Workshop on Spectral and Shape Analysis in Medical Imaging (SeSAMI)


** LNCS Proceedings are out: LNCS Vol. 10126 **

Workshop is on Friday Oct. 21 9AM-4PM, in Room Athenaeum CC I (Lobby Level)

We gladly announce the workshop for Spectral & Shape Analysis in Medical Imaging (SeSAMI), which is held in conjunction with the conference on Medical Image Computing and Computer Assisted Interventions (MICCAI 2016) in Athens, Greece, October 21.

Today’s image data often represents continuous and time-­varying phenomena, usually with a geometric structure. Shape and geometry processing methods are receiving increased attention, for example, due to their higher sensitivity to local variations relative to traditional markers, such as the volume of a structure. Spectral & Shape Analysis provide a wealth of opportunities for studying complex data. This workshop targets theoretical contributions as well as exciting applications in medical imaging, including (but not limited to):
  • Spectral Shape Analysis
  • Spectral Image and Mesh Segmentation
  • Spectral Classification and Dimensionality Reduction
  • Shape Representation, Processing and Modeling
  • Shape Segmentation and Registration
  • Statistical Shape Analysis
  • Longitudinal Shape Analysis and Processing
  • Medical Applications focused on Spectral and/or Shape Analysis
This workshop will provide an invaluable opportunity for researchers to present recent work on Spectral & Shape Analysis, as well as the intersection of these two domains.  In Medical Image Computing or Computer Aided Intervention, the understanding of shapes and their geometrical representations enables the modeling of organs, of anatomical structures or functional entities.  This analysis is supported by spectral methods that help in understanding geometric or high-­dimensional data representation, for example, in population studies, disease data across subjects, time or modalities.  

Program: October 21

09:00 - 09:10

Introduction (Martin Reuter) PDF

09:10 - 09:30

A volumetric conformal mapping approach to spectral clustering of white matter fibers

Author: Vikash Gupta, Gautam Prasad, Paul Thompson

09:30 - 09:50

Deep spectral-based shape features for Alzheimer's Disease classification

Author: Mahsa Shakeri, Herve Lombaert, Shashank Tripathi, Samuel Kadoury

09:50 - 10:10

Functional maps for brain classification on spectral domain

Author: Simone Melzi, Alessandro Mella, Letizia Squarcina, Marcella Bellani, Cinzia Perlini, Mirella Ruggeri, Carlo Altamura, Paolo Brambilla, Umberto Castellani

10:10 - 10:30

Volume Representation of Parenchymatous Organs by Volumetric Self-organizing Deformable Model

Author: Shoko Miyauchi, Ken'ichi Morooka, Tokuo Tsuji, Yasushi Miyagi, Takaichi Fukuda, Ryo Kurazume

10:30 - 11:00

Coffee Break

11:00 - 12:00

Invited talk: Guido Gerig

Longitudinal Analysis of Shape and Appearance to Study Group and Subject-Specific Growth and Disease Processes

12:00 - 12:20

Reducing variability in anatomical definitions over time using longitudinal diffeomorphic mapping

Author: Daniel Tward, Chelsea Sicat, Timothy Brown, Arnold Bakker, Michael Miller

12:20 - 12:40

Spatio-temporal shape analysis of cross-sectional data for detection of early changes in neurodegenerative disease

Author: Claire Cury, Marco Lorenzi, David Cash, Jennifer M Nicholas, Alexandre Routier, Jonathan Rohrer, Sebastien Ourselin, Stanley Durlemann, Marc Modat

12:40 - 13:00

Longitudinal Scoliotic Trunk Analysis via Spectral Representation and Statistical Analysis

Author: Ola Ahmad, Herve Lombaert, Stefan Parent, Hubert Labelle, Jean Dansereau, Farida Cheriet

13:00 – 14:00  

Lunch

14:00 - 15:00

Invited talk: Tom Fletcher

Probabilistic Models on Manifolds for Regression and Dimensionality Reduction of Shape Data

