Adaptively Optimized Combination (AOC) of Phased-Array MR Spectroscopy Data in the Presence of Correlated Noise: Compared with Noise-Decorrelated or Whitened Methods
A method for adaptively optimized combination (AOC) of MR spectroscopic data from a coil array was recently introduced. The superior performance of the AOC method is evident when compared with the methods that assume uncorrelated noise between coil elements. However, it is unclear whether the AOC method represents the most optimal combination in the presence of correlated noise, when compared with the noise-decorrelated or whitened methods that specifically tackle the correlated noise between coil elements.Methods:
A new, unified theoretical framework was developed to illustrate the relationship between the AOC method and three noise-decorrelated or whitened methods, namely, noise-decorrelated combination (nd-comb), whitened singular value decomposition (WSVD), and improved WSVD (WSVD+Apod). Simulation-based comparisons and in vivo human brain experiments on a 3 Tesla (T) MRI scanner were performed using an 8-channel phased-array head coil.Results:
Compared with the noise-decorrelated or whitened methods, the AOC method consistently yielded the best combination in terms of the robustness against noise and maintaining the combined spectrum from distortion, and the superior performance was most evident at a low signal-to-noise ratio (SNR).Conclusion:
The AOC method represents the theoretical optimal combination in the presence of correlated noise between coil elements, whereas the three noise-decorrelated or whitened methods are asymptotically optimal.