Evaluation of an audio-based monitoring system for detecting retching and vomiting in gastroparesis: an observational cohort study
4 affiliations
- 1Takeda Development Center Americas Inc.
- 2Takeda Pharmaceuticals U.S.A. Inc.
- 3University of Melbourne
- 4Department of Medicine, Division of Gastroenterology, Hepatology, and Nutrition, University of Pittsburgh School of Medicine
Abstract
This study evaluated a novel, audio-based digital monitoring system for detecting retching and vomiting (expulsion) events in a noninterventional, longitudinal study in 25 participants with gastroparesis (GP). Participants used an audio-based system comprising an iOS device (iPhone) and Watch in their homes for 14 days. Retching and expulsion events were quantified using human-annotated counts from audio recordings and daily electronic patient-reported outcome (ePRO) counts. Device compliance was calculated as recorded h divided by the total h (24 × 14). Compliance was high: 83.4% for the iPhone and 69.4% for the Watch. Event timing analyses from the audio devices revealed elevated retching and expulsion activity in the morning, with symptoms persisting during nighttime h. Watch and iOS event counts showed strong correlation (r > 0.85), though device agreement with ePRO counts was lower (r < 0.68). While ePRO-based expulsion counts showed stronger within-participant correlation with patient global impression of frequency/severity (PGI-F/PGI-S) measures, device counts better distinguished participants across PGI-F/PGI-S levels. Weekly averaging of event counts improved consistency across both audio-based and ePRO measures relative to daily event counts. Preliminary findings suggest audio-based technologies may provide an objective approach to detecting vomiting and retching, with future automation needed to support scalable implementation.
Related
- Office blood pressure measurement versus ambulatory blood pressure monitoring for detecting undiagnosed hypertension in an apparently healthy Mexican adult cohort: complementary central hemodynamic and arterial stiffness assessment
- Increasing Adoption Trends of Gastric Per-Oral Endoscopic Myotomy Over Gastric Electrical Stimulation for Refractory Gastroparesis: A Nationwide Analysis.
- Refractory Gastroparesis in Ehlers-Danlos Syndrome: Failure of Pyloric-Directed Therapy Despite Normal Pyloric Physiology.
- Comparative evaluation of diffusion models for detecting gray matter microstructural alterations in post-stroke depression patients
- Peer Evaluation Rating Behavior in Differentiating Student Performance in Undergraduate Medical Education: An Observational Study
BibTeX
@article{gong2026evaluation,
title = {{Evaluation of an audio-based monitoring system for detecting retching and vomiting in gastroparesis: an observational cohort study}},
author = {Yishu Gong and Jasmin Imsirovic and Brian Tracey and Nicholas Stylopoulos and Yanwei Zhang and Muna Tahir and Jing Yuan and Xuefang Wu and Lisa Graf and Thayabaran Kathiresan and Adam P. Vogel and David J. Levinthal},
journal = {npj Gut and Liver},
year = {2026},
month = nov,
doi = {10.1038/s44355-026-00086-8},
url = {https://doi.org/10.1038/s44355-026-00086-8},
note = {CC BY},
}
RIS
TY - JOUR
TI - Evaluation of an audio-based monitoring system for detecting retching and vomiting in gastroparesis: an observational cohort study
AU - Yishu Gong
AU - Jasmin Imsirovic
AU - Brian Tracey
AU - Nicholas Stylopoulos
AU - Yanwei Zhang
AU - Muna Tahir
AU - Jing Yuan
AU - Xuefang Wu
AU - Lisa Graf
AU - Thayabaran Kathiresan
AU - Adam P. Vogel
AU - David J. Levinthal
PY - 2026
DA - 2026/11/08
JO - npj Gut and Liver
DO - 10.1038/s44355-026-00086-8
UR - https://doi.org/10.1038/s44355-026-00086-8
AB - This study evaluated a novel, audio-based digital monitoring system for detecting retching and vomiting (expulsion) events in a noninterventional, longitudinal study in 25 participants with gastroparesis (GP). Participants used an audio-based system comprising an iOS device (iPhone) and Watch in their homes for 14 days. Retching and expulsion events were quantified using human-annotated counts from audio recordings and daily electronic patient-reported outcome (ePRO) counts. Device compliance was calculated as recorded h divided by the total h (24 × 14). Compliance was high: 83.4% for the iPhone and 69.4% for the Watch. Event timing analyses from the audio devices revealed elevated retching and expulsion activity in the morning, with symptoms persisting during nighttime h. Watch and iOS event counts showed strong correlation (r > 0.85), though device agreement with ePRO counts was lower (r < 0.68). While ePRO-based expulsion counts showed stronger within-participant correlation with patient global impression of frequency/severity (PGI-F/PGI-S) measures, device counts better distinguished participants across PGI-F/PGI-S levels. Weekly averaging of event counts improved consistency across both audio-based and ePRO measures relative to daily event counts. Preliminary findings suggest audio-based technologies may provide an objective approach to detecting vomiting and retching, with future automation needed to support scalable implementation.
SN - 3004-9806
ER -