Adaptive coil combination
Adaptive coil combination is a method used in Magnetic Resonance Imaging (MRI) to merge signals from multiple receiver coil elements into a single image.
≈ 4 min read
Adaptive coil combination. encyclopegia.org, snapshot enwiki-001338780259. /a/adaptive-coil-combination
Names in other languages
No names in other languages are held for this entry yet.
Article bodies are English and are not machine-translated. A name in another language is not a translation of the article.
Adaptive coil combination is a method used in Magnetic Resonance Imaging (MRI) to merge signals from multiple receiver coil elements into a single image. A weighted sum of the individual coil images is performed with a different weighting vector \mathbf{m} for each pixel.
Each vector \mathbf{m} maximizes the signal-to-noise ratio (SNR) of a region of interest (ROI) around the pixel. \mathbf{m} is calculated using the following equations derived by David O. Walsh: \begin{align} &R_N=\sum_{(x,y)\in{ROI}}\mathbf{n}(x,y)\mathbf{n}^H(x,y)\\ &R_S=\sum_{(x,y)\in{ROI}}\mathbf{c}(x,y)\mathbf{c}^H(x,y)\\ &\mathbf{m}=\textbf{v}_{max}(R_N^{-1}R_S) \end{align} For a system with l coils, \mathbf{n}\in\mathbb{C}^{l} and is a column vector of the noise of each coil at location x,y. This can be obtained by capturing images without a subject, or if noise is assumed to be uncorrelated white, R_N becomes identity. H is the conjugate transpose. \mathbf{c}\in\mathbb{C}^{l} and is the measured value of signal + noise at location x,y. \textbf{v}_{max} denotes the largest eigenvector of R_N^{-1}R_S .
R_S is an estimate of the signal correlation matrix, which works in practice because signal is fairly constant over a small ROI, but thermal noise is white in the image domain so spatial averaging reduces noise-induced bias. The vectors \textbf{m} can be concatenated into a coil sensitivity map and used for techniques like parallel imaging.
Derivation
The following derivation was first published by Walsh. We wish to find a vector \mathbf{m}\in\mathbb{C}^{l} that maximizes SNR over an ROI with p pixels and l coils.
If we put the measured signal in our ROI into a matrix S\in\mathbb{C}^{l \times p} , and measured noise into a matrix N\in\mathbb{C}^{l \times p} we can write the SNR as: \begin{align} &Power_{signal}=\frac{|\mathbf{m}^HS|^2}{l}=\frac{\mathbf{m}^HSS^H\mathbf{m}}{l}\\ &Power_{noise}=\frac{|\mathbf{m}^HN|^2}{l}=\frac{\mathbf{m}^HNN^H\mathbf{m}}{l}\\ &\frac{Power_{signal}}{Power_{noise}}=\frac{\mathbf{m}^HSS^H\mathbf{m}}{\mathbf{m}^HNN^H\mathbf{m}}=\frac{\mathbf{m}^HR_S\mathbf{m}}{\mathbf{m}^HR_N\mathbf{m}}\\ \end{align} Because R_S and R_N are Hermitian, we can perform a simultaneous diagonalization with a new matrix P by requiring: \begin{align} &P^HR_NP=I\\ &P^HR_SP=D \end{align} where I is identity and D is diagonal. By multiplying the two equations we get: \begin{align} &P^HR_SP=P^HR_NPD\\ &R_SP=R_NPD\\ &R_N^{-1}R_SP=PD \end{align} It can be seen that P and D are the eigenvector and eigenvalue matrices respectively of R_N^{-1}R_S .
Performing a change of basis with \mathbf{q}=P^{-1}\mathbf{m} results in: \frac{Power_{signal}}{Power_{noise}}=\frac{\mathbf{q}^HP^HR_SP\mathbf{q}}{\mathbf{q}^HP^HR_NP\mathbf{q}}=\frac{\mathbf{q}^HD\mathbf{q}}{\mathbf{q}^H\mathbf{q}} This is the Rayleigh quotient and so the maximum value of \mathbf{q} corresponds to the maximum eigenvector of D, which is \textbf{q}_{max}=[1,0,\cdots,0] when D is sorted by descending order. Therefore \textbf{m}_{max}=P\textbf{q}_{max}=\textbf{v}_{max}(R_N^{-1}R_S) .
Text derived from https://en.wikipedia.org/w/index.php?oldid=1338780259 under CC BY-SA 4.0; modifications have been made.