STIA’10: Spatio-Temporal Image Analysis for Longitudinal and Time-Series Image Data
Guido Gerig, University of Utah (email@example.com)
Thomas Fletcher, University of Utah (firstname.lastname@example.org)
Xavier Pennec, INRIA Sophia Antipolis (Xavier.Pennec@sophia.inria.fr )
June 13th, 2010 Full paper submissions submissions (Submission).
July 18th, 2010 Notification of acceptance (please note change of date)
August 18th, 2010 Camera ready papers due
Sept. 24 Workshop MICCAI 2010 F-W4
BEST PAPER AWARDS (4 winners were selected by the audience among the 16 presentations):
<![if !supportLists]>· <![endif]>DTI Longitudinal Atlas Construction as an Average of Growth Models, Gabriel Hart (UNC), Yundi Shi (UNC) , Hongtu Zhu (UNC) , Mar Sanchez (Emory University) , Martin Styner (UNC), Marc Niethammer (UNC Chap. H)
<![if !supportLists]>· <![endif]>4D registration of serial brain's MR images: a robust measure of changes applied to Alzheimer's disease, Marco Lorenzi (INRIA Sophia Antipolis / Asclepios team) , Nicholas Ayache (INRIA Sophia Antipolis / Asclepios team) , Giovanni Frisoni (IRCCS Fatebenefratelli), Xavier Pennec (INRIA, Asclepios )
<![if !supportLists]>· <![endif]>Group-wise Spatio-Temporal Registration and Segmentation of Fetal Cortical Surface Development, Georg Langs (MIT), Gregor Kasprian (Medical University of Vienna), Eva Dittrich (Medical University of Vienna) , Mario Bittner (Medical University of Vienna) , Peter Brugger (Medical University of Vienna), Daniela Prayer (Medical University of Vienna)
<![if !supportLists]>· <![endif]>A spatio-temporal model for joint segmentation and registration of cardiac cine MR images, An Elen (K.U.Leuven), Dirk Loeckx (K.U.Leuven) , Jan Bogaert (K.U.Leuven) , Frederik Maes (K.U.Leuven), Paul Suetens (K.U.Leuven)
Clinical imaging increasingly makes use of longitudinal image studies to examine subject-specific changes due to pathology, intervention, therapy, neurodevelopment, or neurodegeneration. Moreover, dynamic organ changes as seen in cardiac imaging or functional changes as measured in perfusion imaging, just to name a few, by definition result in time-series image data presenting volumetric image data over time. Expressions such as development, degeneration, disease progress, recovery, monitoring or heart cycle inherently carry the aspect of a dynamic process – suggesting that imaging at multiple time points may be applied. The detection and characterization of changes from baseline due to disease, trauma, or treatment require novel image processing and visualization tools for qualitative and quantitative assessment of change trajectories. Whereas longitudinal analysis of scalar data is well known in the statistics community, its extension to high-dimensional image data, shapes or functional changes poses significant challenges. Cross-sectional analysis of longitudinal data does not provide a model of growth or change that considers the inherent correlation of repeated images of individuals, nor does it tell us how an individual patient changes relative to a change over time of a comparable healthy or disease-specific population, an aspect which is highly relevant to decision making and therapy planning. Although successful early results were presented for image regression in aging studies, such standard regression is not optimal for longitudinal data because repeated observations from the same individual would ignore the correlation between repeated measurements and thus violate the Gauss-Markov assumption of independence. Moreover, individual change trajectories often need to be interpreted in relationship to a population growth model, which in turn is the hidden group model given a representative set of individual trajectories, and require a common framework based on the use of hierarchical linear models (HLM). Other typical driving applications are concerned with registration of serial data of the cardiac cycle, sampled at different time points, or measuring object shape changes via shape regression, both requiring new image registration and modeling approaches.
The special nature of longitudinal or repeated, time-series data of individual subjects, with the inherent correlation of structure and function across the sequence of images, resulted in the development of a variety of new image processing and analysis approaches tackling the challenging issues of registration, segmentation and analysis in the presence of geometric and contrast changes over time. New methodologies are rapidly evolving, often focusing on the specific application at hand.
This workshop will cover a comprehensive discussion of multiple approaches and new advances for spatio-temporal image processing of longitudinal image data but also aims at a dialogue to define the generic nature of algorithms, methods, modeling approaches, and statistical analysis for optimal analysis of such data. This in turn will lead to a discussion on solutions achieved so far, summary of the state-of-the-art and open issues to be tackled in future research.
Goals of the workshop are the following:
The primary focus of this workshop will be on the generic nature of algorithms, methods, modeling approaches and statistical analysis. We are specifically looking for papers that introduce novel methodology for spatiotemporal analysis of images. We will invite paper submissions related to the following topics:
Provisional list of topics:
We plan to start with a series of invited talks presenting major representative areas covering clinical motivation and methodological aspects of longitudinal imaging and image analysis. We will then continue with oral presentations of high-quality research selected from submitted papers (reviewed and selected by a paper review committee). Other paper submissions will be selected for short poster presentations. The presentations will be moderated by a session chair and will be followed by a joint discussion on methodologies developed so far and challenging open issues to be tackled in the future. We also plan to compile a list of freely available software and to collect a list of publicly available image databases of longitudinal image data to serve as testbeds for the scientific community. We will further distribute a list of books, book chapters and scholarly articles relevant to the specific topic. Presentations and associated papers will be available electronically to registered workshop attendees. Would the quality of submitted papers be highly innovative and original, we might plan to invited the best papers as a special issue in a scientific journal.
To our knowledge, this will be the first MICCAI workshop on the topic spatio-temporal image analysis, but related to a seminal tutorial on “Detection and Quantification of Evolving Processes in Medical Images” organized by Nicholas Ayache at MICCAI 2004. The topic is initiated by the need for information exchange and brainstorming w.r.t. this new, rapidly evolving image analysis sub-discipline.
Participants will get pdf copies of papers and eventually presentations (ready on CD at workshop).
There is a rapidly growing interest in the analysis of time-series data. Recent examples are the very successful MICCAI 2008 and 2009 workshops on image analysis of the early developing brain, where a modeling of change trajectories of growth and brain maturation were key topics. The increased focus on personalized medicine or subject-specific analysis includes processing of time-series data such as pre-/post-therapy or modeling of lesion evolution via parameterized models. Clinical studies of aging or , e.g., include series of follow-up scans to stage and model the effect of aging and to determine the onset of accelerated degeneration.
Rapidly evolving advanced imaging technology can routinely measure volumetric data in short time intervals, creating 4D datasets that require new, efficient processing, visualization, and quantitative analysis techniques.
The target audience will therefore be researchers interesting in or already involved in research and development of methods for studying growth or change patterns in longitudinal and time-series image data. This workshop aims at contributing to a fundamental understanding of data, processing methodology and statistical concepts but also to a review of existing methods, procedures and problem solutions. We expect to create discussions between potential users and researchers in order to inform about existing technology and to lay the ground for future research.