MICCAI 2014, SEP 14-18, Boston

Interactive Medical Image Computing (IMIC) Workshop

Part of MICCAI 2014: September 14, 2014 - Boston, USA

The Interactive Medical Image Computing (IMIC) Workshop was held in Boston, on Sunday September 14, in conjunction with MICCAI 2014. This page includes summary information and links to workshop content. We look forward to futher events on these topics.

The workshop format encourages hands-on demonstrations to stimulate interaction between developers and the user community of clinicians, neuroscientists, and biologists. Our focus is on original methods and elegant implementations that establish a human-machine dialog. Evaluation will be based on productivity, accuracy, simplicity, and overall user experience.

Attendees are encouraged to bring challenging real-world datasets and allow active experimentation with various demo tools on different data.

Workshop Location

Room 217 of the Martin Conference Center at Harvard Medical School. Room floorplan.

Best Paper Prizes

Out of a very strong field of contributions, the following efforts earned special recognition and shared the $300 prize money based on their scientific merit.

1. Best paper award based on the off-site reviews: "Tracking by Assignment Facilitates Data Curation" Florian Jug (MPI-CBG) Tobias Pietzsch (MPI-CBG) Dagmar Kainmüller (MPI-CBG) Gene Myers (MPI-CBG)

2. Best paper award for impact and usability: "tipiX: Rapid Visualization of Large Image Collections" Adrian Dalca (MIT) Ramesh Sridharan (MIT) Rost Natalia (MGH) Polina Golland (MIT)

3. Best paper award for innovative idea: "How to use level set methods to accurately find boundaries of cells in biomedical images? Evaluation of six methods paired with automated and crowdsourced initial contours" Danna Gurari (Boston University) Diane Theriault (Boston University) Mehrnoosh Sameki (Boston University) Margrit Betke (Boston University)

Workshop Schedule

Oral Session I : Microscopy imaging, Cell tracking, Segmentation

Chairs: Tammy Riklin Raviv and Yi Gao
9:00-9:15 Tracking by Assignment Facilitates Data Curation Jug, Florian
9:15-9:30 Informed Segmentation: A Framework for Using Context to Select an Algorithm and a Case Study Using Humans in the Loop. Gurari, Danna
9:30-9:45 How to use level set methods to accurately find boundaries of cells in biomedical images? Evaluation of six methods paired with automated and crowdsourced initial contours. Gurari, Danna
9:45-10:00 An interactive 3D segmentation for the Medical Imaging Interaction Toolkit (MITK)(available in MITK Workbench). Fetzer, Andreas

Oral Session II: Segmentation

Chairs: Steve Pieper and Bjoern Menze

11:00-11:15 Intelligent Initialization and Interactivity: Optimizing Level Sets for T1-weighted White Matter Segmentation. Xue, Wenzhe
11:15-11:30 Probabilistic model for interactive 3D segmentation. Hershkovich, Tsachi
11:30-11:45 A Concept for the Application of a Hierarchical Image Subdivision to the Segmentation Editing Problem. (movie, slides) Heckel, Frank
11:45-12:00 An Effective Interactive Medical Image Segmentation Method Using Fast GrowCut. Zhu, Liangjia

Oral Session III : Visualization/Registration/Denoising

Chairs: Allen Tannenbaum and Hans-Peter Meinzer
14:00-14:15 Interactive 2D/3D Image Denoising and Segmentation Tool for Medical Applications. Urschler, Martin
14:15-14:30 A Control Framework for Interactive Deformable Image Registration. Kolesov, Ivan
14:30-15:00 An interactive visualization tool for Nipype medical imaging pipelines. Sridharan, Ramesh
14:45-15:00 tipiX: Rapid Visualization of Large Image Collections. Dalca, Adrian (online demo)

Demo Session I: 10:00-11:00

Part I 10:00-10:30
1. An interactive visualization tool for Nipype medical imaging pipelines. Sridharan, Ramesh
2. Tracking by Assignment Facilitates Data Curation. Jug, Florian
3. How to use level set methods to accurately find boundaries of cells in biomedical images? Evaluation of six methods paired with automated and crowdsourced initial contours. Gurari, Danna
4. An interactive 3D segmentation for the Medical Imaging Interaction Toolkit (MITK). Fetzer, Andreas
5.Intelligent Initialization and Interactivity: Optimizing Level Sets for T1-weighted White Matter Segmentation. Xue, Wenzhe
Part II: 10:30-11:00
6. A Concept for the Application of a Hierarchical Image Subdivision to the Segmentation Editing Problem. Heckel, Frank
7. Medical image navigation by gesture recognition for Leap Motion Thong, William*
8. Exploration and validation of time-lapse microscopy images using CellProfiler Tracer Bray, Mark-Anthony
9. Image-based Interactive Assessment and Therapy Simulation of Aortic Coarctation Hennemuth, Anja

