is a major challenge for personalised medicine. Leveraging combinations of parametric regressions and machine learning approaches, the personalised online super learner (POSL) can achieve such dynamic and personalised predictions. We adapt POSL to predict a repeated continuous outcome dynamically and propose a new way to validate such personalised or dynamic prediction models. We illustrate its performance by predicting the convection volume of patients undergoing hemodiafiltration. POSL outperformed its candidate learners with respect to median absolute error, calibration-in-the-large, discrimination, and net benefit. We finally discuss the choices and challenges underlying the use of POSL.
format Preprint
id arxiv_https___arxiv_org_abs_2310_08479
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Personalised dynamic super learning: an application in predicting hemodiafiltration convection volumes
Chatton, Arthur
Bally, Michèle
Lévesque, Renée
Malenica, Ivana
Platt, Robert W.
Schnitzer, Mireille E.
Methodology
Applications
Machine Learning
Obtaining continuously updated predictions is a major challenge for personalised medicine. Leveraging combinations of parametric regressions and machine learning approaches, the personalised online super learner (POSL) can achieve such dynamic and personalised predictions. We adapt POSL to predict a repeated continuous outcome dynamically and propose a new way to validate such personalised or dynamic prediction models. We illustrate its performance by predicting the convection volume of patients undergoing hemodiafiltration. POSL outperformed its candidate learners with respect to median absolute error, calibration-in-the-large, discrimination, and net benefit. We finally discuss the choices and challenges underlying the use of POSL.
title Personalised dynamic super learning: an application in predicting hemodiafiltration convection volumes
topic Methodology
Applications
Machine Learning
url https://arxiv.org/abs/2310.08479