From NCIGT Wiki
Jump to: navigation, search
Home < CC:Meetings

Back to U41:Computation

Back to Events

The Computational Core Meetings are Held every other Wednesday at 10am in Hollenberg.



January 17

Attendees: Matthew Toews, Sandy Wells

  • Discussed B-mode ultrasound alignment and visualization technique.


November 30

Attendees: Petter Risholm, Sandy Wells

  • Discussed MIA paper.

November 23

Attendees: Petter Risholm, James Balter, Sandy Wells

  • Discussed R01 Grant application

October 26

Attendees: Matthew Toews, Sandy Wells

  • Discussed 3D reconstruction of 2D b-mode ultrasound data.

October 20

Attendees: Matthew Toews, Sandy Wells, Mehdi Moradi, Petter Risholm, Golby Lab

  • Presentation in Golby lab (Toews): "MR and Ultrasound Image Fusion for Image Guided Neurosurgery"
  • Discussed alignment of AMIGO ultrasound image data, for identifying tumor margins pre/post operatively.

October 11

Attendees: Petter Risholm, James Balter, Sandy Wells

  • Worked on R01 Grant application.

September 9

Attendees: Matthew Toews, Sandy Wells

  • Discussed PAMI paper.

August 30

Attendees: Matthew Toews, Sandy Wells

  • Discussed interpolation strategy for deformable registration from invariant feature correspondences.

August 23

Attendees: Petter Risholm, James Balter, Sandy Wells

  • Worked on R01 Grant application.

August 9

Attendees: Petter Risholm, Sandy Wells

  • Discussed progression of MIA paper

July 28

Attendees: Petter Risholm, Firdaus Janoos, Sandy Wells

  • Discussed INLA method for posterior inference.

July 18

Attendees: Petter Risholm, Sandy Wells, James Balter

  • Discussed uncertainty in radiotherapy, and in particular grant application.

July 18

Attendees: Petter Risholm, Sandy Wells

  • Discussed temperature estimation.

June 22

Attendees: Petter Risholm, Sandy Wells

  • Discussed IPMI presentation.

June 17

Attendees: Matthew Toews, Sandy Wells, Golby Lab

  • Discussed a technique for ultrasound-ultrasound alignment.

June 9

Attendees: Matthew Toews, Sandy Wells, Golby Lab

  • Presented approach and requirements for ultrasound-MR registration in AMIGO

May 10

Attendees: Matthew Toews, Sandy Wells, Stephen Aylward, Danielle Pace

  • Discussed new ultrasound-mri data from duke university, and recommendations for future data collection.

May 6

Attendees: Matthew Toews, Sandy Wells

  • Discussed results for progress report, learning from US/MR image pairs for feature-based registration.

May 2

Attendees: Matthew Toews, Sandy Wells

  • Discussed a learning-based feature selection approach to MR-US image alignment: learning from aligned, pre-operative US/MR pairs.

April 27

Attendees: Petter Risholm, Sandy Wells

  • Discussed First Monday presentation.

April 21

Attendees: Matthew Toews, Sandy Wells

  • Discussed a feature-based approach to MR-US image alignment, via gradient orientation histograms.

April 11

Attendees: Petter Risholm, Sandy Wells

  • Continued discussion on temperature estimation.

April 8

Attendees: Petter Risholm, Sandy Wells, James Balter

  • Discussed grant submission.

Attendees: Matthew Toews, Firdaus Janoos, Dominik Meier

  • Discussed registration directions:
 * MR-to-ultrasound
 * Robust atlas-to-subject

April 6

Attendees: Petter Risholm, Sandy Wells

  • Discussed "temperature" estimation.

March 30

Attendees: Matthew Toews, Sandy Wells

  • Discussed research directions - robust feature-based registration strategy, high-dimensional image registration.

March 9

Attendees: Matthew Toews, Sandy Wells

  • Discussed/reviewed MICCAI submission.

