A cardiac short axis is a transverse reference direction or imaging view relative to a specified cardiac long axis. For left-ventricular imaging, a short-axis plane is planned across the chamber rather than through its base-to-apex length. The chamber and anatomical landmarks must be specified; a patient's horizontal body plane is not automatically a cardiac short axis plane.
A primary cardiovascular magnetic resonance imaging protocol plans short-axis images using linked long-axis views and covers the ventricle from base to apex.[ref] A cardiac short-axis stack turns successive cavity outlines into volumetric information.
For an original geometrical construction, let be the unit vector along the chosen cardiac long axis. A transverse plane through satisfies
Parallel planes at different axial positions sample different portions of the ventricle. With measured cavity areas and represented slice thicknesses , a simple volume estimate is .
An oblique slice also illustrates a useful caution. An ideal circular cylinder cut at angle away from its perpendicular section has an elliptical cross-sectional area . At this is about larger. This does not prove that an entire consistently acquired oblique stack overestimates volume: integrating all areas with the correct normal spacing and complete coverage still recovers volume. Orientation, spacing and coverage must be treated together. The major practical question is whether the chosen stack and image segmentation represent the complete intended cavity, not whether one slice happens to look circular.
A cardiac short-axis stack is a series of images across a cardiac chamber at successive positions approximately transverse to its cardiac long axis. In ventricular volumetry, the series must cover the intended cavity from its basal boundary toward the cardiac apex. A primary cardiovascular magnetic resonance imaging protocol specifies linked long-axis planning and short-axis coverage.[ref]
The stack supplies a spatial sum rather than a single diameter. Let be the segmented cavity area in slice at phase , and let be the represented distance along the stack normal. A basic approximation is
All areas in a given sum must refer to a compatible phase of the cardiac cycle. Image segmentation, slice coverage and phase matching are separate inputs to the volume estimate.
An original example uses four representative areas of 2, 6, 10 and 6 cm, each assigned 1 cm of axial extent. Their represented volume is 24 cm, or 24 mL. Omitting the basal 6 cm contribution reduces the result to 18 mL without changing the tracing quality of the other slices. If slice thickness is 0.8 cm but center spacing is 1 cm, the numerical integration convention must also account for the intervening extent; treating thickness and represented spacing as interchangeable changes the estimate.
Phase mixing causes a different error. An area from one slice near end-filling and another near end-ejection do not generally describe a cavity that existed at one time. A stack can therefore be complete spatially yet inconsistent temporally. With correctly matched phases, the derived end-diastolic volume and end-systolic volume can be used for stroke volume and ejection fraction; those formulas do not repair missing anatomy or mismatched acquisition phases.

Articles by others on the same topic (0)

There are currently no matching articles.