Mbirn: Training:Events Update
2005.06.21 (S.Pujol)
- fMRI data resampling in Java for 2D visualization on the Client side of the Statistical Analysis tutorial
- Version 1.1 of the Slicer tutorial for loading and viewing data.
- Documentation template for the software developed by Core 1&2 during the Programming week
2005.04.29 (R. Gollub/S.Pujol)
- Focus of the fMRI tutorial development on data file format
- Wish-list from the Brain Imaging Laboratory for the May 26-27 Workshop at Dartmouth-Hitchcock Medical Center
- File format management of Darmouth's datasets
- 3D reconstruction of the hippocampus
- Tractography within the memory system for further connectivity studies
- Co-registration of different imaging modalities
- Joint effort with the Dissemination team to support the generation of external documentation during the June Programming Week
2005.04.12 (R. Gollub/S. Pujol)
Update
- fMRI Tutorial development continues to progress, both the software code and the html documentation are being improved
- We are also working out the details for what information to collect from new users when they access the tutorial
- beta testing planned in three phases,
- among developers and fMRI statisticians (R. Gollub, D. Greve, R. Hoge, M. Vangel, W. Plesniack, N. White, C-F. Westin)
- among NAMIC personnel (May 26-27 Dissemination event, end of June Programmers week)
- among larger audience of fMRI statisticans (K. Worsely, T. Nichol, S. Smith, H. Stern, L. Friedman) July and August
- Still to be determined is the number and sites for server hosting. This will be addressed during the Programming Week.
2005.04.06 (S.Pujol)
Update of the current work
- Upgrade of the documentation of the fMRI Statistical Analysis tutorial to a learner-oriented guide,
- Centralization of the training materials disseminated on the wiki via cross-links to Core 5,
- Testing phase of the fMRI Client-Server application.
2005.03.17 (S.Pujol)
Update of the current work
- Update of the HST web-based fMRI Statistical Analysis tutorial with features from the interface of the new Client/Server application.
- Assist end-users in fMRI Analysis training by adding to the tutorial detailled answers to the questions of the HST web-based course.
- Planning and content of the Slicer Training workshop at MGH (June/July).
- Definition of evaluation tools for Training workshops.
The main goals are 1) the evaluation of the knowledge learned during the sessions, 2) the identification of further developments based on the feedback from the attendees. This can be done via a first web-based questionnaire using SurveyMonkey tool and workgroup/discussion sessions.
- Review of the litterature in education for possible publications of the results of Core 5 Training materials.
2005.03.16 (S.Pujol)
Outcome of the 2005.03.14 Training/Dissemination TCON
Team Discussions In attendance: Randy Gollub (MGH), Ron Kikinis (BWH), Tina Kapur (Epiphany Medical), Steve Pieper (BWH), Sonia Pujol (BWH).
Core 5 and Core 6 joint efforts on key delivery to Core 3: Training sessions on Slicer3D
Please, visit the NAMIC TCON information http://www.na-mic.org/Wiki/index.php/TCON:Main to get the text summary of the discussion.
2005.03.05 (R. Gollub)
Updates below are based on the outcome of the 2005.03.4 Morphometry BIRN meeting held in Miami.
Team Discussions
In attendance: Randy Gollub (MGH)/ Jorge Jovicich (MGH)/ Susumu Mori (Johns Hopkins)/ Steve Pieper (BWH) and Allen Song (Duke).
Susumu Mori of Johns Hopkins and Allen Song of Duke University are experts in DTI (and more), Allen is an MR physicist who have expertise in the acquisition and correction of DTI data, Susumu is focused on the analysis and visualization of the DTI data. Both of them have devoted considerable resources to the development of various training and educational software tools as they did their work and both would like to collaborate in one fashion or another.
Here is a complete list of ideas/information that were discussed:
1- They really liked the idea of the NAMIC Training Site becoming the Wiki Pedia for Medical Imaging Educational Resources and offered to help point us to materials they knew about.
2- We discussed having Sonia attend the ISMRM meeting to take the Educational Courses so she could bring those resources to bear- i.e. learn how to organize the material and get any additional web links, open source teaching materials to post. And get up to current speed with the state of the art in teaching this material. I have attended this in the past (1996 I believe) and it was great, but my materials will be very out of date.
3- We discussed having Sonia (and possibly me too) travel to NC after the fMRI tutorial is finished to visit not only Guido's, but also Allen's lab to show them the tutorial and to see the written material and software tools that they have developed. For MRI acquisition: Allen has a software tool that synthetically generates MRI data given a set of acquisition parameters (TE. TI, FOV, )- T1, T2, PD. Looks like a GE scanner. He imagines that it could be expanded to include DTI. He also has a simple visualization software tool that is written in ITK for 3D on the fly viewing of fMRI activation data that he thinks could readily be adapted for DTI results, perhaps even structural MRI. He also is going to explore the possibility that the publisher of his textbook might be willing to allow him to post some of the materials on our site, particularly the material on the included CD that is already in html. This includes a glossary and more.
4- Susumu has a BEAUTIFUL viewing software tool for visualizing DTI data, but he is not sure how he will share/distribute it. He already has a large database of DTI data on his web site that he is open to the scientific community for sharing along with other material there that we can investigate. We will stay in touch with him as we go.