March 7

Attendees: Petter Risholm, Sandy Wells

  • Discussed final revision of MICCAI paper.

March 2

Attendees: Petter Risholm, Sandy Wells

  • Reviewed MICCAI paper.

February 22

Attendees: Petter Risholm, Sandy Wells

  • Discussed MICCAI paper.

February 7

Attendees: Matthew Toews, Sandy Wells

  • Discussed robust atlas-subject alignment for neurosurgery.

February 3

Attendees: Petter Risholm, Sandy Wells

  • Discussed how to compute surrogate points for aligning images in radiation therapy.

January 21

Attendees: Matthew Toews, Sandy Wells

  • Discussed a total variation regularization method
 -> link to information geometry <a href=></a>.
 -> optimization methods: mean field annealing.
 -> representations: distribution field.

January 26

Attendees: Petter Risholm, Sandy Wells

  • Discussed how to estimate the registration uncertainty based on a stochastic gradient descent algorithm.

January 13

Attendees: Matthew Toews, Sandy Wells

  • Discussed a total variation regularization method based on the log odds image representation and KL divergence measure.

January 12

Attendees: Petter Risholm, Sandy Wells

  • Discussed online estimation of registration uncertainty.

January 4

Attendees: Tina Kapur, Sandy Wells

  • discussed new potential BWH radonc dbp -- gyn cancer


December 21

Attendees: Tina Kapur, Sandy Wells, James Balter, Robert Cormack

  • Reviewed radonc research projects for NCIGT at BWH and Michigan

December 16

Attendees: Matthew Toews, Sandy Wells

  • Discussed inter-subject registration results, IPMI publication.

December 14

Attendees: Tina Kapur, Sandy Wells, James Balter (by skype)

  • Continued discussion of the role of uncertainty estimation for head and neck cancer for the Michigan DBP. James shared data with Petter Risholm.

December 8

Attendees: Matthew Toews, Sandy Wells

  • Developed bounds on max achievable mutual information for the distribution field image representation.

December 6

Attendees: Petter Risholm, Sandy Wells, James Ross, George Washcko

  • Discussed results for IPMI paper.

November 24

Attendees: Matthew Toews, Sandy Wells

  • Discussed powerset expansion for subject-atlas mutual information registration, bounds on max achievable mutual information.

November 23

Attendees: Tina Kapur, Sandy Wells

  • Discussed the role of uncertainty estimation for head and neck cancer for the Michigan DBP.

November 11

Attendees: Matthew Toews, Sandy Wells

  • Reviewed results for subject-atlas and intersubject registration, minimizing distribution entropy via powerset expansion.

November 2

Attendees: Tina Kapur, Sandy Wells

  • Reviewed the arc therapy project at Michigan, and potential ct-xray registration solutions

October 27

Attendees: Matthew Toews, Sandy Wells

  • Reviewed results for subject-atlas and intersubject registration, comparing distribution fields vs. raw intensities.
  • Discussed methods for improving distribution field uncertainty: entropy calculation and set theory.

October 20

Attendees: Matthew Toews, Sandy Wells

  • Discussed subject-atlas registration via a probabilistic distribution field atlas.

October 19

Attendees: Tina Kapur, James Balter, Sandy Wells

  • Met with James Balter, Michigan DBP PI and discussed project

October 5

Attendees: Tina Kapur, Sandy Wells

  • Discussed needs of radiation oncology DBP

September 28

Attendees: Matthew Toews and Sandy Wells

  • Discussed the distribution field image representation for registration; the relationship between image intensity blurring and the uncertainty in the registration mapping.

September 13

Attendees: Petter Risholm, Tina Kapur, Sandy Wells Discussed radiotherapy application of Petter's uncertainty registration method.

September 10

Attendees: Matthew Toews and Sandy Wells

  • Discussed probabilistic formulation for image similarity, combining the distribution field representation and the inverse hypergeometric distribution.
  • Application to magnetic resonance -> ultrasound registration

September 9

Attendees: Petter Risholm and Sandy Wells Discussed MR/US registration of prostate.

