Objective: Gastrointestinal magnetic resonance imaging (MRI) provides rich spatiotemporal data about the movement of the food inside the stomach, but does not directly report muscular activity on the stomach wall. Here we describe a novel approach to characterize the motility of the stomach wall that drives the volumetric changes of the ingesta.
Methods: A neural ordinary differential equation was optimized to model a diffeomorphic flow that ascribed the deformation of the stomach wall to a continuous biomechanical process. Driven by this diffeomorphic flow, the surface of the stomach progressively changes its shape over time, while preserving its topology and manifoldness.
Results: We tested this approach with MRI data collected from 10 rats under a lightly anesthetized condition, and demonstrated accurate characterization of gastric motor events with an error in the order of sub-millimeters. Uniquely, we characterized gastric anatomy and motility with a surface coordinate system common at both individual and group levels. Functional maps were generated to reveal the spatial, temporal, and spectral characteristics of muscle activity and its coordination across different regions. The peristalsis at the distal antrum had a dominant frequency and peak-to-peak amplitude of [Formula: see text] cycles per minute and [Formula: see text] mm, respectively. The relationship between muscle thickness and gastric motility was found to be distinct between two functional regions in the proximal and distal stomach.
Conclusion: These results demonstrate the efficacy of using MRI to model gastric anatomy and function.
Significance: The proposed approach is expected to enable non-invasive and accurate mapping of gastric motility for preclinical and clinical studies.