The ZODIAC Respiratory Disease Phenotype Observatory: an IAEA International Cooperative Study for Early Detection of New Pandemics (The IAEA CT Artificial Intelligence -Cooperative Study- ICAI Project)

Every zoonotic disease has two components: the animal and the human, with an interface in-between. Once the pathogen has crossed from animals to humans, it becomes a “human-disease” and has the potential to generate widespread infection and progress from an outbreak confined to one community or geographical location to a worldwide pandemic.
Each disease, e.g. Coronavirus disease 2019 (COVID-19), has radiologic characteristics, or "patterns" that, when precisely identified, are useful for selecting the appropriate method of management, predicting prognosis, and evaluating therapeutic response, thus providing valuable information to support a personalized medicine approach. Unfortunately, when facing an emerging disease, these patterns are not known.
Medical imaging studies have an important role in the diagnosis and follow-up of many infectious diseases. It became apparent early in the course of the COVID-19 pandemic that, although molecular analysis with reverse transcriptase polymerase chain reaction (RT-PCR) is the cornerstone of diagnosis, chest imaging including computed tomography (CT) scan and baseline X-ray also plays a critical role for several reasons, such as 1. The concern for less than optimal sensitivity of PCR testing, 2. The delay in molecular diagnosis due to a shortage of testing kits and laboratory personnel, and 3. The ability to evaluate and characterize the disease extent and multiple organ involvement, as well as long term consequences (sequela).
Medical imaging technologies such as CT positron emission tomography (PET) are able to detect lesions smaller than 0.5 mm or to assess biological activity in the body without a biopsy, allowing the identification of specific characteristics of the disease in each patient. Each study generates hundreds or even thousands of complex images, making analysis difficult to the naked eye.
Radiomics is a method which allows the extraction and analysis of hundreds of diagnostic imaging studies (e.g. CT) using characterization algorithms to extract valuable data, also known as ‘Big data’. It has the potential to discover typical findings of a disease that are not visible to the naked eye, which  increases diagnostic accuracy and greatly contributes to individualized therapy planning. More importantly, radiomics can now be complemented by the emerging fields of artificial intelligence, machine learning and deep learning techniques for further process automation.
This CRP will focus on evaluating the patterns of lung involvement in chest CT and x-ray studies of COVID-19 patients , with the goal of identifying specific characteristics associated with different virus variants and determine if there are racial, ethnic, regional, and clinical differences in the development of the disease complications and the specific medical imaging manifestations.