Wavelet denoising of vaginal pulse amplitude

Authors


  • This work was supported by a training grant for the second author from the National Science Foundation (EPS-0447689).

Address reprint requests to: Nicole Prause, Department of Psychology, Idaho State University, Pocatello, ID 83209, USA. E-mail: nprause@isu.edu

Abstract

Vaginal pulse amplitude (VPA) has been the most commonly analyzed signal of the vaginal photoplethysmograph. Frequent, large, and variable-morphology artifacts typically have crowded this signal. These artifacts usually were corrected by hand, which may have introduced large differences in outcomes across laboratories. VPA signals were collected from 22 women who viewed a neutral film and a sexual film. An automated, wavelet-based, denoising algorithm was compared against the uncorrected signal and the signal corrected in the typical manner (by hand). The automated wavelet denoising resulted in the same pattern of results as the hand-corrected signal. The wavelet procedure automated artifact reduction in the VPA, and this mathematical instantiation permits the comparison of competing methods to improve signal:noise in the future.

Ancillary