Imaging data sets (artificial intelligence)

Changed by Andrew Murphy, 21 Jan 2021

Updates to Article Attributes

Body was changed:

The aggregation of an imaging data set is a critical step in building artificial intelligence (AI) for radiology. Imaging data sets are used in various ways including training and/or testing algorithms. Many data sets for building convolutional neural networks for image identification involve at least thousands of images but smaller data sets are useful for texture analysistransfer learning, and other programs. 

Many commercial AI products are built on proprietary data sets or specific hospital data sets not available due to concerns over patient privacy. There are however several imaging data sets of radiological images and/or reports publicly available at the following websites:

Additionally, The Cancer Imaging Archive contains links to many open radiology data sets including the following:

  • -<a title="COVID-19 Open Annotated Radiology Database (RICORD)" href="https://pubs.rsna.org/doi/10.1148/radiol.2021203957">COVID-19 Open Annotated Radiology Database (RICORD)</a> expert annotated COVID-19 imaging dataset. 1000 chest x-rays and 240 thoracic CT exams</li>
  • +<a href="https://pubs.rsna.org/doi/10.1148/radiol.2021203957">COVID-19 Open Annotated Radiology Database (RICORD)</a> expert annotated COVID-19 imaging dataset. 1000 chest x-rays and 240 thoracic CT exams</li>
  • +<li>
  • +<a title="RSNA Pulmonary Embolism CT (RSPECT) dataset" href="https://pubs.rsna.org/doi/full/10.1148/ryai.2021200254">RSNA Pulmonary Embolism CT (RSPECT) dataset</a> 12,000 CT studies</li>

ADVERTISEMENT: Supporters see fewer/no ads

Updating… Please wait.

 Unable to process the form. Check for errors and try again.

 Thank you for updating your details.