Miccai 2008 Segmentation Workshop

Workshop title

3D Segmentation in the Clinic: A Grand Challenge II Competition in Coronary Artery Central Lumen Line Extraction, MS Lesion Segmentation and Liver Tumor Segmentation

Place, Time, WWW

• MICCAI 2008, New York University, New York, NY
• Room # to be announced
• Whole day workshop, 8am-5pm, September 6, 2008

• Detailed information on the workshop webpage and the MS lesion challenge

Keywords

Segmentation, Comparison, Evaluation, Multiple Sclerosis, Liver Tumor, Cardiac Imaging

Sponsor

TBA

General Organizers

Wiro Niessen, Theo van Walsum, Michiel Schaap, Coert Metz Biomedical Imaging Group Rotterdam, Departments of Radiology and Medical Informatics Erasmus MC Room EE 2151, Erasmus MC, PO BOX 2040, 3000 CA Rotterdam, The Netherlands Phone: +31 10 704 4119 Email: {w.niessen, t.vanwalsum, michiel.schaap, c.metz}@erasmusmc.nl Web: http://www.bigr.nl

Martin Styner Psychiatry and Computer Science, University of North Carolina at Chapel Hill CB 7160, Sitterson Hall, Chapel Hill, NC 27599-7160, USA Phone: +1 919 962 1909 Email: martin_styner@ieee.org Web: http://www.cs.unc.edu/~styner/

Simon K. Warfield Computational Radiology Laboratory, Department of Radiology Children’s Hospital and Harvard Medical School 300 Longwood Avenue Boston MA 02115 Email: simon.warfield@childrens.harvard.edu Web: http://www.crl.med.harvard.edu

Xiang Deng, Guangwei Du Corporate Technology, Siemens Ltd. China 7 Wangjing Zhonghuan Nanlu, Beijing, 100102, P. R. China Phone: +86 10 6476 6777 Email: xiang.deng@siemens.com

Objectives

This workshop is a continuation of the successful MICCAI 2007 workshop

The goal of this workshop is to quantitatively evaluate the performance of 3D image segmentation and tracking algorithms for three clinical applications, namely coronary artery tracking, multiple sclerosis lesion segmentation, and liver tumor segmentation.

The competition will be conducted using a set of clinical image databases that cover a wide range of patients and pathologies. Three sets of images will be provided to all participants. The first set, which includes ground truth manual annotations will be provided in February 2008 and can be used by contestants for fine-tuning the algorithms. The second set will only include the raw imaging data and will be used to evaluate the methods before the workshop and provide results for the workshop papers. The third set will be used at the workshop to evaluate all submitted algorithms on-site. The evaluation results of each separate competition will be summarized in a top rank journal. Submissions of the same algorithm already submitted to the main conference (or other workshops) are allowed, as well as submission of already published methodologies - the focus of this workshop is segmentation and tracking accuracy and stability under real-world conditions.

The second major goal of the workshop is to foster indepth discussions and technical interchange among researchers. For this reason, the workshop consists of a combination of in-depth presentations of selected algorithms, poster, and discussion sessions.

Program

TBA