Artificial intelligence for the evaluation of profiles, trajectories and management of vulnerability in health: COVID-19, frailty of the elderly and cancer
COVIPREDS
Characterization and prediction of the occurrence of severe or lethal forms of COVID-19 using data from the AP-HP DHS
Underway
31/03/2020
-
01/10/2022
Main objectives
This project aims to improve knowledge of the profiles of patients hospitalized with COVID-19 and their management, by studying the characteristics at entry and the therapeutic management associated with the clinical course. The objective is to characterize the therapeutic management likely to limit the passages in intensive care and, in intensive care, to limit the number of deaths and the length of stay.
Translated with www.DeepL.com/Translator (free version)Un objectif secondaire est d’identifier les caractéristiques à l’entrée associées à un pronostic défavorable, afin que les cliniciens susceptibles d’orienter les patients bénéficient de l’information disponible sur le bénéfice attendu de la prise en charge et les incertitudes associées.
Translated with www.DeepL.com/Translator (free version)Un objectif secondaire est d’identifier les caractéristiques à l’entrée associées à un pronostic défavorable, afin que les cliniciens susceptibles d’orienter les patients bénéficient de l’information disponible sur le bénéfice attendu de la prise en charge et les incertitudes associées.

Auteur / Crédit Photo : François Marin/AP-HP
Challenge
The occurrence of severe or lethal forms of Covid-19
Methodology/Technology used
Data Mining
Expected outcomes
The expected clinical and epidemiological outcomes are:
- Identification of potentially effective or deleterious therapies for the management of patients with COVID-19 in conventional hospitalization and in intensive care
- Construction of predictive models of the expected prognostic benefit (vital status and length of stay) in case of admission to the ICU (potential decision support tool for clinicians).
The expected results in terms of methodology are:
- Development and implementation of generic methods for the analysis of care pathways using DHS data
- Identification of potentially effective or deleterious therapies for the management of patients with COVID-19 in conventional hospitalization and in intensive care
- Construction of predictive models of the expected prognostic benefit (vital status and length of stay) in case of admission to the ICU (potential decision support tool for clinicians).
The expected results in terms of methodology are:
- Development and implementation of generic methods for the analysis of care pathways using DHS data
Members
AP-HP, Inria & Centrale Supélec
Other projects
Description
Learning a deep representation of patient records for event prediction and patient segmentation
Names of partners involved
AP-HP, Inria