Connectionists: Bio-Medical Image Captioning Dataset and Benchmark at ImageCLEF 2017

[Carsten Eickhoff]

Automatic image understanding is one of the key challenges on the way to
medical artificial intelligence. Clinical resources such as radiology or
biopsy images are traditionally hard to process using state-of-the-art
machine learning and computer vision methods.

To foster research in this direction, ImageCLEF 2017 is hosting a dedicated
task on caption prediction for bio-medical images and provides a dataset of
160,000 pairs of clinical images along with their manually assigned
captions.

For registration, dataset download and more information, please visit:
http://www.imageclef.org/2017/caption

Timeline:

   - 01 Feb. 2017: development data release
   - 15 Mar. 2017: test data release
   - 04 May  2017: deadline for submission of runs by the participants
   - 15 May  2017: release of processed results by the task organizers
   - 26 May  2017: deadline for submission of working notes papers by the
   participants
   - 17 Jun.  2017: notification of acceptance of the working notes papers
   - 01 Jul.   2017: camera ready working notes papers
   - 11.-14 Sep. 2017: CLEF 2017 Conference, Dublin, Ireland

Organization:

   - Carsten Eickhoff, ETH Zurich, Switzerland
   - Immanuel Schwall, ETH Zurich, Switzerland
   - Alba García Seco de Herrera, National Library of Medicine (NLM/NIH),
   USA
   - Henning Müller, University of Applied Sciences Western Switzerland,
   Sierre, Switzerland


-- 
Carsten Eickhoff
Postdoctoral Researcher

ETH Zurich - Data Analytics Lab
Universitätstrasse 6
CAB F61.2
8092 Zurich
Switzerland

T    +41 446 337 017
E    ecarsten at inf.ethz.ch
W    http://www.carsten-eickhoff.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/connectionists/attachments/20170209/68d4e5d1/attachment.html>