[ImageJ-devel] BioImaging ROI Spec

[Lee Kamentsky]

I'm pretty sure I'll go to the meeting - both sets of dates are OK for 
me. I'll check in again when the plans become final and make the 
necessary arrangements. I think it would be useful if we could have some 
sort of scratchpad or wiki to discuss the spec prior to going. I have 
one or two use-cases that are challenging and also I want to plan how to 
address scalability issues and possibly benchmark a couple of approaches 
on real data.

--Lee

On 10/8/2012 11:15 AM, Jason Swedlow wrote:
>
> Dear All-
>
> I'm writing following various discussions  regarding the definition 
> and use of a common specification for multi-dimensional regions for 
> biological imaging, and in particular biological microscopy.  I am 
> cc'ing Michael Unser, Wiro Niessen and Antje Keppler on this email for 
> reasons that I will describe below.  I've also included representation 
> from OME, ImageJDev, KNIME, Icy, and CellProfiler.
>
> The first concrete discussion on a common ROI spec was held in April 
> in Barcelona.  Josh Moore's notes were circulated earlier (also, see 
> below).
>
> We now have an opportunity to finish this specification.  This follows 
> from meetings sponsored by Euro-BioImaging (http://eurobioimaging.eu), 
> and in particular The Data Management Work Package (often referred to 
> as "WP11"), led by Wiro and Michael, and which I, through our work on 
> OME, contribute to.  Through WP11, we have funds to sponsor a meeting 
> that moves a common ROI spec towards completion.  The idea is to have 
> a common specification, and example "proof-of-concept" software 
> implementation(s) in at least a few of our projects, so that a region 
> defined in one image software package can be used by another.  The 
> emphasis is on transport and interop, and not, for example, on a 
> single library for defining, using ROIs that must be used by all.  We 
> already know the kind and level of standardization that works, and 
> that which does not.
>
> Given the progress we made in Barcelona, it seems that this is not a 
> fantasy, but an achievable goal.  We don't propose a standard meeting 
> with presentations, etc., but more of a developer meeting or 
> "hackathon", that has been so successfully used by the ImageJ/Fiji 
> teams and others.  We need to work through the specification (ideally 
> making some progress beforehand using the various electronic tools we 
> all know and love), and then get at least rough implementations in place.
>
> As a start, we've used the scijava.org site that was defined at the 
> Dresden ImageJ/Fiji hackathon in Dec 2011.  See:
>
> http://www.scijava.org/
>
> As this was where our projects agreed to work together at the Dresden 
> hackathon, it seemed a place where we could do this work, but this is 
> not at all mandatory— just a place to start.  Whatever we use, it has 
> to be usable for modern code development (in my opinion, that means, 
> it has to be github or better, but other opinions welcome).  Obviously 
> other resources can link to this.
>
> Below “Resources”, there is now a link to:
>
> http://www.scijava.org/roi-model/
>
> On this page are Josh Moore's notes from the Barcelona mtg.  With Lee 
> Kamentsky's permission we can add his SpaceROI.xsd (email from 30 
> April), or a more recent update to it.  Roger Leigh (OME) has added 
> his work on a draft model spec.  He can explain what is going on there.
>
> For the mtg itself, we propose two possible dates— Nov 12-14 or Dec 
> 10-12.  There are obviously several looming conflicts— US 
> Thanksgiving, ASCB in SF, and several others.  I believe there is also 
> an ImageJ hackathon looming.  So far there is strong preference for 
> the November date.  Obviously minimising trans-ocean flights helps 
> productivity and keeps us all in good graces with our families.  Wilma 
> Woudenberg will send out a poll and we can try to get this nailed 
> down.  If those dates don't work, we can go sooner or later, so speak 
> up, but would be great to get this done before end of 2012.
>
> For location, we've discussed Dresden, Dundee, and there may be 
> others.  Through Euro-BioImaging, Dundee has funds to sponsor the 
> meeting, and to support travel for participants (economy only).  If 
> possible, Dundee will do all travel arrangements, to save the fun of 
> travel reimbursements from EU grants.
>
> I wasn't exactly sure who was interested, and we don't yet have a 
> single mailing list that covers all of us, although the SciJava google 
> group (https://groups.google.com/forum/?fromgroups#!forum/scijava 
> <https://groups.google.com/forum/?fromgroups#%21forum/scijava>) has 
> been used for some  discussions.    This problem will get worse when 
> we need our various developers to talk to each other.  I know many of 
> the OME devs read the ImageJdev list, but that is not universal.  Note 
> that some of the email lists included require registration, so do we 
> use scijava at googlegroups?
>
> Ok, this a too long email.  Decisions needed:
>
> -- do we agree to use scijava to start this discussion?
>
> -- can we put materials, discussions we each have there?
>
> -- date for meeting—is there another scheduled event we should 
> piggyback on? (Note I think things like the ImageJ mtg are not a good 
> idea; too many other things going on)
>
> -- everyone OK with Dundee for mtg place??
>
> Note: I'm partial to Dundee, since we have to pay for everything, so 
> it saves admin hassle, but if this can be solved easily, then that's 
> fine.  Wilma will take care of Doodle polls and very likely travel.
>
> We've worked together well in the past, so hopefully we can get this 
> done, and build something that will truly help our scientist users.
>
> Cheers,
>
> Jason
>
> P.S. As Euro-BioImaging will sponsor meeting, the various leaders of 
> WP11 and also Antje Keppler, Euro-BioImaging PM are cc'd.
>
> **************************
>
> Wellcome Trust Centre for Gene Regulation & Expression
>
> College of Life Sciences
>
> MSI/WTB/JBC Complex
>
> University of Dundee
>
> Dow Street
>
> Dundee  DD1 5EH
>
> United Kingdom
>
> phone (01382) 385819
>
> Intl phone:  44 1382 385819
>
> FAX   (01382) 388072
>
> email: j.swedlow at dundee.ac.uk <mailto:j.swedlow at dundee.ac.uk>
>
> Lab Page: http://www.lifesci.dundee.ac.uk/gre/staff/jason-swedlow
>
> Open Microscopy Environment: http://openmicroscopy.org
>
> **************************
>
>
> The University of Dundee is a registered Scottish Charity, No: SC015096 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://imagej.net/pipermail/imagej-devel/attachments/20121011/2f191caa/attachment-0001.html>