Unbiased measurements of reconstruction fidelity of sparsely sampled magnetic resonance spectra

The application of sparse-sampling techniques to NMR data acquisition would benefit from reliable quality measurements for reconstructed spectra. We introduce a pair of noise-normalized measurements,and, for differentiating inadequate modelling from overfitting. Whileandcan be used jointly for methods that do not enforce exact agreement between the back-calculated time domain and the original sparse data, the cross-validation measureis applicable to all reconstruction algorithms. We show that the fidelity of reconstruction is sensitive to changes inand that model overfitting results in elevatedand reduced spectral quality.