A Soft-Robotic Biomimetic Benchtop Model for Esophageal Motility Simulation

Abstract

Large animal models, while valuable, are expensive, time-consuming, and limited to discrete interventional or terminal timepoints, while existing benchtop models do not offer an accurate representation of the esophageal environment. Moreover, current pre-clinical models cannot effectively simulate swallowing dysfunction (dysphagia), restricting progress in understanding motility disorders like achalasia and hindering evidence-based dietary recommendations. In response, we present RoboGullet, a biomimetic soft-robotic model with independent localized longitudinal and circumferential muscle actuation, enabling, for the first time, simulation of both normal and diseased esophageal motility. We further enhance realism with a biohybrid variant, RoboGullet +, incorporating porcine esophageal mucosa/submucosa. We demonstrate this platform’s versatility through three key applications: assessing stent migration, simulating achalasia I-III within clinical diagnostic criteria, and analyzing bolus swallowing. Our findings reveal that: (1) stent migration increases over fivefold when incorporating longitudinal muscle movement versus isolated circumferential; (2) using a viscous non-Newtonian bolus improves high-resolution manometry diagnostic sensitivity of Achalasia III through increasing the Distal Latency diagnostic metric by 20.83%; and (3) stirring Greek-style yoghurt (common non-Newtonian dietary recommendation) significantly improves bolus transit versus unstirred for Achalasia Types I-II patients. This establishes RoboGullet+ as a powerful translational tool, advancing our understanding of esophageal motility and its therapeutic interventions.