September 7

Attendees: Tina Kapur and Sandy Wells

  • Discussed plans for the 2d/3d registration SA of the renewed grant, and times to meet James Balter (DBP lead).

September 2


Sandy Wells and Matthew Toews

  • Discussed the use of the inverse hypergeometric distribution for image registration.

August 24


Sandy Wells Petter Risholm Matt Toews

  • Discussed Matt's recent experiments on feature-based

registration using mutual information.

  • Briefly covered Petter's recent developments of an extension

of his registration methodology to include a nonlinear (rather than linear) elasticity model.

July 22


Sandy Wells and Petter Risholm

  • Discussed the research plan for the next year.

July 21


Sandy Wells and Matthew Toews.

  • Discussed results for intensity profile feature representation, implementation in 3D.

July 15


Sandy Wells, Matthew Toews and Erik Miller

  • Discussed representing image feature appearance via intensity profiles.
  • Discussed a histogram-based image representation and links to bilateral filtering.

July 1


Sandy Wells and Matthew Toews.

  • Discussed feature-based registration and the application of compressed sensing techniques.

June 24

Present: Sandy Wells and Petter Risholm

  • Discussed the layout of the poster for WBIR 2010.

June 23

Attendees: Sandy, Matt, Tina Kapur, Andriy Fedorov. Visitors: Dieter Hahn

  • Discussed potential collaborations with Dieter on registration methodology for MR and Ultrasound. In particular, Dieter demonstrated the work on registration of thyroid cancer images at University of Erlangen.

June 14


Sandy Wells and Matthew Toews.

  • Discussed mutual-information based image feature selection for image registration.

May 28


Sandy Wells and Matthew Toews.

  • Discussed method for fast atlas-subject alignment of anatomical brain scans.
  • Investigated options for accounting for rotation: via invariance or an explicit search over rotation space.


February 9


Sandy Wells and Nicu Archip.

  • Discussed the proposal to organize at workshop at MICCAI 2009 - an opportunity to disseminate our results
  • Discussed potential invited speakers and collaborators for the workshop
  • Discussed the topics and scope of the workshop, potential publications

January 26


Sandy Wells and Nicu Archip.

  • The synergy between ongoing efforts in CC and future plans in view of resubmission have been discussed.
  • Registration and fusion of multi-modality imaging and data - has been explored as a potential theme for the CC.
  • Development and validation of novel imaging biomarkers - an opportunity for development and validation of image registration technology.

January 19

No meeting, Martin Luther King Jr. Day.

January 12


Sandy Wells and Nicu Archip.


April 23


Sandy Wells, Ehud Schmidt, Kinh Tieu, Kim Lawson, Haiying Liu, Noby Hata

Ehud presented on Overview of Registration Issues involved in Interventional Cardiac Electrophysiology

Some major points:

Imaging now adds electrical activity to the functional/anatomical mix Interventional just focuses on large chambers of the heart. Lot of use of MR/CT in this field. Limitations of MRI is that many cardiac devices are not MRI-compatible. Medtronic and St. Jude are two companies leading in the effort to make this possible.

Heart info: Sinus node is computer of heart. Fibrillation is firing of the muscles. In heart attack patients you have rogue asynchronous firing building around the border of infarct that poses the most danger to patient. Geometry and thickness of scar are two factors affecting interventions including contrast agent uptake (longer time than non-scarred tissue).

Under fluoroscopy guidance, mapping part done with preaquired CT-registered to coordinate system in the OR. Color code represents size of voltages.

Acquire MR images day to 2 days before, takes an hour. Points of view: luminal, congenital. Understand pressure, flow rate, scar, anatomy T2 acute injury can capture edema.

EP mapping longer than hour. Regimens for Pre-op to op: multiple contrasts...and Intraop to intraop

Can map deformation field. Predict how things will behave. Derivative of displacement map. Strain and displacement are different. Strain=no deformation. Change in volume of voxel deformation. Not one to one for electrical activation. Elastic deformation-predicting desynchrony.

