Retrospective Study on the Accuracy of AI-Powered Reading of Chest X-Rays in the Diagnosis of COVID-19 Pneumonia in a Tertiary Hospital (CHERISH Study)

With the continuous escalation of COVID-19 cases worldwide, there is a need to simplify and expedite the workflow in detecting the infection. Exploring the accuracy of AI-powered chest x-ray in detecting COVID-19 pneumonia, this project provides a method that can augment COVID-19 screening. Implemented and completed in November 2020, the project produced three models of prototype applications to facilitate automated COVID-19 screening diagnosis. Dubbed as the CHERISH project (A Retrospective Study on the Accuracy of AI-Powered Reading of Chest X-Rays in the Diagnosis of COVID-19 Pneumonia in a Tertiary Hospital), a team of nine researchers led by Dr. Beatrice J. Tiangco from The Medical City used the clinical data of all patients admitted at a tertiary hospital during the COVID-19 pandemic from 1 March to 30 June 2020, to produce an application embedded with AI models to detect patients with COVID-19 pneumonia. Dr. Tiangco's co-investigators are Dr. Ron Yebes from the Department of Radiology, Dr. Christine Chavez and Dr. Miguel Angelo C. Manguiat from Adult Pulmonology, Medical City, Mr. Mario Domingo, CEO and founder of Domingo AI Research Center (DARC), Prof. Jason Albia, the lead researcher of DARC and an associate professor from the Institute of Mathematical Sciences and Physics, College of Arts and Sciences, University of the Philippines Los Banos, Lei Rigi Baltazar, Joverlyn Gaudillo, and Mojhune Gabriel Manzanillo, data scientists also from DARC.