Lagging behind IEEE CVPR,
the call for paper for MICCAI 2010
is
once again
lacking a requirement
for REPRODUCIBLE research.


One year more of inefficient knowledge transfer,
where readers (you) will have to spend months
(thousands of dollars) trying to rebuild from scratch
what authors report in their incomplete papers.

Another year of more than 800 papers published:
without data
without source code
without parameters
without Open Access
not even with hyperlinks...

Promoting Medical research by using outdated
publishing technology from the 18th century.

Papers whose only purpose
is to be cited by future papers.

The time for change is overdue:

"Article-Level Metrics and the Evolution of Scientific Impact"
http://www.plosbiology.org/article/info%...



Luis



---------- Forwarded message ----------
From: gangqin.zhang
Date: Tue, Nov 24, 2009 at 3:54 AM
Subject: First call for papers for MICCAI2010
To: miccai2010


Dear All，

The 13th International Conference on Medical Image Computing and
Computer Assisted Intervention (MICCAI2010) will be held from 20th
to 24th September 2010 in Beijing, China. MICCAI attracts annually
world leading scientists, engineers and clinicians from a wide range
of disciplines associated with medical imaging and computer assisted
surgery.

You are highly encouraged to submit papers for the conference and/or
a proposal for the workshop and tutorial session. If you want to know
more detailed information, please visit the MICCAI2010 website:
http://www.miccai2010.org


MICCAI 2010 – First Call for Papers
20th-24th September 2010, Beijing, China

MICCAI 2010, the 13th International Conference on Medical Image
Computing and Computer Assisted Intervention, will be held from
20th to 24th September 2010 in Beijing, China. The venue for
MICCAI 2010 is the China National Convention Center (CNCC).
Located in the Beijing Olympic Green, the CNCC is right next to
the Bird's Nest (China National Stadium), the Water Cube
(National Aquatics Center) and the National Indoor Stadium.

Topics to be addressed in MICCAI 2010 include, but are not limited to:

General Medical Image Computing
Computer Assisted Interventional Systems and Robotics
Visualization and Interaction
General Biological Image Computing
Brain and Neuroscience Image Computing
Computational Anatomy (statistics on anatomy)
Computational Physiology (virtual organs)
Innovative Clinical/Biological Applications and Surgical Procedures


Important Dates
 1 February 2010 Tutorial and workshop proposals
 1 March 2010 Acceptance of tutorials and workshops
 11 March 2010 Submission of full papers
 20-22 April 2010 Rebuttal of reviews
 18 May 2010 Notification of acceptance
 18 June 2010 Camera ready copy of papers due
 20-24 September 2010 Tutorials, Conference, Workshops

Submission of Papers
We invite electronic submissions for MICCAI 2010 (LNCS style, double
blind review)
of up to 8-page papers for oral or poster presentation. Papers will be reviewed
by members of the programme review committee and assessed for quality and
best means of presentation. Besides advances in methodology, we would also
like to encourage submission of papers that demonstrate clinical relevance,
clinical applications, and validation studies.

Proposals for Tutorials and Workshops
Tutorials will be held and will complement and enhance the scientific programme
of MICCAI 2010. The purpose of the tutorials is to provide educational material
for training new professionals in the field including students,
engineers, clinicians
and new researchers. The purpose of the workshops is to provide a comprehensive
forum on topics that will not be fully explored during the main conference.

Executive Committee
General Chair/Co-Chairs
Tianzi Jiang, Institute of Automation, Beijing, China (Chair)
Alan Colchester, University of Kent, UK
Jim Duncan, Yale University, USA

Programme Chair/Co-Chairs
Max Viergever, Utrecht University & UMC Utrecht, The Netherlands (Chair)
Nassir Navab, TU München, Germany
Josien Pluim, University Medical Center Utrecht, The Netherlands

Tutorial Chair/Co-Chairs
Dinggang Shen, University of North Carolina, USA (Chair)
Alejandro Frangi, Universitat Pompeu Fabra, Barcelona, Spain
Gabor Szekely, ETH Zurich, Switzerland

