The RSNA International COVID-19 Open Radiology Database (RICORD)

The coronavirus disease 2019 (COVID-19) pandemic is a global health care emergency. Although reverse-transcription polymerase chain reaction testing the reference standard method to identify patients with COVID-19 infection, chest radiography and CT play vital role in detection management of these patients. Prediction models for imaging are rapidly being developed support medical decision making. However, inadequate availability diverse annotated data set has limited performance generalizability existing models. To address this unmet need, RSNA Society Thoracic Radiology collaborated develop International Open Database (RICORD). This database first multi-institutional, multinational, expert-annotated set. It made freely available machine learning community as research educational resource imaging. Pixel-level volumetric segmentation clinical annotations was performed by thoracic radiology subspecialists all COVID-19–positive scans. labeling schema coordinated other international consensus panels annotation efforts, European Medical Imaging Informatics, American College Radiology, Association Physicists Medicine. Study-level classification labels radiographs were three radiologists, majority vote adjudication board-certified radiologists. RICORD consists 240 scans 1000 contributed from four sites. anticipated that will ideally lead prediction can demonstrate sustained across populations systems. © RSNA, 2021 Online supplemental material article. See also editorial Bai Thomasian issue.