Table 1.
Acquisition parameters used for acquiring cardiac image series at various resolutions and acceleration factors.
Table 2.
Variations in FOV and spatial resolution in the different subjects.
Resolution is reported for matrix sizes of 192 x (96, 128, 192).
Fig 1.
Representative image, SNR, TSNR, and σP/σT maps from a healthy volunteer.
Images were acquired with a matrix size of 192x192, R = 1, Resolution = 1.67x1.67mm2. The SNR, TSNR, and σP/σT ratio within the left ventricular myocardium ROI (white dashed line) was 51.3± 9.8, 30.5±10.2, and 1.49 ± 0.56 respectively. The noise ratio map is lowest in areas with little movement, such as the muscles of the back and highest in areas with large fluctuations in signal intensity such as blood vessels. In a static phantom and stable imaging environment, TSNR = SNR. In vivo, TSNR ≤ SNR due to additional sources of σP. The σP/σT ratio determines whether σP or σT is the dominant source of noise. When σP is dominant (σP/σT >> 1), SNR gains will not improve TSNR.
Fig 2.
Plots of temporal signal-to-noise ratio (TSNR) versus signal-to-noise ratio (SNR).
Separate plots are for A) all LV myocardium, C) septal myocardium and D) lateral myocardium, from mid-short axis slices. Septal and lateral segments are illustrated in B). Each point represents an average over four subjects while the error bars represent the standard deviation within the ROI averaged over four subjects. bSSFP acquisitions are shown in solid red while GRE acquisitions are shown in outlined blue. The solid black line represents the fit to the PNC model along with the lower and upper 95% confidence interval shown in dotted black. The dashed black line indicates the line of unity. Vertical and horizontal dashed red lines represent when the noise ratio is equal to one and the asymptotic limit of TSNR respectively. Both occur at 1/λ. At low SNR, TSNR varies linearly with SNR while at high SNR, TSNR approaches an asymptotic limit. The lateral myocardium also exhibited higher physiological noise (λ = 0.0242, α = 0.5581) and hence lower asymptotic TSNR than the septum (λ = 0.0162,α = 0.7583).
Fig 3.
Comparison of SNR, TSNR, and the σp/σt for several image matrix sizes and acceleration factors.
A) GRE and B) bSSFP. Matrix sizes of 192×96, 192×128, and 192×192 are shown in blue, green, and yellow respectively. As expected, SNR decreases with increasing matrix size and higher acceleration factors. In GRE, TSNR follows a similar trend and the noise ratio is < 1 for all imaging settings. In bSSFP, TSNR remains relatively constant and the noise ratio is ≥1 at low acceleration factors and approaches 1 only at higher accelerations and larger matrix sizes.