Coronavirus: Ventilator Outcomes Using Artificial Intelligence Chest Radiographs & Other Evidence-based Co-variates
- Conditions
- PneumoniaCovid19
- Registration Number
- NCT04855539
- Lead Sponsor
- King's College Hospital NHS Trust
- Brief Summary
We will determine ventilator outcomes to Coronavirus Infectious Disease 2019 (COVID-19) using artificial Intelligence with inputs of chest radiographs and other evidence-based co-variates.
- Detailed Description
The chest radiograph (chest x-ray) has emerged as the United Kingdom's National Health Service (NHS) frontline diagnostic imaging test for COVID-19, in conjunction with clinical history and key blood markers: C-reactive protein (CRP) and lymphopenia. Typically, every suspected COVID-19 patient presenting to the emergency department is undergoing blood tests and a chest radiograph. Therefore, it has become critical for radiologists to review and "hot" report the chest x-ray urgently.
Primary Objective: Use chest radiographs and clinical data to determine whether patient can survive with a ventilator
Recruitment & Eligibility
- Status
- UNKNOWN
- Sex
- All
- Target Recruitment
- 300
Admitted to intensive care unit (ITU) or equivalent COVID-19 polymerase chain reaction (PCR) positive
No imaging prior to ITU admission
Study & Design
- Study Type
- OBSERVATIONAL
- Study Design
- Not specified
- Primary Outcome Measures
Name Time Method Sensitivity and specificity of a convolutional neural network to predict survival outcome 1 month Defined by sensitivity, specificity, positive and negative predictive values
- Secondary Outcome Measures
Name Time Method
Trial Locations
- Locations (3)
Princess Royal University Hospital
🇬🇧Orpington, Kent, United Kingdom
Guy's and St Thomas' Hospital
🇬🇧London, United Kingdom
King's College Hospital
🇬🇧London, United Kingdom