Description du projet
Décisions d’assistance pulmonaire entraînées par l’IA
Un ventilateur mécanique est une machine qui assure la respiration d’une personne lors d’une intervention chirurgicale ou aux urgences lorsque celle-ci n’est pas en mesure de respirer par elle-même. La ventilation mécanique traditionnelle utilise un tube que l’on insère dans la bouche ou le cou du patient. L’assistance respiratoire extracorporelle (ARE) représente une autre solution pour assister la respiration, car elle facilite l’oxygénation du sang et l’élimination du CO2. L’objectif du projet IntelliLung, financé par l’UE, est de développer un système d’assistance par l’IA visant à optimiser la prise de décisions concernant la ventilation mécanique et l’ARE, afin d’améliorer les soins dispensés aux patients. Il convient de noter que le système aidera les soignants à expliquer aux patients et à leurs proches les stratégies de gestion choisies et fournira des informations sur des maladies particulières. IntelliLung réalisera une évaluation multicentrique du système d’aide à la décision généré.
Objectif
Invasive mechanical ventilation (MV) is one of the most important and life-saving therapies in the intensive care unit (ICU). In most severe cases, when MV alone is insufficient, extracorporeal lung support (ELS) is initiated. However, MV is recognised as potentially harmful, because inappropriate MV settings in ICU patients are associated with organ damage, contributing to disease burden. Studies revealed that MV is often not properly provided despite clear evidence and guidelines. Furthermore, treatment decisions by the healthcare providers, especially regarding MV and ELS, often remain incomprehensible to the patients and their relatives, since flow of information from caregivers to patients is challenged by a number of factors, including limited time and resources, communication problems as well as patients? capacity to comprehend and memorise information. The project proposed herein aims at clinically validating and extending our IntelliLung Artificial Intelligence Decision Support System (AI-DSS) designed to optimise MV and ELS settings to improve the care of ICU patients, alongside caregiver-patient communication. Thereby, best practice evidence-based MV and ELS within safer therapy corridors for longer periods, faster weaning from MV, and improved survival could be achieved - even in non-experienced hands. Additionally, this project will improve the information flow from caregivers to patients and relatives in the ICU setting. Therefore, we will develop a digital solution that allows automatic generation of an extensive plain-language information package for patients and their relatives, communicating highly individualised information on diseases and knowledge-based disease-management strategies, thus facilitating high-quality current and subsequent care through health literacy empowerment and patient-centredness. We will perform a retrospective and prospective multi-centre study to validate our IntelliLung AI-DSS and the patient information software.
Champ scientifique
Mots‑clés
Programme(s)
Régime de financement
HORIZON-RIA - HORIZON Research and Innovation ActionsCoordinateur
01069 Dresden
Allemagne