Annotated image data (anatomical location, pathology)

This is a dataset of medical images collected between 2010 and 2011 for the PACS dump in Vienna and 2006-2009 for the lung CT data, including:

  • DD1: Large scale realistic data dump from MUW / AKH PACS. Anonymized, with a rich variety of anatomical structures, anomalies, and modalities, this is the closest we will get to the real WP9 application scenario in the 3D part (MUW)
  • L1: Lung images. This is a controlled data set with annotation of anomalies/diseases to be used for the development of anomaly recognition/indexing. It encompasses approximately 100 anonymized Lung CTs, with known anomaly. (HES-SO)
  • FBCT1: Full body CT data. This is a controlled data set with annotation of anatomical structures (currently in progress) for use during the development of anatomical structure recognition and retrieval. (MUW, Univ. Washington)
  • TH1: gated lung CTs with contrast agent (aorta). approximately 30 keypoints, approximately 20 patients
  • ImageCLEF data set: containing images for the ImageCLEFmed competition, for many biomedical journals of BiomedCentral, so open access. (HES-SO)
  • Images for the MUW databases are described via visual feature vectors, that are descriptive and can be used to match anomalies/anatomical structures during retrieval.
  • Within the VISCERAL project, 3D radiology images have been manually annotated and will be released to the research community by April 2015.