Fig 1.
Imaging processing for free-breathing, real-time acquisitions.
(A) All slices (10–12 SAX and 6 each of 2 and 4-chamber LAX) are viewed simultaneously for selection of those for analysis. All SAX slices with myocardium and a single 2- and 4-chamber LAX are selected for analysis. A full cardiac cycle for each selected slice is extracted, from which end-diastolic (B) and end-systolic (C) images are identified and endocardial (red) and epicardial (green) borders are traced.
Fig 2.
CMR acquisition protocol.
Fig 3.
Sample two-chamber long-axis images.
Images for the same individual at end-diastole (top row) and end-systole (bottom row) for the conventional gated, end-expiration breath-hold acquisition (first column), real-time, free-breathing acquisition at rest (middle column); and real-time, free-breathing acquisition at peak exercise (right column).
Fig 4.
Bland-Altman plots of left ventricular volumes and function and heart rate.
Comparison of gold standard gated, breath-hold, short-axis evaluation to a real-time, free-breathing, biplane long-axis evaluation. Dotted lines represent mean difference ±1.96 times the standard deviation of the differences. Abbreviations: EDV, end-diastolic volume; ESV, end-systolic volume; SV, stroke volume; EF, ejection fraction; CO, cardiac output; GLS, global longitudinal strain; HR, heart rate.
Table 1.
Free-breathing LAX acquisitions versus breath-hold and free-breathing SAX acquisitions for resting LV volumes and function, and heart rate.
Table 2.
SAX versus LAX real-time free-breathing approaches for peak exercise and reserve LV volumes and function.
Table 3.
Test-retest reliability by intraclass correlation of real-time, free-breathing, long-axis evaluation approach.
Fig 5.
Linear relationship between peak cardiac output and peak volume of oxygen consumption.
Cardiac output measured via long-axis, real-time, free-breathing acquisition. Both variables are normalized to body weight.