Three-dimensional (3D) echocardiography is based on the acquisition of volumes containing heart structures (called volumetric or 3D data sets). Once a 3D data set is acquired, additional display maneuvers are required during post-processing in order to visualize the information enclosed. In contrast with live imaging by 2D echocardiography, 3D echocardiography enables in addition the possibility to (re)generate retrospectively various views from a previously-acquired 3D data set.
This chapter will provide a brief overview of the main modalities used by 3D echocardiography to obtain and display information.
Three-dimensional echocardiography - similarly to what the anatomist or the surgeon does - requires that a structure is first exposed by removing the different chamber walls in order to be able to visualize it. This process of virtually removing the unrelevant neighbouring tissue is called cropping (Figure 1), and can be performed either during or after acquisition.
In contrast with 2D images, displaying a cropped image requires also the definition of the viewing perspective (i.e. since the same 3D structure can be visualized en face either from above or below, as well as from any desired view angle)..
Slicing refers to a virtual “cutting” of the 3D data set into one or more 2D (tomographic) grey-scale images (Figure 2).
There are fundamentally three ways of displaying information contained in a volumetric data set: volume rendering, surface rendering and (multi-)slice rendering.
The echo system can simulate a real 3D appearance of the anatomy of a cardiac region by displaying a stereoscopic image called “volume rendering” or “3D rendering”.
There are several techniques used to obtain a 3D rendering of a cardiac structure (i.e. mitral valve etc) on the flat 2D monitor of the echo machine.
- Opacity function - modulates the grey-level of the 3D data set: opaque regions (pixels) correspond to tissue, while transparent regions to blood (Figure 1). When 3D flow information is added, color Doppler encoding of blood flow is simultaneously displayed with tissue 3D rendering (Figure 3).
- Tissue colorization - is used to enhance the depth perception: the distances from the examiner’s viewpoint are estimated by the system and the rendered data is displayed using various depth-encoding color maps (Figure 4).
- Additional functions (shading, smoothing etc) - can be added for improving the quality of 3D rendering.
- Stereo rendering - is achieved by displaying the 3D data set from two slightly different viewing angles and can be appreciated with the use of special red/cyan glasses (Figure 4, Panel B).
Once a 3D rendering of a cardiac structure is optimized, it can be displayed in various ways (layouts) on the echo monitor:
- as a single 3D rendered image
- displayed side-by-side with 2D images (2 orthogonal or 3 longitudinal slices at 60° obtained from the 3D rendering) (Figure 5)
- as a rotating 3D rendering (from left to right and backwards), to allow a wider view angle without any intervention from the examiner (Video)
Volume rendering is preferentially used to display cardiac anatomy and relationship with neighbouring structures.
An alternative way of displaying cardiac structures and preserve the 3D visual perception is to generate solid surface models in a 3D scene (Media: Surface rendering of the left ventricle.jpg). In contrast with 3D rendering, the information of the tissue beneath the surface is missing.
The prerequisite of this technique is for the echo system to know the blood-tissue boundaries of the structure to be displayed. Therefore, automatic or semi-automatic (requiring human interventions) outlining of the structure is needed before generating the surface rendering.
After obtaining the 3D surface rendered model, various functions - colorization, shading, brightness, tissue transparency - are automatically applied to increase depth perception and to outline regional characteristics (LV segmentation, abnormal surface shape in mitral valve prolapse etc).
Wire-frame represents a variant of surface rendering, looking like a cage-like image (Media: Wire-frame rendering of the left ventricle.jpg). It can be applied solely, or combined with solid surface rendering.
Surface rendering is mainly used to display the size, dynamic shape and function of cardiac chambers or valves (Figure 6).
From a 3D data set, one can obtain stereoscopic images of heart structures in motion, but also various 2D views of the same structures. Slicing allows the examiner to obtain several tomographic (2D) views from the same volumetric data set, which can be reviewed side-by-side on the same screen layout.
The number of slices (from 1 to 12) can be manually selected by the examiner, according to the purpose.
The cut-planes can be either individually elected by the examiner (according to the best orientation and angulation that displays the anatomic feature to be assessed, particularly for those impossible to obtain by standard 2D echocardiography) or along several pre-defined long-axis or short-axis cuts.
To obtain the desired views, slice orientation within the 3D volume can be adjusted by 2 maneuvers: translation (parallel) or angulation (rotation relative to initial plane position).
Slice rendering can be used to assess valvular stenotic orifices or leaflet prolapse, regional wall motion (Media: Multi-slice display of left ventricle.jpg) or to obtain anatomically correct 2D images in difficult cases (e.g. correcting for foreshortening or off-axis views by 2D echocardiography).
- ↑ Lang RM, Badano LP, Tsang W et al. EAE/ASE recommendations for image acquisition and display using three-dimensional echocardiography. Eur Heart J Cardiovasc Imaging 2012; 13(1): 1-46
1. Badano LP, Muraru D. Three-dimensional echocardiography in clinical practice, in Badano LP, Lang RM, Zamorano JL (eds). Textbook of Real-Time Three Dimensional Echocardiography, Springer-Verlag London 2011:33-44.