Workshop Chair/Co-Chairs
Bram van Ginneken, Radboud University Nijmegen, The Netherlands (Chair)
Yong Fan, Institute of Automation, Beijing, China
Polina Golland, Massachusetts Institute of Technology, USA
Tim Salcudean, University of British Columbia, Canada


Best regards,
Gangqin Zhang

--------------
Gangqin Zhang
General Secretary of MICCAI2010
Reasearch Center for Computational Medicine(RCCM),LIAMA
Institute of Automation,Chinese Academy of Sciences
http://www.miccai2010.org/
http://www.rccm.org.cn/
Tel:+86-10-62613669
Fax:+86-10-62647458
----------------------------------------------------------
2009-11-23

Hi Luis,

I agree that it would be better if MICCAI reviewing guidelines did more to encourage reproducible research, but I think you are being overly pessimistic. After all, some of the burden of responsibility lies with the authors; we can still submit papers accompanied by parameters, code, and data, even if the guidelines do not specifically require it. Similarly, MICCAI reviewers can still make comments ranging from e.g. a suggestion that a submission would be strengthened if code were made available, to e.g. a demand that tuned parameter values be reported fully for the requirement of transparency as well as reproducibility.

Also, to some extent MICCAI can be driven by developments in Open Access rather than the other way around. For example, several of the papers from MICCAI'09 that caught my eye had used data from either OASIS or ADNI. These data-sets encourage reproducible research indirectly, not by demanding it, but rather by being among the best data-sets that people could use to evaluate their algorithms -- i.e. more carrot than stick.

I do agree with you that CVPR is really leading the way here, but the situation is slightly complicated by the sensitivity of some patient data, in the sense that it is considerably easier to make available a bunch of photos of random objects used in an object recognition study, than a set of confidential images of e.g. intra-operative MR of patients undergoing a procedure. Having said that, for code availability, we might even expect MICCAI to lead CVPR, since there is (arguably) greater motivation to make medical advances widely available than e.g. advances in radar aircraft tracking systems...

All the best,
Ged

Ged,

I guess my pessimism with MICCAI is aggravated by the optimism that IEEE CVPR motivated last week.

I agree with you, there is certainly a fraction of the burden that falls on the author's shoulders, and another fraction that falls on the reviewer's own discretion. It is however frustrating to see that "novelty" (which is useless in the context of scientific research) is held at a higher value than "reproducibility" (that is essential to scientific research).

Note also, that even if you are willing an able to share your source code and data, the publication methodology used by MICCAI doesn't allow you to submit this material to them. You will have to store your data and source code in a public repository (somewhere else), hopefully one that allow you to create unique identifiers, and then you will include the links to your data and source code in the paper that you submit to MICCAI. In this way, your additional information will make it to the physical paper print of the proceedings. The essential outcome of the conference, is still an 18th-century-style book printed on physical paper, with none of the rich information that you can generate today with any medium-size home computer. No color, no hyperlinks, no video, no 3D data... For an age where your cell phone allows you to browse the web, it is sad that the publication of medical research is still subject to limitations that are three centuries old.

Yes, there is a lot that MICCAI can learn from Open Access journals, and it is nice to see some authors sharing their data. Maybe it is up to us in the community to create the carrots that will provide recognition to authors who do the right thing, but by the same token, it is absurd that as a collective we still choose to assign reputation to a venue that doesn't serve the needs of the field (except of course, for generating citation entries that we could use stuffing up resumes).

It is true that medical data has the additional complication of having to protect patient information, but that is not an excuse for disposing of the need for reproducibility. Instead, it is simply a condition that tell us to adhere to certain practices when preparing the data that will be disseminated. It is still the case that NIH has a policy of Data Sharing, that is required for Federally funded research.

Ignoring a reproducibility requirement impose an enormous economic weight on the community. We are still the same people who will have to "reverse engineer" every paper to recover the actual functional part of the work: source code and data. How much time and funds are wasted every year by forcing researchers and graduate students to read beyond the lines of what can be written in 10 pages of text, trying to rediscover how a particular method was implemented and used ? Such level of inefficiency is a direct and sad misuse of public (and private) funds.


Thanks for your comments.
It is good to see that other people in the community also care for these issues.


Best Regards

Luis