“…Indeed, a considerable amount of information is lost by considering only these averages and calls for a different statistical approach for the diagnosis of the patient brain condition. Several complementary methods have been developed to quantify the predictive value of the MMN for comatose patients: some are based on t-test in time (100-200 ms) to check the presence of a detectable peak of the mismatch negativity, others are based on wavelet transform, multivariate, cross-correlation and probabilistic methods [4,5,6,7,8,9,10,11]. Finally, we still missing a consensus about the quality of the existing features [12,2,13] to predict the outcome of post-anoxic coma.…”