Artificial Intelligence - SARS-CoV-2 (COVID-19) Risk Evaluation (AI-SCoRE)

San Donato Group (GSD)

Status: Completed

Conditions: Covid19

Study type: Observational

Funder types: Other; Industry

Identifiers

NCT04834934

AI-SCoRE

Details and patient eligibility

About

The management of COVID-19 patients in overwhelmed hospital facing the pandemic is a clinical challenge.

The improvement of decision making may allow a better allocation of available resources and a better treatment of patients at higher risk.

Chest CT has been widely adopted for COVID-19 pneumonia diagnosis. Several experiences documented the capability of Artificial Intelligence to improve and fasten COVID-19 pneumonia detection, mainly using chest X-ray.

Aim of the present study was to develop and validate an Artificial Intelligence approach integrating clinical and imaging data (automatically extracted through the adoption of dedicated neural networks) for the creation of a cloud platform capable of performing automatic patients risk stratification. Such an approach could be used for triage of COVID-19 patients in the emergency department, with the aim to improve healthcare personnel decision-making and allocation of resources during health emergencies.

Enrollment: 2,000 patients

Sex: All

Ages: 18 to 100 years old

Volunteers: No Healthy Volunteers

Inclusion criteria

• confirmed SARS-CoV-2 infection with RT-PCR
• non contrast chest CT scan performed within 72 hours after admission to the emergency department

Exclusion criteria

• age < 18 ys

Trial design

2,000 participants in 2 patient groups

COVID-19 first wave patients

Description:

1700 patients retrospectively enrolled in 15 Italian hospitals from 16/2/2020 to 29/4/2020.

COVID-19 second wave patients

Description:

300 patients prospectively enrolled in IRCCS San Raffaele Hospital from 19/10/2020 to 31/12/2020.