Advanced simulation technology hasn't been done but would be an interesting project.

Cardiologists are used to working in x-ray with lexicon of LAO, RAO, left anterior and oblique of xray tube, mostly built on orientation. Not used to Slicer dimensions.

Three scans a day at BWH. Preop mapping in OR. Another problem is acquiring with low resolution breath holds. Acquisition done with breath holds or with high resolution navigation/echo (respiratory gating)Need to navigate on pre-op data possibly acquired during different cardiac phases.

Preop-MRI to intraop xray is a must to account for respiration

Modality of choice is still x-ray-one can do a lot faster. Intraop done on x-ray . shift up and down with coordinate frame. Intraop to intraop-patient moves as heart rate changes.

Can we account for changes in heart rate? Real-time imaging is acoustically very loud.

Only use MRI when need it. Need to navigate on pre-op data possibly acquired during different cardiac phases.

Today-purely fluoroscopy. St Jude has EM tracking device that brings pre op CT do registration. Shows 3D map. Use fluoroscope all of the time and push contrast all of the time.

April 2

CC Meeting of 4/2/08

Present: Sandy Wells Kinh Tieu Wendy Plesniak Raul San Jose Kirby Vosburgh Haiying Liu Noby Hata Necu Archip Sonia Pujol

We reviewed the status of the third year milestones, and the mechanics of producing the progress report. Individual components of the progress report are due on April 23, for integration into a CC report that is in turn due internally by the end of April.

We planned to hold a meeting of the CC that will focus on the registration problems that are currently on the horizon in the NCIGT, and beyond. That meeting will be held on April 23.

At that meeting, brief presentations will be requested from the following people about the registration problems that they are engaged or interested in:

  • Kirby Vosburgh or Raul San Jose
  • Necu Archip
  • Noby Hata
  • Steve Haker
  • Ehud Schmidt
  • Sonia Pujol

February 6

CC Meeting of 2/6/08

In attendance:

Sandy Wells Kinh Tieu Sonia Pujol Haiying Liu Kirby Vosburgh Necu Archip Noby Hata

Discussed the June 12 EAB meeting and May 30 progress report, including plans to bring publications listing up to date at that time using SPL Publications database.

We discussed the status of the milestones that were proposed for the current year.

  • Spatial Neuroinformatics

Sandy reported on recent progress with Clare Poynton in predicting EPI distortions from subject's structural MRI scans, and preliminary work on methodology for ultrasound / MR registration.

  • HPC

Necu summarized the status of the database of MRT scans for registration evaluation, and said that is was well received by the community. He summarized a recent publication in MRI distortion correction as well as some work in CT/PET biopsy guidance.

  • Endoscopic Navigation

Kirby reported that Jeff Stohl has left for Siemens in Mtn View, and that Raul San Jose will be taking over some of that work.

Kirby also reported on progress in the pancreas staging by navigated laparoscopic ultrasound, with an IRB at MGH for open surgery, with humans in Nova Scotia on the horizon. He also described interest in GI and lung applications at BWH, and pointed out that Jeff had a partial solution for automatic segmentation of the small bowel. Kirby reported related progress in whole body CT (registration ?) by Kilian Pohl, and noted that an Australian student will visit for a year and likely work on "whole body registration".

Sandy proposed that the CC have some meetings focussed on registration in the March time frame.

  • Slicer Engineering

Noby described a recent workshop on slicer for IGT that attracted 20 NIH funded participants. Noby also pointed out some recent progress in slicer - tracker interoperability.

  • other topics

The group discussed the potential utility of obtaining an optical tracking device for experimental purposes.

Sonia describe recent work in evaluating accuracy of a vessel navigation system for publication, work with Kai Freirichs.


November 28


October 24

Ultrasound-CT/MR Registration brainstorming led by Sandy.

October 3

Continuation of discussion from previous meeting with Nicholas Ayache.

