• Home
  • Program
  • Challenges
  • Submission
  • Important Dates
  • Keynotes
  • People
TO ALL PARTICIPANTS

We had a great workshop this year which was held at the Harvard Medical School in conjunction with the MICCAI 2014 conference. A great day with great presentations and two excellent invited talks by Prof David Haynor and Prof Nassir Navab. Thanks everyone for making this workshop a success and see you next year at CSI 2015!

Jack, Ben, Tobi, Shuo

PS: The slides of Prof David Haynor's presentation are available here.
Picture
Winners of the Workshop Awards

This year's awards were sponsored by Microsoft Research. We had two categories, 'Winner of Vertebra Segmentation Challenge' and
'Best Workshop Paper'. And the winners are...
Best Workshop Paper 2014
An Active Optical Flow Model for Dose Prediction in Spinal SBRT Plans. Jianfei Liu, Jackie Wu, FangFang Yin, John Kirkpatrick, Alvin Cabrera, Yaorong Ge. Duke Cancer Center, University of North Carolina at Charlotte, USA


Honorable Mention:

Localisation of Vertebrae on DXA Images using Constrained Local Models with Random Forest Regression Voting. Paul Bromiley, Judith Adams, Tim Cootes. University of Manchester, UK

Automated Radiological Measurement of Spinal MRI. Meelis Lootus, Timor Kadir, Andrew Zisserman. Oxford University, Mirada Medical, UK

Automated 3D Lumbar Intervertebral Disc Segmentation from MRI Images. Xiao Dong, Guoyan Zheng. Southeast University, China, University of Bern, Switzerland


Picture
Picture
Winner of Vertebra Segmentation Challenge 2014
Interpolation-Based Shape-Constrained Deformable Model Approach for Segmentation of Vertebrae from CT Spine Images. Robert Korez, Bulat Ibragimov, Bostjan Likar, Franjo Pernus, Tomaz Vrtovec. University of Ljubljana, Slovenia


Honorable mention:

Vertebrae Segmentation in 3D CT Images based on a Variational Framework. Kerstin Hammernik, Thomas Ebner, Darko Stern, Martin Urschler, Thomas Pock. Graz University of Technology, Austria
Picture
Picture
CSI 2014 WORKSHOP

The spine represents both a vital central axis for the musculoskeletal system and a flexible protective shell surrounding the most important neural pathway in the body, the spinal cord. Spine related diseases or conditions are common and cause a huge burden of morbidity and cost to society. Spine imaging is an essential tool for assessing spinal pathologies and there is need for advanced computerized methods that support the physician in diagnosis, therapy and intervention.

Venue

The 2nd Workshop on Computational Methods and Clinical Applications for Spine Imaging (CSI 2014) is held in conjunction with MICCAI 2014 in Boston. The workshop will be held on Sunday, September 14 at the Joseph B. Martin Conference Center at Harvard Medical School (HMS), 77 Avenue Louis Pasteur. The workshop will be in conference room 214.There will be buses between Kresge Auditorium at MIT and the Martin Conference Center at HMS in the morning (departing at 7:00am and at 7:30am) and in the evening (departing at 5:15pm). Alternatively, you can use public transportation or taxi.

CSI 2014 is an event supported by the SpineWeb Initiative.

There will be invited talks from two internationally recognized experts in spinal imaging. Prof. David R Haynor and Prof. Nassir Navab will share their expertise and knowledge in the field of clinical spinal applications.

Scope of the Workshop

We are inviting researchers to submit papers on computational methods in the area of spine imaging and image analysis. Topics of interest include, but are not limited to:
  • Clinical applications of spine imaging
  • Computer-aided diagnosis of spine conditions
  • Image-guided spine intervention and treatment
  • Segmentation, registration, detection, and localization of spinal anatomy
  • Statistical modelling of shape and appearance
  • Novel imaging and visualization techniques

Challenges

An important aspect of this year’s workshop is to include computational challenges. Researchers are encouraged to evaluate novel and existing methods on two important image analysis tasks. The first challenge concerns full vertebrae segmentation, the second challenge is on vertebrae localization and identification. For both challenges a set of annotated training data is made available (~20 for segmentation, ~200 for localization/identification) through the SpineWeb Initiative. This data is intended to be used for development, parameter learning, and self-evaluation. Evaluation datasets will be released for both challenges. Participants are asked to run their methods on this data and submit their results which will be evaluated and ranked by the organizers. Challenge results will be presented and discussed during the workshop.
LATEST NEWS

31 July 2014
Workshop Final Program
Picture
Picture
Powered by Create your own unique website with customizable templates.