Antral contraction rate estimation from dynamic antral scintigraphy using Hilbert-Huang transform

Abstract

Conventional analysis of the gastric antral contraction rate (ACR) utilizes the Fourier transform (FT) which does not effectively capture the non-stationary property of dynamic antral scintigraphy (DAS). In this study, we showed that application of Hilbert-Huang transform (HHT) on DAS yielded better estimates of ACR. Specifically, the time activity curves were extracted from the DAS data of 18 healthy volunteers and subjected to FT and HHT analyses. Comparison of the mean, standard deviation (SD), and root mean square error (RMSE) of ACR estimated by both methods showed that the proposed HHT method yielded significantly smaller SD (p<0.00001), smaller relative SD (13.3% versus 53.7%) and RMSE (0.72 cpm versus 1.59 cpm). Moreover, the HHT method also achieved lower relative SD of the frequency values from the intrinsic mode functions. Overall results indicated that the HHT method outperformed the conventional FT method in estimating the ACR from DAS. We anticipate that our approach will lead to development of effective noninvasive diagnoses of gastrointestinal tract diseases using DAS.