Effect of noise on peak heights calculated using an exponentially modified gaussian peak shape model

Abstract Most computer-based methods for finding chromatographic peak heights are relatively crude, relying on finding an appropriate baseline, then measuring a maximum signal height relative to the baseline. The error in finding a precise signal height of a weak signal can be increased by noise spikes. In this article, data are presented to show that the use of the exponentially modified gaussian peak shape model can effectively increase the quality of height measurements of peaks deliberately degraded to near undiscernability by dilution.