5- Most importantly, they were both very enthusiastic about utilizing the M-BIRN calibration data set for educational purposes. They have no objection to working with us to make educational modules based on the lessons learned from this work. It is likely that this will be an on-going process that will take 1-3 years to fully implement.
6- They both agreed to be advisors to us in the overall planning for NAMIC MRI and DTI teaching. We could set a goal for Core 5 to have a draft "big picture" plan for what exactly we will teach and how to have a session at the October BIRN All Hands meeting in San Diego devoted to getting this group's input.
Steve P and Ron would like to add to our highest priority work list bringing Slicer training and dissemination for end users into final polished form. We, with Sonia's enthusiastic agreement, have approved her taking on a key role, together with Tina, in running the Slicer workshops being planned for the near future. We generated an initial action plan for this:
1- Steve P will be responsible (together with Alex Yarmakovic) for drafting a one page synopsis of Slicer 2.4 functionality
2- Steve P with Nicole, Marianna, and various SPL folk will gather together all currently available Slicer documentation, tutorials, presentations etc
3- Steve P or his designee will go through this and prepare a status report on Slicer training
4- Ron or his designee with organize a conference call to initiate "the NAMIC process" for training/dissemination development and support- the "how-to" for Cores 2, 5 and 6 to work together with Slicer as the initial software tool to get this treatment. The agenda for that meeting will be: -report on the status of Slicer (see 3 above) -plan for what deliverables we want (e.g. downloadable modular tutorials in pdf format, web based FAQ, moderated email users list, materials for Workshops, update log) and how they will be NAMIC branded -conceptual overview of division of responsibilities to implement the agreed upon plan.
2005.02.28 (S.Pujol)
Updates below are based on the outcome of the 2005.02.20-22 NAMIC AHM meeting.
Team Discussions In attendance: Guido Gerig (UNC)/ Greg Jones (Utah)/ Sonia Pujol (BWH) with input from Steve Pieper (BWH) and Carl-Fredrik Westin (BWH).
Milestones and web-access of Core 5 tools were appreciated by NIH project managers as an excellent source of output material from NAMIC to the scientific community.
Key points
I) Identified need for multi-level training in DTI, with materials targeted for several audiences. From the most basic level to teach vocabulary and concepts to clinicians and novice learners to the most advanced to teach the applications engineers and algorithm developers about the relations among the analysis tools. Design of tutorial materials classified by the time required to complete (e.g. DTI in 10 minutes- designed for busy clinicians).
II) One goal of training is to improve DTI data quality for anyone who accesses our materials. Calibration of acquisition and analysis methods.
Suggested topics/domains for interactive educational material in this domain include: 1. Acquisition parameters: influence of the spatial resolution/ Scan Type (e.g. EPI vs spiral)/ Slice thickness, b-value, SNR, and gradient encoding directions, (One idea is a tutorial based on visualization of the same brain structure with different acquisition parameters), 2. Sources of geometric distortion, ways to minimize during acquisition and methods for correction during imaging processing, 3. Signal to Noise: 1.5 T vs 3T. III) Another goal of training is the understanding of the attributes of tissue that are revealed by DTI for anyone who accesses our materials: Visualization.
Suggested topics/domains for interactive educational material in this domain include 4. Tensors, 5. Glyphs , 6. Tractography.
Suggestion was made that for the Advanced Training we may need to help develop materials for training in the use of the NAMIC Image Analysis research tools (in coordination with the Core 1 developers, the Core 2 engineers and the Core 6 disseminators- see below). Tools for possible educational effort include 5. Clustering (C-F Westin – BWH), 6. Statistical Analysis (Guido Gerig – UNC), 7. QBall Imaging (David Tuch – MGH + Slicer3D).
Existing interactive tools that are available for use and additional development by the Training Core include Fiber Viewer (Guido Gerig, UNC) and Slicer3D DTI Module (C-F Westin, BWH). Another interesting software discussed was Diffusion Visualizer (Image Computing & Analysis Lab, Tokyo) used by David Tuch at MGH, but this is not open source.
Additional Ideas and Suggestions Collect material, make it available on NAMIC website (DTI library) Design of NAMIC format presentation to “brand” all our materials
Conclusion/Vision for how the Training Core fits into NAMIC efforts
Core1/Core 3 Communication issues --> Core 5 = interface between neuroscientists and computer scientists,
Teach the underlying fundamental medical imaging concepts/content through the tools; do not teach the tools, Open-Source oriented meeting (ITK), Software sharing for our community --> Image Analysis tools and all our training materials will be put in ITK when possible but will always be open source: NA-MIC toolkit.
Action Plans
- Received positive endorsement to go forward with our overall plans as presented.
- Need to develop overall framework for teaching DTI. Acquisition/calibration issues may be able to be partnered with Morphometry BIRN (Susumu Mori, Allen Song, Anders Dale, etc), update to come soon from Randy Gollub.
- Need to develop an overall strategy for coordination with the Core 1 developers, the Core 2 engineers and the Core 6 Disseminators.
- Organize upcoming immediate training/tele-teaching activities including NAMIC at MICCAI 2005, and workshops in Toronto and NIH.
