AI-Powered Predictions: Identifying COVID Patients at Risk of Ventilator Dependence

Development of an AI Nomogram for COVID-19 Ventilator Prediction

Research Overview

Researchers at Case Western Reserve University have created an integrated clinical and CT-based AI nomogram, known as CIANE, designed to predict which COVID-19 patients may require ventilator support. This innovative tool aims to enhance the early identification of patients who could benefit from intubation and invasive mechanical ventilation, potentially reducing disease progression and mortality rates.

Study Details

The study involved a comprehensive chart review of 869 patients who tested positive for SARS-CoV-2 in January 2020. These patients were treated at Renmin Hospital of Wuhan University, Hubei General Hospital, and University Hospitals in Cleveland. All participants were in the mild stage of the disease and did not require any form of respiratory assistance at the time of evaluation. Baseline clinical characteristics and chest CT scans were collected for analysis.

Functionality of CIANE

The process begins with uploading a chest CT scan as a digitized image. CIANE utilizes advanced algorithms to detect and analyze specific features within the scans that may not be visible to the human eye. This analysis helps determine whether a patient is at an elevated risk of respiratory distress and may need ventilator support.

Clinical Implications

This prognostic tool is designed to aid healthcare professionals in identifying patients who require early interventions to mitigate disease progression. With increasing vaccination rates, the early identification of candidates for ventilator support remains crucial, especially in light of the ongoing global shortage of ventilators exacerbated by the emergence of the Delta variant.

Reference

Hiremath A, Bera K, Yuan L, Vaidya P, Alilou M, Furin J, Armitage K, Gilkeson R, Ji M, Fu P, Gupta A, Lu C, Madabushi A. Integrated Clinical and CT based Artificial Intelligence nomogram for predicting severity and need for ventilator support in COVID-19 patients: A multi-site study. IEEE J Biomed Health Inform. 2021 Aug 13; PP. doi: 10.1109/JBHI.2021.3103389. Epub ahead of print. PMID: 34388099.