Big data and targeted machine learning in action to assist medical decision in the ICU - 02/11/18

Doi : 10.1016/j.accpm.2018.09.008

Romain Pirracchio ^a,^b,^c,^d,^⁎ , Mitchell J Cohen ^e, Ivana Malenica ^a, Jonathan Cohen ^a, Antoine Chambaz ^f, Maxime Cannesson ^g,^h, Christine Lee ^h, Matthieu Resche-Rigon ^d, Alan Hubbard ^a
the ACTERREA Research Group
^a Division of biostatistics, School of Public Health, university of California Berkeley, CA, USA
^b Department of anesthesia and perioperative medicine, university of California San Francisco, CA, USA
^c Service d’anesthésie-réanimation, hôpital Européen Georges-Pompidou, université Paris Descartes, Sorbonne Paris Cite, 75015 Paris, France
^d Service de biostatistique et informatique médicale, hôpital Saint-Louis, Inserm UMR-1153, université Paris Diderot, Sorbonne Paris Cite, 75010 Paris, France
^e Department of surgery, university of Colorado Denver, Colorado, USA
^f MAP5 (UMR CNRS 8145), université Paris Descartes, 75006 Paris, France
^g Department of anesthesiology and perioperative medicine, university of California Los Angeles, CA, USA
^h Department of bioengineering, university of California Irvine, CA, USA

^⁎Corresponding author at: Service d’anesthesie-réanimation, hôpital Européen Georges-Pompidou, 20, rue Leblanc, 75015 Paris, France.Service d’anesthesie-réanimationhôpital Européen Georges-Pompidou20, rue LeblancParis75015France

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Friday 02 November 2018
Cet article a été publié dans un numéro de la revue, cliquez ici pour y accéder

Abstract

Historically, personalised medicine has been synonymous with pharmacogenomics and oncology. We argue for a new framework for personalised medicine analytics that capitalises on more detailed patient-level data and leverages recent advances in causal inference and machine learning tailored towards decision support applicable to critically ill patients. We discuss how advances in data technology and statistics are providing new opportunities for asking more targeted questions regarding patient treatment, and how this can be applied in the intensive care unit to better predict patient-centred outcomes, help in the discovery of new treatment regimens associated with improved outcomes, and ultimately how these rules can be learned in real-time for the patient.

Le texte complet de cet article est disponible en PDF.