JMIR Publications

ABSTRACT

Background:

Diet-related diseases such as type 2 diabetes, which require strict dietary management to slow down disease progression, call for innovative management strategies. Conventional diet monitoring places a memory burden on patients, who may not accurately remember details regarding their meals; as a result, conventional diet monitoring frequently falls short in preventing disease advancement . Recent advances in sensor and computational technologies have sparked interest in developing eating detection platforms.

Objective:

This review investigates central hemodynamic and thermoregulatory responses as potential biomarkers for eating detection.

Methods:

We searched peer-reviewed literature indexed in PubMed, Web of Science, and Scopus on June 20, 2022, with no date limits (PROSPERO CRD42022360600). We included English articles demonstrating the impact of eating on central hemodynamics and thermoregulation in healthy individuals. To evaluate the overall study quality and assess the risk of bias, we designed a customized tool inspired by the Cochrane assessment framework. This tool has four categories: high, medium, low, and very low. Two independent reviewers conducted title and abstract screening, full-text review, and study quality and risk of bias analysis. In instances of disagreement between the two reviewers, a third reviewer served as an adjudicator.

Results:

Our search retrieved 11,450 studies, and 25 met our inclusion criteria. Among the retrieved studies, eight (32%) were classified as high quality, 13 (52%) as medium quality, and four (16%) as low quality. Furthermore, we found no evidence of publication bias in any of the included studies. A consistent postprandial elevation in heart rate, cardiac output, and stroke volume was evident in no less than 95% of the studies that investigated these variables. Finally, we observed grossly inconsistent postprandial thermoregulatory effects.

Conclusions:

Our findings demonstrate that central hemodynamic responses, particularly heart rate, hold promise for wearable-based eating detection, given that cardiac output and stroke volume cannot be measured by any presently available non-invasive medical or consumer-grade wearables.