Bayesian Functional Mixed-Effects Model with Gaussian Process Responses for Wavelet Spectra of Spatiotemporal Colonic Manometry Signals

TL;DR

This paper introduces a Bayesian functional mixed-effects model utilizing Gaussian processes and wavelet transforms to analyze complex spatiotemporal colonic pressure signals, enabling detailed spectral comparisons across conditions.

Contribution

It presents a novel statistical framework combining wavelet analysis and Gaussian process modeling for functional responses in colonic manometry data.

Findings

Successfully identified spectral differences between colonic regions.

Detected changes in pressure signals in response to a meal.

Provided a fast, detailed analysis method for manometric signals.

Abstract

Objective: We present a technique for identification and statistical analysis of quasiperiodic spatiotemporal pressure signals recorded from multiple closely spaced sensors in the human colon. Methods: Identification is achieved by computing the continuous wavelet transform and cross-wavelet transform of these recorded signals. Statistical analysis is achieved by modelling the resulting time-averaged amplitudes or coherences in the frequency and frequency-phase domains as Gaussian processes over a regular grid, under the influence of categorical and numerical predictors that are specified by the experimental design as a mixed-effects model. Parameters of the model are inferred with Hamiltonian Monte Carlo. Results and Conclusion: We present an application of this method to colonic manometry data in healthy controls, to determine statistical differences in the spectra of pressure signals between different colonic regions and in response to a meal intervention. We are able to successfully identify and contrast features in the spectra of pressure signals between various predictors. Significance: This novel method provides fast analysis of manometric signals at levels of detail orders of magnitude beyond what was previously available. The proposed tractable mixed-effects model is broadly applicable to experimental designs with functional responses.