SIIM-FISABIO-RSNA COVID-19 Detection Radiograph Dataset
| Modality | Data Format | Publisher | Licence |
| Radiograph | DICOM | SIIM-FISABIO-RSNA COVID-19 Detection Radiograph Dataset | Non-Commercial |
COVID-19, caused by the SARS-CoV-2 virus, is a respiratory illness that emerged in late 2019 and quickly escalated into a global pandemic. Compared to the seasonal flu, it is approximately five times more deadly and has profoundly impacted public health, economies, and daily life. Its symptoms vary widely, from mild respiratory issues to severe lung damage, leading to significant morbidity and mortality rates.
Traditional diagnostic methods for COVID-19 primarily include polymerase chain reaction (PCR) tests, which identify the virus’s genetic material but can take several hours or even days to yield results. Chest radiographs, on the other hand, can be obtained more quickly but are often difficult to interpret accurately due to similarities between COVID-19 and other types of pneumonia.
About COVID-19 Detection Radiograph Dataset
The SIIM-FISABIO-RSNA COVID-19 Detection Dataset represents a critical advancement in the ongoing struggle to diagnose and manage COVID-19 more efficiently. It comprises a large collection of 6,334 chest scans in DICOM (Digital Imaging and Communications in Medicine) format. This standard allows for the inclusion of rich metadata alongside the image itself.
A panel of experienced radiologists has meticulously labeled these scans to identify both the presence of opacities and the overall appearance of the lung tissue. The dataset is geared towards aiding the development of computer vision models capable of detecting and localizing COVID-19-related abnormalities in chest radiographs.