Development of a Scoring and Prediction Model for Weaning Success in ARDS Patients Using Ventilation Parameters Combined with Artificial Intelligence and Deep Learning Techniques

Overview

This study aims to develop an AI-supported scoring model to optimize the weaning processes of ARDS patients from mechanical ventilation. Retrospective analysis will be conducted on the data of 25,000 patients, focusing on ventilator parameters and hemodynamic variables. The model will be designed to contribute to clinical decision support systems.

The aim of this study is to develop an artificial intelligence and deep learning-supported scoring system using ventilator parameters obtained during the mechanical ventilation process in patients diagnosed with ARDS. This system seeks to predict and optimize the weaning process, facilitating successful liberation from mechanical ventilation. In this context, our study will analyze data from 25,000 patients obtained from the Metavision system. From this data pool, ARDS patients will be filtered and divided into two groups: those successfully weaned from mechanical ventilation (weaned) and those who were not (non-weaned). The ventilator parameters of both groups, including oxygenation indices, driving pressure, and total mechanical power, will be examined in detail. The collected data will be analyzed using artificial intelligence and deep learning algorithms to develop a scoring system capable of predicting patients' weaning processes. This system is designed to guide clinicians in patient management and enhance the success of weaning procedures. The results of this study aim to contribute to more efficient and safer management of the weaning process for ARDS patients. Furthermore, the implementation of AI-supported scoring systems in intensive care units is expected to promote widespread adoption and improve the quality of patient care.

Conditions