3D histologically-detailed reconstruction of individual beating hearts: tools and application

The heart is an electrically controlled mechanical pump, whose dysfunction is incompatible with life. Both normal and disturbed activity are closely related to the fine architectural detail of the tissue that makes up the heart. Thus, the electrical signals that initiate contraction must first travel along every single of the millions of muscle cells in the heart, before each individual cell will shorten at its prescribed timing. Similarly, the forces produced by individual cells interact with, and are transmitted through, cell chains often referred to as muscle fibres. These muscle fibres are ordered in complex three-dimensional (3D) arrangements, joined together by non-muscle 'connective' tissue. This assembly is fundamental to healthy cardiac function, allowing the muscle not only to shorten, but also to thicken, in order to push blood out of the cardiac chambers and into the arteries that supply all organs of the body. A precise understanding of detailed cardiac tissue architecture would be of great importance for diagnosis of cardiac diseases, prediction of their progression, identification of useful treatment approaches, and even for doctors' teaching and training (the heart is one of the organs where 'learning by mistake' is not a viable option!). This clinical relevance is contrasted by the fact that, traditionally at least, establishing the architecture of any tissue meant 'slicing it up'. The study of cadaver organs, while historically fundamental in establishing anatomical insight, is equally not helpful to determine the properties of the beating heart. Recent improvements in non-invasive techniques, such as Magnetic Resonance Imaging (MRI), allow increasingly detailed insight into structure and function of internal organs. Even though these recordings are not yet sufficiently detailed to reliably identify fibre orientation in a patient's heart, the technology is moving in that general direction, and it is important that we start now to develop the tools required to handle the vast amount of data that doctors will be able to extract. This represents a major challenge, as it requires a combination of skills and expertise not usually present in a single lab or clinic. These include: advanced MRI techniques to explore the fine-structure of heart muscle while beating; automated image analysis and alignment of data from multiple sources; creation of computationally usable three-dimensional (3D) data sets and characterisation of their changes over the cardiac cycle (3D+Time); comprehensive validation of these techniques by comparison to the current gold-standard of histology for the whole organ; 3D+T visualisation and user-interaction; application of all of the above to an individual within a time-frame that makes 'clinical sense' (hours, not months). This project undertakes to develop exactly these technologies by combining the expertise of leading teams in cardiac MRI (Cardiovascular Medicine at the John Radcliffe Hospital Oxford), bio-medical studies (Oxford University Department of Physiology, Anatomy & Genetics), and image analysis and computational modelling (Oxford University Department of Engineering Science). These teams will jointly implement and validate the whole range of tools required to efficiently reconstruct individual beating hearts from entirely non-invasive imaging techniques, based on proof-of-principle work in small rodents, but taking care that all algorithms are scalable to be adapted, in future, to human heart sizes. The longer-term vision is that after a clinically-indicated cardiac MRI, doctors will be able to look at a 3D+T representation of the patient's heart, zoom-in on any relevant detail (a coronary vessel blockage or a damaged part of tissue), assess treatment options, and predict outcomes for the specific individual before the patient even enters the operating theatre. Much of this vision is still far ahead. Nonetheless, this proposal will make an important step towards this goal.

Tissue structure is a key determinant of cardiac mechano-electric function in health and disease. Knowledge of individual 3D cardiac histo-architecture and its dynamic alteration during the cardiac cycle (3D+T) would offer significant benefits for basic research, clinical application, and teaching. This study proposes: to establish Diffusion-Tensor (DT) Magnetic Resonance Imaging (MRI) for non-invasive structural characterisation of living myocardium at different mechanical states; to develop image processing tools to efficiently analyse and combine data from anatomical, functional and DT-MRI with cross-validation by whole-heart serial histology; and to apply these techniques to one test case of regional tissue inhomogeneity (infarct). Image analysis will include the development of suitable registration and segmentation algorithms to combine imaging scans and to generate realistic finite element meshes to be used in computer simulations of electro-mechanical heart action. Image acquisition will focus initially on characterisation of Langendorff-perfused hearts ex vivo, arrested in two different mechanical states (slack, contracture), before moving on to gated acquisition of data in the beating heart. We will implement an image acquisition pipeline, starting with the whole animal in vivo, followed by ex vivo and in vitro MRI, and concluded by histological reconstruction of 6 hearts each for control, sham and infarct (half fixed in systole, the rest in diastole). Sham and infarct (the test case for local tissue anisotropy; induced by permanent occlusion of the left coronary artery) rats will be scanned longitudinally (before and at 3 time points post-intervention). The developed methods will provide a validated tool-box as proof-of-principle before application to larger hearts (contributing to the technological basis for future patient-specific tools), and comprehensively validated datasets and models for individualised beating hearts in norm and disease.

Several areas/groups of beneficiaries of the proposed project can be identified: Firstly, the project will facilitate and benefit pre-clinical research. Given the multi-disciplinary nature of the project, the three PIs and their Departments represent large sections of this group, including basic bio-research, image analysis/modelling of biological systems, cardiovascular medicine, and magnetic resonance imaging. Furthermore, the PIs have strong collaborations at national and international levels (as partially reflected in this proposal). The outcome of this proposal is of direct relevance for the research of our teams and their collaborators in areas such as structure-function studies, mechano-electric interactions research, cardiac pump function, arrhythmias and defibrillation modelling (for an illustration, see lists of publications of the PIs). Secondly, in the longer run, the project will serve the medical community by paving the way for development of tools and models to aid the diagnoses, planning of interventions, and tailoring of treatments and, hence, help to improve prognosis for patients with heart disease, whether congenital or acquired. Computer models developed under this proposal will further contribute to education and training of medical practitioners by providing mechano-electrical information, combined with structural, anatomical and functional data of the heart, in 3D+T. More specifically, (patho-)physiological processes can be illustrated to explore conditions found clinically in cardiac patients. Thirdly, the project embraces the concept of the 3Rs (Reduction, Refinement, and (partial) Replacement of animal-based research). Advancing MR techniques to be applicable, non-invasively, to high-fidelity structure-function studies enables a reduction in the number of animals required for research. This finds reflection in our study design, where each heart serves as its own control (imaging multiple mechanical states; using multiple imaging modalities; going from in vivo to ex vivo and in vitro), where disease development is studied in a partially longitudinal design, and where (in as far as possible) non-biological phantoms are used for method development. Stringent intra-individual control and validation provides for more powerful statistical analyses in an otherwise highly inhomogeneous biological model. The development and cross-validation of computer models further helps to reduce the use of animals for research, development, and teaching. Fourth, it is clear that there is significant potential for commercial benefit. This includes manufacturers of imaging devices (e.g. see Letter of Interest from the Agilent Technologies MR Imaging group) who will be able to assess the benefit of newly developed MRI techniques; providers of biomedical image analysis and visualisation software who will have access to new image alignment and structure extraction tools; and - in the longer run - pharmaceutical companies who will eventually benefit from the availability of new technologies to test the efficacy and side effects of drugs using either the newly developed imaging techniques or the computational models that will arise. Last but not least, considering the burden heart disease places on the individual and their families, the economy, and the social and health systems in the developed world, the techniques and tools provided by this proposal will be of general societal benefit, as the outcome of this work will fundamentally contribute to the understanding of normal and patho-physiological mechano-electrical activity of the heart.