Demo Session II: 15:00-16:00

Part III: 15:00-15:30
1. tipiX: Rapid Visualization of Large Image Collections. Dalca, Adrian
2. A Control Framework for Interactive Deformable Image Registration. Kolesov, Ivan
3. Interactive 2D/3D Image Denoising and Segmentation Tool for Medical Applications. Urschler, Martin
4. Probabilistic model for interactive 3D segmentation. Hershkovich, Tsachi
5. An Effective Interactive Medical Image Segmentation Method Using Fast GrowCut. Zhu, Liangjia
Part IV: 15:30-16:00
6. Informed Segmentation: A Framework for Using Context to Select an Algorithm and a Case Study Using Humans in the Loop. Gurari, Danna
7. Cardiac Image Modeller Right and Left Ventricle. Gilbert, Kathleen
8. inTag, CMRSegToosl and C-DTI. A suite of tools for analysing cardiac MR images. Romero R., William A.
9. MIRAN - An Interactive Image Segmentation and Meshing Environment Borsic, Andrea
10. Software Demo: MITK Diffusion. Hering, Jan
~~~~~

Schedule

Important Dates

Online Demo Submissions Open Until August 15, 2014

Paper Submission Deadline: June 9 June 12, 2014

Acceptance Notification: July 5, 2014

Camera Ready Manuscript Required: July 11, 2014

Demo Submission Deadline: August 15, 2014

Workshop: Sunday, September 14, 2014

~~~~~

Topics

Visualization of large multiparametric, multimodal, and/or multisubject imaging data

Interactive control of image segmentation or registration algorithms

Software architectures to optimize the use of high-performance computing (e.g. clusters or GPUs)

Novel interaction or visualization devices such as native 3D displays, head-coupled tracking, hand tracking, gesture detection, custom user input devices and related technologies

Interactive simulations of body systems, disease processes, and interventions

Education and training systems that use interactive methods

Other research that involves interactive computing applied to medical image computing

~~~~~

IMIC

Interactive systems are widely used in research and clinical practice, yet most scientific venues make it impossible to fully communicate exactly how they work. IMIC combines live demos, hands-on examples, and scientific publications to support exchange of ideas among a community of like-minded researchers. IMIC seeks to showcase and disseminate the best examples of dynamic interaction with medical imagery.

Organizing Committee

  • Tammy Riklin Raviv, The Department of Electrical and Computer Engineering, Ben-Gurion University
  • Steve Pieper, Isomics, Inc.
  • Yi Gao, Department of Electrical & Computer Engineering and the Comprehensive Cancer Center, The University of Alabama at Birmingham
  • Bjoern Menze, TU Munchen

On-site referees for best paper prize

  • Koen Van Leemput
  • Andrey Fedorov
  • Steve Pieper

Program Committee

  • Koen Van Leemput
  • Kilian Pohl
  • Anastasia Yendiki
  • Demian Wasserman
  • Andriy Fedorov
  • Jim Miller
  • Marcel Prastawa
  • Ivan Kolesov
  • Peter Karasev
  • Isaiah Norton
  • Dagmar Kainmueller
  • Florian Yug
  • Moti Friman
  • Sylvain Bouix
  • Yaniv Gur
  • Ender Konukoglu
  • Peter Savadjiev
  • Lichao Wang
  • Tina Kapur
  • Andrey Fedorov

Paper Submission Process

We are using an automated submission management system.

Online Paper Submissions Closed

Submissions will be carefully peer-reviewed by a selected group of researchers with extensive experience in IMIC. Workshop presentations are expected to represent a cross section of the best available research in this area. As this is an emerging field that has not been historically well-covered in the MICCAI literature, accepted submissions will set a precedent that can be expected to be widely discussed.

Note that the online submission system allows upload of supplementary files, so you may include a video, links to project information or online content, and/or a description of your live demo.

Materials from the workshop will be collated and made available via the InteractiveMedical.org web site as a record of the event and a resource to the field.

Paper format
  • Papers should be formatted using the LNCS style files.
  • Papers should have 4-8 pages, but shorter papers accompanied by videos or live demonstrations will also be given full consideration.
  • Submissions with video and/or proposing live demonstration will be given higher priority for oral presentation.
Parallel Submissions

Submitted work has to be original, not identically submitted in parallel to other conferences or workshops. We accept papers that provide further detail on algorithms also validated in the MICCAI challenge workshops. We ask authors to reference the challenge contribution explicitly.

Review process

The review of the papers will be double-blind to the extent physically possible. Submissions should be anonymous according to the MICCAI guidelines.

Previous MICCAI submissions

If desired authors can upload the MICCAI reviews and rebuttal to be considered by the workshop chairs.

Patent disclosure policy

The purpose of the workshop is to exchange ideas and methods to advance the state of medical care. Attendees of the workshop and readers of the proceedings are expected to be able to implement and use the presented methods in their own work. Any restrictions on the usability of presented methods, whether for commercial or for non-commercial applications, must be disclosed.

This information should be included in a separate section at the end of text in the final ('camera-ready') version of the manuscript, and it will be printed in the proceedings in order to help your readers to make informed decisions.

Why IMIC?

The Medical Image Computing literature traditionally favors fully-automated analysis algorithms that offer the potential high throughput, objective, and reproducible results for large data collections. However fully-automated techniques are not able to handle many time-critical tasks, nor can they handle tasks that require contextual or general knowledge not readily available in the images alone. This workshop addresses the topic of human interaction with algorithms for initialization, steering, quality control, and visualization for problem domains such as segmentation and registration, object detection, tracking. The workshop format encourages hands-on demonstrations to stimulate interaction between developers and the user community of clinicians, neuroscientists, and biologists. Attendees are encouraged to bring challenging real-world datasets and allow active experimentation with various tools on different data. In the call for participation in the workshop we ask for extended abstracts or short papers that are accompanied by live demos. Our focus will be on original methods and elegant implementations that establish a human-machine dialog. Evaluation will be based on productivity, accuracy, simplicity, and overall user experience.