September 19

Agenda: To present an overview of the research in the CC to Nicholas Ayache using the U41 CC wiki pages: Slicer Engineering (Hata/Liu), HPC (Archip), Spatial NeuroInformatics (Tieu). Each SA overview will take approx 15 minutes, and we will use the rest of the time for discussion.

September 5

Agenda: Checking back in at the end of the summer. Plan regarding milestones for the fall

August 22

Agenda: Slicer 3 IGT for Neurosurgery discussion and planning.

August 8


(personnel on grant are: Wells, Archip, Goldberg-Zimring, Liu, Plesniak, Tieu, Stoll, Jakab, Anderson)

July 25


Attendees: Sandy, Kirby, Tina, Haiying, Kinh, Nicu

June 6

Continued from last meeting: Radiation Oncology Invited Speaker: Xiang Hong, Radonc, BWH

May 30

No Meeting.

May 23

Radiation Oncology Invited Speaker: Xiang Hong, Radonc, BWH

May 16

No Meeting.

May 9

  • Regular meeting cancelled.
  • Chris Johnson visiting Sandy for cross-P41 collaboration with Utah.

May 2

No Meeting.

April 25

Continued Registration Discussion.

Apr 18

No meeting.

Apr 11

No meeting.

Apr 4


  • registration methods for IGT. what is being worked on in the group, and where the gaps are.
  • GZ to introduce collab with imaging core

Attendees: Eigil, Daniel, Kinh, Gary (Imaging Core), Sandy, Tina, Haiying, Nicu


Registration was discussed in the context of the following four applications:

  • Cardiac (Eigil)
    • Data: 4D preop to 2D intraop
    • Objective function: MI
    • Registration Methods: SVD, ICP, KF
  • Neurosurgery in the MRT (Nicu, Daniel)
    • Data: 7 MRT cases of 1 preop DTI+2-3 intra-op MRI
    • Registration Methods (and Obj Function): Comparison of FEM (?), B-Spline (?), and Demons (MI)
  • Prostate in the closed bore (3T) with ultrasound guidance (Steve Haker)
    • Data: to be gathered
    • Registration method: FEM (matlab implementation)
  • Laproscopic Ultrasound (Raul)
    • Details to be checked with Raul

Mar 14

Location: Hollenberg Time: 10am.

Note: DGZ on vacation.

Agenda: continuation of endoscopic navigation discussion and formalization of specific aim.

Mar 07

Location: SPL.

Agenda: continue endo discussion w/ demos in SPL.

Feb 28

Location: 10am Hollenberg.


  • Continue endoscopic guidance discussion from last meeting.
  • All CC investigators to fill in updates for yr2 (April 30 2006-April 30 2007) here: CC:Yr2_Progress_Updates. These updates should be completed by Wednesday, Feb 28th.

Attendees: Nicu, Wendy, Jeff, Haiying, Simon, Kim, Sonia, Daniel, Kinh, Tina, Sandy


  • Wendy Plesniak, the GUI designer for the Slicer 3 group, is in the cc and available to the team for consultations on GUI design of IGT applications.
  • ERC at JHU has an interest in neuroendoscopy, and it would make sense to start a seed project with them (peter kazanzides and Russ Taylor) as part of the endoscopy specific aim, and then spin off an R01.
  • There are three current efforts in endoscopy that we need to consider in creating this SA:
    • jeff's work on gastric endoscopy (slicer 2.4?)
    • neuroendoscopy work between Simon and student (SIGN)
    • surgical simulation work of raimundo (what clinical application and software??)
  • Rob Howe's student Yuri Ishihara is also working in this area and we can connect with
  • We will meet next week at the SPL for demos of current endoscopy applications
  • reviewed updates from investigators.

Feb 14

Time: 10am Location: Hollenberg.


  • Jeff Stoll to discuss his research.
  • Scientific and collaborations updates from all investigators to Sandy. (This will continue to the next meeting).
  • Schedule meetings from now to end of July.

