MICCAI 2006 Review process

A careful review and selection process was executed in order to secure the best possible program for the MICCAI 2006 conference. We received 578 scientific papers from which 39 papers were selected for the oral program and 192 papers for the poster program.

3 anonymous independent scientific reviewers were assigned to evaluate each paper. Reviewer affiliations were carefully checked against author affiliations to avoid conflicts of interest. The review process was run as a double blind process; authors could not now reviewer names and vice versa. A special procedure was devised for papers from the universities of the organizers, whereby an international program co-chair assigned reviewers. Communication with these reviewers was carried out through a set of anonymized e-mail accounts, thus upholding a double blind review process for all papers submitted to MICCAI. In the end we received 98 % of the reviews we asked for.

In general, the papers submitted to MICCAI 2006 were of a very high quality. 236 papers received an average overall score of 3.5 [scale is 1..5, 5 being best] or above, an additional 252 papers received an average overall score of more than 2.5 (and less that 3.5).

The MICCAI program committee consists of the local organisers, 2 internationally selected co-chairs, and 15 internationally selected area chairs. Each person a leading expert in his/her field. Each area chair was assigned 40 papers from which she/he formed a recommendation for the program committee based on the scientific reviews as well as their own assessment.

The entire program committee met in Copenhagen for a 2-day meeting on May 18-19th. All 578 papers and their corresponding reviews were printed and accessible during discussions. During this meeting, the area chairs in 5 groups of 3 - from their joint pool of 120 papers - identified 60 potential papers for the oral program, 140 papers for the poster program, and 80 potential papers for the poster program.

In a second round, 5 committee members selected the oral program from the 60 papers identified as potential oral papers. The criteria for selection of these papers were quality, total coverage of MICCAI topics, and suitability for oral presentation. Independently, the 80 potential poster papers were cut down to 31 accepted poster papers. This procedure involved sequential comparison of papers in samples of 5 by pairs of area chairs. This entire procedure was set up to select papers by paper to paper comparison and not by the scoring provided by the reviewers, which forced the committee members to argue for the decision in each individual case.

As can be inferred form the above, each paper was examined by 3 reviewers, and non-oral papers where further scrutinized by 3-5 program committee members, while the papers for the oral program where further scrutinized by 8 program committee members.

We believe that we have carried out a careful and fair selection process. Also, we believe that this process has resulted in close to the best possible program. Obviously, since the number of good papers submitted to MICCAI 2006 exceeds the number of slots in this single-track 3-day meeting, some good papers will be rejected. We hope that these will find their way into the MICCAI workshops or be published elsewhere.

Finally, we would like to thank the scientific reviewers and the program committee for their tremendous help in preparing the MICCAI 2006 program.

See you in Copenhagen

Mads Nielsen, Jon Sporring, and Rasmus Larsen