Purpose The purpose of this work is to compare the behavior of the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) in contrast-enhanced MR angiography with background suppression performed by either a Dixon-type or subtraction-type method. utilizing one two and three echoes. Specifically the SNR and CNR ratios were compared as the concentration of contrast material in the blood changed. Results Empirical measurements of the SNR and CNR ratios compared favorably with the ratios predicted by theory. As the Bafetinib (INNO-406) contrast concentration was reduced the SNR advantage of the Dixon techniques increased asymptotically. In the ideal case the SNR improvement over subtraction contrast-enhanced MR angiography was at least twofold for one- and two-echo Dixon techniques and at least a factor of for the three-echo Dixon technique. Conclusion: Expressions showing a contrast concentration-dependent SNR and CNR improvement of at least a factor of two when Dixon-type contrast-enhanced MR angiography is used in place of subtraction-type contrast-enhanced MR angiography were derived and validated with phantom and in vivo experiments. = + η where the received signal and the true signal are complex vectors and η is proper complex (zero-mean) Gaussian noise with covariance Ψ Bafetinib (INNO-406) (22). The SNR of a single coil element is given in Eq. [1]. refers to the complex conjugate of a vector or matrix (·)* refers to the conjugate (Hermitian) transpose and (·)refers to the nonconjugate transpose of a matrix. [·] denotes the expected value Tr(·) denotes the matrix trace det (·) denotes the matrix determinant and ?· denotes the real part of a vector or matrix. Additionally (·)refers to the × (echoes × coils) matrix whose columns represent the multiecho signal from each of the coils. A matrix of Dixon system parameters = γΔ= 2πΔis the gyromagnetic ratio Δis the chemical shift in Hz and TE is the echo time of the acquired image. is a Mouse Monoclonal to Goat IgG. 2 × 1 column vector containing the true water and fat signals (W and F). is a 1 × matrix of Bafetinib (INNO-406) the sensitivities of the coil elements and × matrix of (zero-mean) proper complex Gaussian noise whose rows have covariance Ψ. That is is an echo (row) index and δis the vector which selects the and × 1. A Kronecker delta function (δ= [1 0]) is used to select from = [0 1] could be used. Thus the maximum likelihood estimate is the number of echoes is a weighted root sum of squares aggregate. SNR and CNR for Single-Echo Dixon CE-MRA Mirroring Eq. [4] the signal equation for single-echo multicoil Dixon acquisition can also be described in matrix form as matrix containing the signal from each coil of a single-echo acquisition. The matrix of Dixon parameters is now a 1 × 2 matrix and matrix containing zero-mean proper complex Gaussian noise with covariance Λ. Because is assumed to be real valued in the single-echo case the water and fat signals are constrained to have equal phase ?0 at time can be estimated from via least squares minimization: be real is part of the minimization process. The solution for the single-echo + δ+ + matrix containing the complex difference between the images matrix containing (zero-mean) proper complex noise with covariance ΛΔ = 2Λ. The multiplier of 2 is a result of the subtraction operation on the noise from the individual images (with covariance Λ). The reconstruction can again be solved using least squares minimization: is the number of echoes and echoes are evenly spaced around the unit circle Eq. [22a] reduces to: or Δθ (which depend on TE) on the SNR and CNR is shown in Figure 1d-f. The number of echoes where σ2 is the measured variance in the Bafetinib (INNO-406) noise. No noise correction was necessary as the complex (dual- and three-echo) or real (single-echo) image was used for noise estimation. Ratios of measured Dixon SNR to measured subtraction SNR were computed and compared against the ratio predicted by theory. In this study consistent with our experimental method the signal equation for an SPGRE sequence was used to calculate the signal terms of the theoretical SNR ratio. Phantom Experiments An anthropomorphic fat-water phantom was constructed to simulate an abdomen Bafetinib (INNO-406) with subcutaneous fat by layering solid vegetable shortening around a bovine gelatin Bafetinib (INNO-406) (b-gel) center (Fig. 2a). A cylindrical void in the center of the phantom allowed vials of.