This dataset Image AddedThese retrospective NIfTI image datasets consists of unenhanced chest CT images of CTs: 

• First dataset - from 632 patients with COVID-19

• infections at

• initial point of care, and
• Second dataset - a second set of 121 CTs from 29 patients with COVID-19 infections with serial / sequential CTs.

The initial images for both datasets were acquired at the point of care in an outbreak setting from patients with Reverse Transcription Polymerase Chain Reaction (RT-PCR) in an outbreak setting with NIFTI files without annotations or bounding boxes. The images were retrospectively acquired after CT of patients with RT-PCR confirmation for the presence of SARS-CoV-2.

NIFTI CT images were converted from DICOM images. CT reconstruction algorithm was soft tissue without intravenous contrast.  

Image AddedPatients presented to a health care setting with either symptoms &/or 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 . There may be follow-up versions of this data set which include more sources of CT data or metadata, or annotated data to potentially make this dataset a useful tool and resource for developing algorithms for medical applications or data analysis challengesexams were performed without intravenous contrast and with a soft tissue reconstruction algorithm.  The DICOM images were subsequently converted into NIfTI format. The second dataset also had other follow up CTs, in addition to the initial point of care CT.

A multidisciplinary team trained several models using portions of this data set (the first dataset, along with additional CTs and manually annotated images and other data &/or other CT’s)from other sources. A classification model derived in part from portions of this data (& other data) can be found the first dataset is described in a Nature Communications manuscript at:  https://wwwXXX  or [DOI pubmed PMID].EMBARGOED FOR NOW / WILL PROVIDE WHEN EMBARGO LIFTEDModels partly derived from portions of this (& other) data may be found at:  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/resourcescontainers/nvidia:clara:clara_ai_covid19_pipelineA web-based research-only model (website for research use only) has CT drag-and-drop functionality. Upload of CT yields a return email that contains the results. This model is also partly derived from portions of this (& other) data:  https://marketplace.arterys.com/model/nvidiacovidCT.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. The second data set of 121 serial / sequential CTs in 29 patients is reported in a Scientific Reports manuscript at  https://doi.org/10.1038/s41598-021-85694-5.