We have Hollenberg for the following Wednesdays at 10am:

2/28 - continue endoscope meeting

3/7 - registration for igt
















Attendees: SW, KT, JS, NA, HL, ES, TK, RSJE, KL


  • Jeff presented his research. Discussion will continue next week.
  • Group will provide updates for yr2 on the wiki for Sandy's presentation to Ferenc
  • Kim Lawson introduced dissemination/training/research goals to the group
  • "IGT week" idea was discussed, similar to the NAMIC project/programming week.


Nov 29, 2006

Location Medical Staff Reception Conference Room B

Agenda: Information discussion led by Daniel Goldberg-Zimring on his work in the u41.

Nov 22, 2006


Agenda: Nicu to continue talk from Nov 8th meeting.

Attendees: Sandy Wells, Nicu Archip, Arne Hans, Noby Hata, Kinh Tieu, Simon DiMaio, Raimundo Sierra, Haiying Liu, Tina Kapur

Highlights: Nicu provided a quick summary from last time, and continued the talk for which slides are linked in the Nov 8th meeting. In addition to what is in the slides, he is planning to work on the non-rigid registration of A-mode ultrasound that will be collected in the experiments that Noby will be performing after attaching transducers to the modified mayfield clamp used in neurosurgery. The rough timeline for the data collection completion is June, 2007.

Another relevant point regarding Alex Golby's request for Nicu's non-rigid registration software to be available to her clinically is that Daniel Goldberg is now working on that. There was some discussion around this point, and we might need to follow it up in another session when Daniel is present (he had a conflict today and did not attend the meeting).

Nov 15, 2006

Location: Hollenberg. (Please note, not in Ron's office)

Agenda: AMIGO Network Planning Discussion.

Expected Attendees: Ron, Bob, Ben, Mike, Sandy, Noby, Simon.

Nov 8, 2006


Neculai Archip will present his ongoing efforts in the area of image guided liver interventions (RFA and biopsy). He will also describe his current work on validation of DTI deformable registration, estimation of white matter fiber tracts displacement during brain tumor resection, together with a novel framework for brain tumor validation.

Attendees: Sandy, Tina, Ron, Nicu, Kinh, Arne, Noby, Daniel GZ, Simon DiMaio. (Haiying was at a prostate case where his new Slicer biopsy module was being used).


  • Nicu presented his work on deformable registration for needle tracking in RF ablation of liver tumors. This work is with Stu Silverman's group. The slides are available [/IGT/images/0/03/Archip-talk-11-08-2006.ppt?IGTWikiDB_session=502bf7acf8368d0be9e1db0c7f4916e5 here] (~15MB).
  • He also started the presentation of a PET/CT guided biopsy. The plan is to continue his presentation in the coming weeks.
  • Non-rigid registration was identified a commom theme among multiple projects in the U41, including Steve haker's work that is part of the u41 and the BRP. Tina to follow up with Steve to see if he can present at a cc meeting.
  • Nikos from W&M is preparing a NA-MIC collaboration grant of which Nicu is also a part.
  • It came up that the ITK b-spline registration "didn't work" for this liver registration problem, and Tina will follow up with Bill Lorensen in NA-MIC to see if there is an opportunity to either educate us on correct use of the module, or to fix it.

Oct 11, 2006


  • 5 minute informal no-slide summary of current NCIGT U41 research (each CC researcher)

Sep 27, 2006

Location: Medical Staff Recep. Room B, 75 Francis.

