Statistical Atlases and Computational Models of the Heart:
Imaging and Modelling Challenges
~ 5th OCTOBER 2012 ~
Recently, there has been considerable progress in cardiac image analysis techniques, cardiac atlases and computational models, which can integrate data from large-scale databases of heart shape, function and physiology. Integrative models of cardiac function are important for understanding disease, evaluating treatment, and planning intervention. However, significant clinical translation of these tools is constrained by the lack of complete and rigorous technical and clinical validation, as well as benchmarking of the developed tools. For doing so, common and available ground-truth data capturing generic knowledge on the healthy and pathological heart is required. This knowledge can be acquired through the building of statistical models of the heart. Several efforts are now established to provide web-accessible structural and functional atlases of the normal and pathological heart for clinical, research and educational purposes. We believe all these approaches will only be effectively developed through collaboration across the full research scope of the imaging and modelling communities.
Topics of interest include:
Efficient and robust statistical representations of cardiac morphology and morphodynamics
Quantitative analysis of cardiac images through segmentation and motion/deformation estimation techniques
Atlas construction methods.
Sharing and reuse of computational cardiac anatomical, mechanical and electrophysiological models
Strategies for the personalization of cardiac computational models
Parameter sensitivity quantification and identification of relevant parameters in complex computational models
Integration of multimodal data in a common reference space
Clinical translation of imaging and modeling techniques
Statistical analysis of regional heart shape and wall motion characteristics across population groups.
Atlas-based physiological analysis of subject-specific characteristics.
This workshop will follow on from the successful STACOM’10 and STACOM'11 workshops, which attracted over 50 participants and were published in the Springer LNCS series. This year the workshop will be focused again on both cardiac image analysis and simulation tools in order to advance towards their application in clinical environments. Challenges using data from human, phantom and animal stufies will be organized on segmentation and motion tracking.
This forum will provide a forum for the discussion of the latest developments in the areas of heart mapping, including atlas construction, statistical modeling of cardiac function across patient groups, cardiac computational physiology, model personalization, ontological schemata for data and results, atlas based functional analysis, and integrated functional/structural analyses, as well as the clinical applicability of these methods. The workshop will be of interest to computer scientists working in imaging and computational modeling, but also to experts in cardiology, radiology, biology and physiology. Through this workshop we would also particularly like to engage a new generation of early career researchers in working at these interfaces.
This year, four challenges are organized within the workshop:
!! Please note that we accept both regular papers and challenge papers for the main workshop !!
Oscar Camara (Universitat Pompeu Fabra, ES)
Tommaso Mansi (Siemens Corporation, Corporate Research and Technology, US)
Mihaela Pop (University of Toronto, CA)
Kawal Rhode (King’s College London, UK)
Maxime Sermesant (INRIA, FR)
Alistair Young (University of Auckland, NZ)
Leon Axel (New York University, USA)
Nicholas Ayache (INRIA, FR)
Piet Claus (KUL Leuven, BE)
Mathieu De Craene (Philips, FR)
Hervé Delingette (INRIA, FR)
Alberto Figueroa (King's College London, UK)
Peter Hunter (University of Auckland, NZ)
Reza Razavi (King’s College London, UK)
Puneet Sharma (Siemens Corporation, Corporate Research and Technology, US)
Nic Smith (King’s College London, UK)
Nathan Wilson (Open Source Medical Software Corporation, US)
Graham Wright (University of Toronto, CA)
Academic Objectives and Relevance to MICCAI
There has been increasing interest in the MICCAI community on the clinical impact of the image analysis and computational models developed in our field, as well as in benchmarking and comparison of these techniques. The evaluation of these techniques has been particularly challenging in cardiology comparing to other communities (e.g. brain) due to the intrinsic dynamic nature of the heart, which makes the generation of ground-truth data difficult. In addition, researchers in the cardiac domain have lacked a forum and platform for developing a cardiac mapping community and compare their work with common data. However, segmentation, motion analysis and computational models of cardiac electrophysiology are now well advanced to be included in benchmarking studies. This workshop seeks to bring together leading researchers in the field of heart mapping and showcase the latest technology in this field by testing it with generated ground-truth data based on human, phantom and animal studies. This will foster collaborations and develop the field of cardiac mapping and modeling by giving researchers a sense of community. Such collaborative initiatives have proven particularly helpful in the brain mapping community, and we hope to mirror a similar success in the cardiac domain.
Contacts for organizers: