Noise analysis in magnetic resonance electrical impedance tomography at 3 and 11 T field strengths

Abstract

In magnetic resonance electrical impedance tomography (MREIT), we measure the induced magnetic flux density inside an object subject to an externally injected current. This magnetic flux density is contaminated with noise, which ultimately limits the quality of reconstructed conductivity and current density images. By analysing and experimentally verifying the amount of noise in images gathered from two MREIT systems, we found that a carefully designed MREIT study will be able to reduce noise levels below 0.25 and 0.05 nT at main magnetic field strengths of 3 and 11 T, respectively, at a voxel size of 3 x 3 x 3 mm(3). Further noise level reductions can be achieved by optimizing MREIT pulse sequences and using signal averaging. We suggest two different methods to estimate magnetic flux noise levels, and the results are compared to validate the experimental setup of an MREIT system.

Direct detection of neural activity in vitro using magnetic resonance electrical impedance tomography (MREIT)

“…Each measure, ϕ NC,k , ϕ d,k or ${\varphi }_{d,k}^{\mathrm{italicAVG}}$, was also affected by instrument noise ε , which depends inversely on magnitude image SNR (Sadleir et al, 2005). The predicted baseline noise in subtracted phase images ϕ d,k , is (Sadleir et al, 2005)…”

“…Specifically, we determined an average over a portion of the magnitude image in the test chamber interior, and dividing this value by the standard deviation of a portion outside the test chamber. The result was multiplied by a factor of 0.655 to account for the difference between noise distributions inside and outside the test chamber (Sadleir et al 2005). The scans used in determining magnitude SNR for each of the 24 ganglia experiments was the second PRE scan, i.e., I 1+ .…”

“…Conductivity distributions within the chamber σ ( x , y , z ), including any heterogeneous tissue regions, will dominate current flow characteristics and therefore magnetic flux density distributions. Conductivity characteristics may in general depend on time, thus $$\mathit{J}=-\sigma (x,y,z,t)\nabla V(x,y,z).$$ The phase contribution in MR images due to externally applied currents, γT c B z (Sadleir et al, 2005), therefore depends on sample chamber shape, location of current application ports and the internal conductivity distribution, that is $$B_z=B_z(x,y,z,\sigma (t)).$$ Consider transverse ( xy plane) phase images with or without current, recovered from a sample chamber containing live tissue in experimental segment k = PRE , POST or DEAD . From Table 1, we observe that within each segment, images were acquired for NC , 1+ , 1− , 2+ and 2− , sampling properties within the chamber over an approximate 27-minute period. For NC , we measure…”

“…The phase contribution in MR images due to externally applied currents, γT c B z (Sadleir et al, 2005), therefore depends on sample chamber shape, location of current application ports and the internal conductivity distribution, that is…”

Analysis and Blocking of Error Propagation by Region-Dependent Noisy Data in MREIT

“…To improve the quality of reconstructed conductivity images, we carefully investigate noise characteristics in measured B z data, which are extracted from MR phase images. From the noise analyses by [21] and [19], the noise standard deviation in measured B z data is inversely proportional to the signal-to-noise ratio (SNR) of the corresponding MR magnitude image. Considering numerous factors affecting the magnitude image SNR [4], it is evident that the SNR varies from one region to another.…”