Intelligent Lung Support for Mechanically Ventilated Patients in the Intensive Care Unit

Informations projet

IntelliLung

N° de convention de subvention: 101057434

DOI

10.3030/101057434

Date de signature de la CE 24 Mai 2022

Date de début 1 Septembre 2022

Date de fin 31 Août 2027

Financé au titre de

• Health

Coût total

€ 5 980 347,50

Contribution de l’UE

€ 5 980 347,00

Coordonné par

TECHNISCHE UNIVERSITAET DRESDEN
Germany

Periodic Reporting for period 2 - IntelliLung (Intelligent Lung Support for Mechanically Ventilated Patients in the Intensive Care Unit)

Période du rapport: 2024-03-01 au 2025-08-31

Résumé du contexte et des objectifs généraux du projet

Mechanical ventilation remains a life-saving intervention for critically ill patients in intensive care units (ICUs). However, finding the optimal ventilator settings for each individual patient is highly complex. Suboptimal ventilation can risk further lung injury and complicate recovery. The IntelliLung project addresses this critical challenge by developing and validating a novel Artificial Intelligence Decision Support System (AI-DSS) designed to assist clinicians in optimizing ventilator settings.

The project's primary objective is to enhance patient safety and outcomes in critical care. By leveraging AI that learns from clinical data, the system aims to provide personalized ventilation recommendations, helping to reduce the risk of ventilator-induced complications. A parallel objective focuses on improving communication and health literacy. The project is developing digital tools to provide patients and their families with clear, individualized information about the treatment process and the role of AI assistance, thereby fostering understanding and trust during a vulnerable time.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

Since its start, the IntelliLung consortium has made significant strides in both technical development and clinical preparation. A foundational achievement has been the successful development and initial validation of the core AI algorithm using retrospective data. This work has demonstrated the system's potential to learn effective ventilation strategies.

Concurrently, the technical architecture of the integrated IntelliLung platform has been finalized. This includes the development of a secure communication middleware to connect with hospital equipment and a user-friendly interface for clinicians. Rigorous testing has been conducted to ensure the platform's robustness and reliability in a clinical environment.

On the clinical front, the multi-centre prospective study is now actively underway. Patient recruitment has successfully commenced at the lead clinical site, with additional sites finalizing their technical and ethical approvals to join the study imminently. The project's comprehensive, plain-language information materials for patients and relatives are being utilized in the active site, playing a crucial role in supporting the informed consent process and enhancing health literacy among participants and their families.

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

IntelliLung moves beyond static clinical guidelines by creating an adaptive AI system that can tailor its recommendations to the unique and changing physiological state of each patient. A key innovation is the project's dedicated focus on building a trustworthy AI. This involves integrating advanced techniques for explaining the AI's recommendations to clinicians and implementing methods to detect and mitigate potential biases in the algorithm, ensuring its recommendations are fair and equitable across different patient groups.
By combining a technically advanced AI-DSS with a strong emphasis on clinical usability, transparency, and patient communication, IntelliLung is pioneering a holistic approach to the integration of AI in critical care environments.