15:00 - 15:20

Statistical Shape Model with Random Walks for Inner Ear Segmentation

Author: Esmeralda Ruiz Pujadas, Hans Martin Kjer, Gemma Piella, Miguel Angel González Ballester

15:20 - 15:40

Volumetric Image Pattern Recognition using Three-way Principal Component Analysis

Author: Atsushi Imiya

15:40 - 16:00

Shape Preservation based on Gaussian Radial Basis Function Interpolation on Human Corpus Callosum

Author: Umut Orcun Turgut, Didem Gökçay

16:00 – 16:30  

Coffee Break


Topics

Specific topics of interest include, but are not limited to:

  • Medical Shape Analysis
  • Shape Representation and Processing
  • Shape Modeling
  • Shape Segmentation
  • Shape Registration and (dense or coarse) Correspondence
  • Shape based Image Segmentation
  • Statistical Shape Analysis
  • Longitudinal Shape Analysis and Processing
  • Spectral Shape Analysis
  • Spectral Image and Mesh Segmentation
  • Spectral Classification
  • Spectral Dimensionality Reduction
  • Spectral Graph Methods
  • Medical applications of all of the above.
(Note: In this workshop, the word "spectrum" refers to eigendecompositions of bounded operators in functional analysis.  This includes generalizations of matrix eigenvalues and vectors to operators, such as the Laplacian, and excludes spectral imaging (spectroscopy) as a topic. The word "shape" describes geometric representations, e.g., triangle or tetrahedra meshes, point clouds, volumetric or implicit representations.)

Important Dates

June 15 (23:59PST)  Submission of Title/Abstract*
June 15 (23:59PST)  Submission of Final PDF
June 20  Acceptance Notification
October 14  Camera Ready Submission
October 21  Workshop Day (Athens)
* Creation of submission is mandatory by June 15th - Title/Abstract may be modified until June 24th


Paper Submission
We will accept the submission of short and full papers (single column, LNCS style, Latex package here). The review is single-blind, submissions should include authors and affiliations. Reviewers will be chosen among experts in the medical image computing community. The most original contributions will be selected for presentations. Depending on the number of submissions, we plan on publishing the workshop proceedings in Lecture Notes in Computer Science (LNCS) or the MIDAS journal.
We are glad to announce that your revised papers could be published in a Springer LNCS Series, provided that final versions are extended to a minimum of 12 pages. Camera-ready files (sources+pdf in one zip file) are due on October 14. Sources should be complete, self-contained, including signed LNCS copyright form, bibliography and LNCS style files.
Please note that at least one co-author must register to the workshop, and the early bird registration deadline is August 19.
[ LNCS Copyright Form to sign: /LNCS-Copyright_Form-MICCAI-SESAMI16.pdf ]


Presentation Format
All oral presentations will be 15 minutes long, followed by 5 minutes of discussion. In addition to the presentation of original work, keynote speakers, experts in the field, will present an overview of their work.

Advisory Panel
Stanley Durrleman, Inria
Guido Gerig, NYU
Ender Konukoglu, Harvard Medical School
Julien Lefevre, Université Aix-Marseille
Diana Mateus, Technische Universität München
Washington Mio, Florida State University
Marc Niethammer, UNC Chapel Hill
Stephen Pizer, UNC Chapel Hill
Robert Pless, Washington University
Kilian Pohl, SRI
Yonggang Shi, University Southern California
Martin Styner, UNC Chapel Hill
Ross Whitaker, University of Utah

Organizers

Martin Reuter <mreuter at nmr.mgh.harvard.edu>

Harvard Medical School, Athinoula Martinos Center for Biomedical Imaging, Charlestown, USA

Christian Wachinger <wachinger at csail.mit.edu>

Ludwig Maximilian University, Munich, Germany

Hervé Lombaert <herve.lombaert at inria.fr>

Inria Sophia-Antipolis, France

Floor Plan (Room is on Lobby Level, Athenaeum CC I)