This retrospective NIfTI image dataset consists of unenhanced chest CTs from 632 patients with COVID-19 infections. The images were acquired at the point of care in an outbreak setting from patients with Reverse Transcription Polymerase Chain Reaction (RT-PCR) confirmation for the presence of SARS-CoV-2.

Patients presented to a health care setting with a combination of symptoms, exposure to an infected patient, or travel history to an outbreak region. All patients had a positive RT-PCR for SARS-CoV-2 from a sample obtained within 1 day of the initial CT.  CT exams were performed without intravenous contrast and with a soft tissue reconstruction algorithm.  The DICOM images were subsequently converted into NIfTI format. 

A multidisciplinary team trained several models using portions of this TCIA data set, along with additional CTs and manually annotated images from other sources. A classification model derived in part from this data is described at:  https://doi.org/10.1038/s41467-020-17971-2.  The NVIDIA-related frameworks and models specific to this publication are available at no cost as part of the NVIDIA Clara Train SDK at https://ngc.nvidia.com/catalog/containers/nvidia:clara:ai-covid-19. This includes both inference-based pipelines for evaluation, as well as model weights for further training or fine tuning in outside institutions.  In addition, a web-based version of this model (for research use only) with drag-and-drop functionality for evaluating individual scans can be found at https://marketplace.arterys.com/model/nvidiacovidCT.  Uploading a CT yields a return email that contains results.