As known the SNR evaluates the efficiency of Adaptive filter used to separate heard sound from lung sounds but is it enough to judge? maybe there is an additional method to investigate. Please recommend me about that method .
I think that a simple way to evaluate the efficiciency of any adaptive filter could be to mesasure the residual correlation (or spectral flatness) in the output signal. This could be use for example when your adaptive filter acts as a decorrelation filter (or whitening filter).