Attendees: (conflicts in schedule: Raimundo Sierra at MIT, Noby at Northeastern MEMS )



  • Research Members of the Computation Core (as of June 1, 2006): Wells (CC PI), Warfield, Hoge, Mitsouras, Zientara, Goldberg-Zimring, DiMaio, Liu, Tieu, Hans, Sierra, Kikinis (dir of research for the U41).
  • Infrastructure Members of the computation core: Genoa, Jakab, Zlatanov, Anderson.
  • Mailing list u41-cc created. Folks should subscribe here:
  • Dates for future meetings (CC pages will be updated to reflect these):
    • Oct 11
    • Nov 1
    • Nov 8
    • Nov 15
    • Nov 22
    • Nov 29
    • Dec 6
    • Dec 13
    • Dec 20

Items to be discussed in future meetings:

  • 3-4-5 minute informal no-slide summary of current NCIGT U41 research (each CC researcher)
  • Provide update on network/IT requirements discussion (Noby, Simond D)

Sep 13, 2006



  • System administration for the Cluster
  • System administration for the Fat Nodes

Attendees: Sandy, Sonia, Nicu, Simon D, Haiying, Noby, Tina


  • Discussed requirements for the network/IT infrastructure of the AMIGO suite. Noby will follow up with Simon and Nicu to refine requirements and make a proposal for the room. As part of the discussion we identified the following physical areas in the hospital where our research takes place (in addition to the dedicated research lab areas) and which should be covered under the solution:
    • 221 3T Scanner (EP Ablation)
    • MRT
    • L1 Interventional CT Scanner (Nicu)
    •  ?? where are the neuroendoscopies done??
    • Interventional Radiology (Sonia)
  • Question: Does U41 have a cluser in y2 budget. (Answer: probably not. We brought this up in a meeting with Ron right after this meeting and that was his informed guess. We can find out the final answer from Ferenc when he reviews the budget.)
  • We might want to consider AFS...
  • Going forward we would like to try out a weekly meeting at this time (Wed at 10am) where we talk about research (40 min) and infrastructure (20 min). We'll start next week with an overview of the vision/goals/SAs of the CC. Tina will provide some links on the agenda page for the meeting.

Aug 23, 2006


  • Neculai Archip
  • Noby Hata
  • Haiying Liu
  • Sonia Pujol
  • Eigil Samset
  • Sandy Wells


  • Sandy described his recent visit to Oxford, launching a collaboration

with Mark Jenkinson and Clare Poynton on the topic of using registration and physics technologies to address distortions in EPI imaging.

  • Noby and Eigil described progress on the "SIGN" project. A release is

envisioned for the October IGT meeting, as well as an "article" in the Insight Journal. Noby described potential support for futre SIGN development with a STTR grant involving Kitware, Ken Cleary, and Stephen Aylward.

  • A discussion ensued about the Aurora system being late in coming back

from NDI. Simon DiMaio is the local point of contact.

  • Noby pointed out that the group has purchased a new tracking system

from Acension.

Aug 9, 2006

Wells will describe the structure of the u41

Archip will describe the status of the HPC specific aim

July 5, 2006

Location: Hollenberg Conf. Room

  • Wells describe structure of Computation Core
  • DiMaio describe ERC and Prostate Core

present: DiMaio, Liu, Wells, Archip, Pujol, Hata

Simon DiMaio did a nice presentation on the ERC collaboration, the "MRT Core", the Tempany BRP, and some related projects.

We also discussed general background on ERCs and P41s

June 21, 2006

Present: Wells, Hata, Liu, Poynton, Pujol, Archip, DiMaio, Samset

Big Picture: NCRR, P41's and Structure of the U41 and Amigo

The NIH environment was described, along with a request for criteria to be used in evaluating P41s

Slicer / MRT Update: Noby updated usage of Slicer in MRT

  • 2.6 RC 4 avaiable and in regular use
  • warping code run on "new cluster" at 1249 Boylston St.
  • new clusters coming: SPL and MRT locations (20 core machines?)
  • described minimally invasive laser brain ablation with MRT thermal monitoring

Slicer 3.0 / SIGN update: Eigil described developments in 3.0 architecture and related developments of SIGN (Slicer Image Guided Navigator?)

