SPIE Membership Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:
SPIE Photonics West 2018 | Call for Papers

OPIE 2017

OPIC 2017




Print PageEmail PageView PDF

Biomedical Optics & Medical Imaging

euHeart: integrated cardiac care using patient-specific cardiovascular modeling

A project based on biophysics aims to improve the diagnosis, planning, and treatment of cardiovascular disease.
1 June 2008, SPIE Newsroom. DOI: 10.1117/2.1200804.1126

Cardiovascular disease (CVD) is the cause of over 4.35 million deaths in Europe each year, including nearly half of all non-accidental deaths, and significantly reduces quality of life.1 CVD is most often a consequence of atherosclerosis, an inflammation of the arterial blood vessels, which manifests itself in diseases such as coronary artery disease (CAD), congestive heart failure (HF), and cardiac arrhythmias (irregular heart beat). Early detection and prediction of how cardiac diseases progress are key requirements for improved treatment outcomes and a reduction in mortality and morbidity.

The quantity and diversity of currently available data, which includes measurements of cardiac wall motion,2 chamber flow patterns,3,4 coronary perfusion,5 and electrical mapping techniques,6 present significant opportunities to improve clinical care of CVD. However, the clinical practice of using population-based metrics derived from different image sets often recommends contradictory treatment plans due to individual variability in pathophysiology.5 Despite advances in imaging techniques, determining optimal treatment strategies for patients remains problematic. Exploiting the full power of imaging technologies and the information produced from many different patients requires the ability to integrate multiple types of anatomical and functional data into a consistent framework.

Multiscale modeling as an integrated framework

An exciting and highly promising integration strategy involves the personalization of mathematical models based on biophysical measurements. Such models can capture the complex and multifactorial cause and effect relationships that link the underlying pathophysiological disease mechanisms. This in turn provides the capacity to locate factors that are not directly observable but play a key role in the pathology of disease, such as measures of pump efficiency and tissue stress, to assist treatment decisions. The most recent international efforts to develop these types of models have been organized in two initiatives: the International Union of Physiological Sciences (IUPS)-sponsored Physiome10 and the Virtual Physiological Human (VPH) project.11 Within these frameworks the heart is arguably the most advanced current exemplar of an integrated organ model and represents an excellent organ system to demonstrate the translation of models to clinical application.

Figure 1. Examples of models of the heart. (a) Anatomical model of the cardiac chambers, myocardium, and great vessels.7 (b) Patient-specific electromechanical simulation (shaded) computed from a segmented myocardium (wireframe) built from cine magnetic resonance images. The color code indicates the amount of active stress along the fibers. (Image courtesy of Hervé Delingette, Maxime Sermesant, and Nicholas Ayache, INRIA, France.8) (c) Geometric and coronary blood flow model.9 (d) Aortic flow streamlines. (Image courtesy of Rod Hose, University of Sheffield, UK, and Hendrik Von Tengg-Kobligk, German Cancer Research Center, Heidelberg, Germany.)

Detailed and anatomically finite models of the heart now accurately represent elements of both cardiac anatomy and detailed microstructure (see Figure 1). These mathematical descriptions serve as spatial frameworks for embedding functional cellular models of electrical activation, and the resultant tension generation that produces cardiac contraction. These cell and organ components have been combined through the application of continuum equations to simulate whole-organ cardiac electro-mechanics, and coronary and aortic blood flow: see Figure 1(b–d).

However, despite advances there remains a significant translational barrier to customizing and applying models for human clinical application. This is because the vast majority of cardiac models are currently developed and validated using data collected from invasive measurements in animal populations under controlled conditions. While these models are useful for proof-of-concept demonstrations of function, the development of mechanistic concepts, and interpretation of specific animal data, there are inherent limitations in the way animal models accurately mimic disease. Moreover, the relevance of insights gained from such animal models to human health remains difficult to determine.

The euHeart project

The current challenge is to quantify and personalize current frameworks for individual patients using extensive but minimally invasive clinical measurements. The goal of integrating and applying this information in patient-specific contexts is the focus of a four-year European Commission project—euHeart—which will soon be funded as part of the VPH section of the Seventh Framework Programme for research.12 This project will develop a library of several models of the heart and the great vessels using the established mark-up languages CellML and FieldML.13,14 Along with robust and effective personalization strategies, these models will enable the description of normal cardiac and pathological conditions for the major CVDs.

More specifically, models for HF and arrhythmias will incorporate descriptions of ion-channel kinetics into the organ-level solution of reaction-diffusion equations to compute the activation sequences in ventricular and atrial tissue. These applications will also couple the mechanisms for myofilament force generation at the subcellular level to the fluid-mechanical properties of the heart chamber and wall to capture contractile dynamics. Work on CAD will relate perfusion, the delivery of arterial blood to capillaries, to the coronary flow of the whole heart. Mechanical contraction and coronary perfusion will in turn be linked to the constitutive properties of aortic wall tissue to connect heart function and the pressure-flow behavior of the circulation.

This library of models and tools will be integrated into clinical applications to validate the descriptive capabilities of the approach and to demonstrate the plausibility of improving clinical outcomes—when compared to current best practices—for diagnosis, interventional planning, and treatment optimization. Model validation will be carried out on cohorts of patients, and one small-scale multicenter trial, including approximately 120 patients, will be performed to demonstrate the clinical benefit to determine the optimal lead placement in cardiac resynchronization therapy, a treatment that improves the coordination of the heart's contractions.

The long-term outcome of the euHeart program will be a consistent, biophysics-based framework for quantitative data integration, interpretation, and knowledge extraction using the new tools developed. Our hope and expectation is that this will support a paradigm shift away from predefined clinical indices that determine treatment options toward the true personalization of care based on specific individual physiology.

Olivier Ecabert
X-ray Imaging Systems
Philips Research Europe
Aachen, Germany

Nicolas Smith
Computing Laboratory
University of Oxford
Oxford, United Kingdom