Zu den Inhalten springen

4D Imaging / Cardiac NanoDynamics

The 4D Imaging Group combines experimental and computational research into cardiac structure (3D) and function (i.e. over Time). The spatial domains studied can be split roughly into nano-to-micro and micro-to-macro levels, while the temporal domain requires resolution from milliseconds (say to resolve a cardiac contraction cycle) to months (for disease-related remodelling).


Precise 3D nano-structural insight into the highly compartmentalised and densely-populated environments inside cells of the cardiovascular system is only beginning to emerge. In collaboration with teams at Boulder, Bristol, and Heidelberg, we are pursuing electron microscopic (EM) tomography – an imaging approach where thin fragments of a cell are studied using high voltage EM from over 200 viewing angles. With this approach one is able to reconstruct 3D EM data with a resolution of ≈4*10‑9 m (Fig. 1).

Nano-to-Micro + Time

Given that the samples studied by EM are fixed, the temporal domain involves taking multiple snapshots of structure, either over the progression of cardiovascular disease, of at different stages of the cardiac cycle. It is possible, therefore, to observe changes in sub-cellular compartment architecture, such as contraction-induced lateral compression of T-tubules (Fig. 2).

Other techniques, such as micro-manipulation, can be used in real time to control or monitor electrical, ionic, and mechanical properties of cardiovascular cells (Fig. 3).


Cardiac cells offer a convenient mid-point between molecular and systemic research, as they can be isolated from organ or biopsy material and studied literally ‘in isolation’. Projecting between cell and organ behaviour is not easy, as cells integrated into the normal tissue are less easily manipulated and observed. Here, computational modelling helps in reconstructing 3D histo-anatomical features from serial imaging data (combining non-invasive imaging such as high-resolution MRI, with whole-heart histology). An example of this approach is shown in Fig. 1 of the Bioinstrumentation pages. These data-sets can then be integrated to build computational models of individual cardiac structure and function, or to extract cell-level deformation data for any position inside the heart.

Micro-to-Macro + Time

Deformation of the heart is more complex than we think. Even though modern computer simulations can be based on highly detailed insight into structural and functional properties of heart muscle – thus far they tend to fail in reproducing normal tissue deformation. We believe that this is due largely to an under-explored structural property of heart muscle – its layering into sheetlets (a bit like filo-pastry) that can not only slide past one-another, but also change the angle at which they intersect. First evidence of this has been obtained in magnetic resonance studies (Fig. 4), and this is now being pursued in computational models.