Special issue: Functional imaging and modelling of the heart

This special issue of Medical Image Analysis presents modelling studies of cardiac anatomy and physiology. The studies represent a selection from scientific research introduced at the Fourth International Conference on Functional Imaging and Modelling of the Heart (FIMH). The conference was held on the 7th to 9th of June 2007 at the University of Utah, Salt Lake City, Utah, USA. The FIMH proceedings consist of 48 peer reviewed full articles (Sachse and Seemann, 2007). The 29 oral and 28 poster presentations made the FIMH conference a major event for those interested in cardiac modelling, imaging, and image processing. More than 120 participants from 17 countries attended the FIMH conference. A novel feature was the industry session. Invited speakers from companies, which offer products for cardiac diagnosis and therapy, provided insights into their application and significance of imaging and modelling techniques. The studies in this special issue underline the role of computational modelling and simulation as a research methodology complementary to animal experimentation and clinical research. Furthermore, the studies address concerns regarding the role of computational heart models in clinical diagnosis and therapy. The study of Kroon et al. (2008) describes an approach that estimates fibre orientation in the left ventricle of the heart with numerical optimization based on mechanical modelling. Knowledge on fibre orientation in cardiac muscle is a basic prerequisite for modelling the electrical and mechanical behaviour, because tissue properties are strongly dependent on orientation. However, with current technology it is difficult to measure the orientation in vivo. The model-based approach of Kroon et al. might be particularly suitable for patient specific modelling and simulation, which would profit from accurate reconstruction of fibre orientation. The study of Phatak et al. (2008) presents and evaluates an approach for estimation of strain distributions in the left ventricle by magnetic resonance imaging (MRI) and image registration. The model-based approach applies high resolution, non-tagged cine MRI, which is an imaging modality readily accessible in most hospitals. The approach might be clinically relevant, because analysis of strain distributions facilitates assessment of cardiac performance. Kerckhoffs et al. (2008) showcase a computational study of failing hearts and effects of biventricular pacing. An electro-mechanical model of ventricles was coupled to a model of systemic and pulmonary circulation to provide insights into the relationship between scar size and heart function. These insights might help to predict ventricular remodelling and long-term effects of biventricular pacing. The study of Pop et al. (2008) shows the importance of model parameterization and validation. A ventric-