Words are important too...

July 27, 2014

They say a picture is worth a thousand words, but for the images on Slice, words are important too.

Slice hosts over 9000 images contributed by academics and healthcare professional from around the world. Thanks to these contributions, Slice has a huge variety of images, including macroscopic, microscopic and radiology images, as well as photographs of laboratory specimens and diagrams. Finding images within the image bank is possible because well curated descriptive data (metadata) is a priority and we have devoted significant resources to this.

When designing the Slice image bank, the focus was on making sure have the right metadata for different image types. As a large proportion of Slice images are related to humans (or animals), the database is structured around the individual from which a specimen or photograph was taken. Slice captures general information about the species and gender of an image’s subject, but can also include information about the subject’s clinical history.

To make images easy to find, they are tagged for the body structures and disorders visible in the image (where relevant). SNOMED-CT (Systematized Nomenclature of Medicine – Clinical Terms) is used to standardise the data added to Slice. SNOMED has codes for body structures, disorders, procedures and histology stains, so by using SNOMED, the data on Slice is consistent and errors are avoided during data entry, making images easy to find.

‘Description of image’ includes information about the procedure by which the image or specimen was obtained, the orientation, scale and a description for macroscopic images. Radiology images include information about the procedure, modality and a description while Microscopic images have information about the type of procedure, histological stains, type of imaging used, scanning magnification and a description. Where relevant, Slice can also capture information about the type of cell or cellular structures visible and antibodies used in staining.

The ‘Clinical History’ panel includes de-identified information about the subject’s prior medical history and where relevant any disorders related to the medical history have been tagged. The ‘Case Materials’ section records information for related images.

Acknowledging individuals who have made contributions to Slice ensures that individuals, institutions and professional bodies get credit for sharing their resources. This information is recorded in the ‘Provenance’ section and recommendations for referencing Slice images are available here.

Contributions to Slice only need to be accompanied by relevant information about the type of image, body structures and disorders. We thank you for considering contributing to Slice. To contribute contact the BEST Network at support@best.edu.au.