Commons:SMART-Servier Medical Art

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search

Smart Servier, the website that contains clip arts related to medicine and healthcare, contains over 3,000 images shared under CC-BY 3.0 license. These images are compatible to Wikimedia Commons.


The overall workflow has several steps

  1. Downloading and converting media from and Flickr, setting up file structure (User:OlafJanssen)- download from Flick is done
  2. Translation and describption of files and filenames (Wikimedians)
  3. Upload properly described and named files to Commons (via GLAMwikiToolset, OlafJanssen) - add a draft 'ideal example file' to Commons and further model (how an ideal filepage should look like)
  4. Further, more detailed categorization on Commons (see for inspiration?)
  5. Include images in target Wikipedia articles (and other Wikimedia projects)
How to find target Wikipedia articles?
Match images to Qids (with OpenRefine, work in progress for Bones-cat)
From the Qqids we can the WP-articles in all languages via urls like
For Simple Wikipedia :
Via the API, we can request the length of each article and how many images are included --> we can add the Smart images to articles that have a low/image word ratio (much text, few images is a target, and article with 0 image is a definite target.
  1. Add structured data/Wikidata to images (Link to ISA campaign) --> it would also be handy to already have some Qids for the images pre-upload, as for uploading the images with the GWT-upoadtool. I plan to model every image in (a derivate of, maybe create a custom template) the Photograph Commons template, which has a field for the Qid
  2. Measure impact: uptake and reuse
  3. Communicate about project and impact (for all steps)

Approaches of getting the images[edit]

  1. Hi-res images from Powerpoint slides (with multiple items per image file)
  2. Hi-res images with transparent backgrounds from SMART Servier on Flickr --> see for correspondence wit powerpoint structure
  3. Low-res images from webscraping (with single item per image file)
  4. Hi-res images of single items from manually cutting up image file into single item hi-res files (after 1 is done) --> this is not needed anymore, because of 2)

Part of work and file will be available on

SSA Homepage to refer back to from Commons image description (for instance from pdf version of each powerpoint slide deck:

Description, translation and categorization[edit]

Creating and translation of descriptions for the images and categorization shall be done by a group of volunteers. Filenames on Flickr are in French, this needs translation into English for better/wider acessibilty worldwide.Structured data shall be added to images by using ISA tool.


If you are interested in helping out with this project, please sign up below:

Images & templates[edit]

(assuming the Wikidata items ankle fracture (Q2314265), foot (Q15807) and ankle (Q168002) are the P180-Depicts values for the File in this example)

TODO list 30-6-2021[edit]

Working on implementing the comments/suggestions made by Jarekt: The are several actions I'm working on:

  • 3) Writing a script that can add Wikidata Q-numbers from an Excel to the P180-Depicst tag in the Structured Data of a file. This way we can remove the Qnumber from the {{SMART-Servier Medical Art Image}} template and into the SDoC. I know this is possible, but still need to figure it out.
  • 4) A way to adapt the metadata/infoboxes of the already uploaded Bones-images to the new situation, preferably using VisualFileChange. This has to do with making regular expressions work in that tool, I've tried it, but without succes, so I'll need to get some help with that.

SDoC P180-Depicts - progress[edit]

Using Hay's Structured Search


Cellular biology and histology[edit]

Uptake and reuse[edit]

Communication about this project[edit]

Articles found onthe web[edit]

Subpages in this project[edit]

SMART-Servier Medical Art[edit]

Commons:SMART-Servier Medical Art[edit]

Template:SMART-Servier Medical Art[edit]