  • Slicer big and general purpose, good for pre-op preparation
  • SIGN leaner, for interventional applications
    • share data model (MRML)

Training: discussed integrating interventional aspects into the more general Slicer curricula

Action Items: Construct renewal timeline

Future Agenda Items:

  • Sandy Wells describe structure of U41 Computation Core
  • Simon DiMaio describes ERC and the prostate core
  • Necu Archip describes intra-operative warping
  • Sonia Pujol describes catheter registration project

May 31, 2006

Present: Wells, Samset, Liu, Warfield, Pieper

Future IT issues:

The issue of future administration for IGT infrastructure was today's topic. Some consensus emerged:

  • IT needs will include
    • computer system administration
    • network administration
    • maintaining and installing applications

Previously identified issues include:

  • special administration will be needed for MRT workstation, and associated fat node
    • special login mechanism, or standard one with group restrictions
    • special FS mounting
  • special access to cluster will be needed
    • preemptive use


  • 1 FTE will be needed to administer computer systems in the intervetional setting.
    • s/he will need to be very familiar with the SPL computer and network environment
      • perhaps s/he should spend a preliminary period working under McKie to learn it

Action items:

  • form a committee to address this problem
  • involve Mike Halle

May 17, 2006

Present: Wells, Hata, Samset, Liu, Warfield

Slicer 2.6

  • Usage at MRT going well, Janice starting to use it in place of old slicer.
  • Noby is working on "innovative therapy" documentation (like IRB),it has been developed but not submitted
    • q: Can it be used also for SIGN ? a: not clear
    • this exists already for laser ablation in brain, and for prostate
  • Simon: 2.6 use going well, found bug in threads + DTI, receiving attention
    • MRT / DICOM is a positive development

Future IT issues

  • special administration will be needed for MRT workstation, and associated fat node
    • special login mechanism, or standard one with group restrictions
    • special FS mounting
  • special access to cluster will be needed
    • preemptive use

Data for Collaborators: Should we prepare example sets of IGT data for external collaborators to use? Could it be available in time for the upcoming IGT workshop?

Slicer 3: Eigel is coordinating SIGN development with Slicer 3 team. Specification of new MRML is a potential bottleneck: not ready for showtime yet

Open tracker component development has gone very well. Eigil and Jochen von Spiczak and DiMaio have defined a general device interface that can merges streams of different types of events. Aiming at a Virtual Reality article

Student leaving July 1

MRT: Eigil has been using GE product "raw data server" that provides access to k space data. One must know the MR protocol to make sense of the data. Says it works well.

Jan 18, 2006

Present: Eigel, Noby, Haiying, SW, Simon W.

review u41 evolution and landscape / amigo / MRT

slicer 3.0 evolve with navigation compatability / Eigel

US / MR or CT registration

US will be in amigo, but what?

clearly general utility in future IGT paradigm: preop CT/MR intraop guidance: US

there is some MR compatible US Dan Kacher, Annie Tang

pursue it here?

Eigel has contact student at Mirada

maybe Vossburgh would fund student or postdoc

how about GPU processing for registration? kitware is having a seminar on GPU computation

invite Niceau to meetings?

Warfield had 2.6 reg problem now referred to Pieper? special version of release candidate loads data properly progress is happening

2.6 / MRT target is "working with MRT" Feb 1 IRB and training 2 more mos. techs will continue to use old system, training available for techs on new system old system dies with old MRT, summer timeframe

Jan 4, 2006


  • Sandy Wells
  • Simon Warfield
  • Noby Hata
  • Haiying Liu

Discussion of general structure

  • U41 (grant)
  • Amigo
  • MRT

HPC project and Slicer

  • current project NSF funded with SDSC
  • most processing off line
  • mesh generator with Nikos Charisochoides at William and Mary
  • whole brain working
    • push other data through
  • loading results and siplay in MRT not so good
  • aiming towards more intra op demos

Action Items

  • Slicer Issues: Future Meeting
    • after "Golby" slicer meerting
    • display of float data
    • dicom loading
    • genesis loading
  • More interaction with Slicer team
  • Status of registration in Slicer
  • Schedule presentation for U41 leadership meeting