Instant identification of biomarkers of COVID 19 by applying AI on structured reporting of Chest CT integrating clinical information

clinical trial - Current evidence suggests that chest CT imaging may be an extremely valuable tool in the diagnosis, epidemiology, and therapy response control of COVID-19 cases. It offers high sensitivity, short turnaround times and wide availability, and may thus complement RT-PCR tests, especially in situations of unclear clinical presentation, such as a negative RT-PCR despite strong anamnestic evidence for COVID-19. More importantly, it may offer opportunities to directly assess the stage of progression of the disease as observed directly from the affected lung tissue, and may thus be a method of choice for therapy response assessment in upcoming trials of new therapeutic agents. However, in order to develop it into a suitable tool for these purposes, a reproducible, standardized, and quantitative approach to image diagnostics is required. This proposal aims to develop such a standardized diagnostic and staging procedure for COVID-19 cases, using an AI-supported approach to select clinical attributes that serve as optimal predictors of the presence and stage of the disease.