What is it about?
We collected full days of wrist motion data using smart watches from 351 participants exhibiting everyday life. The result of this data collection is a dataset 10x-100x larger than all previous datasets in the field of automated dietary monitoring (ADM). This data was then processed to create a technology that can detect when the wearer is eating food. Our technology may assist people with weight loss or weight gain strategies.
Featured Image
Photo by Helena Lopes on Unsplash
Why is it important?
Researchers can use our datasets to create improved technologies to detect eating in humans. Nutritionists and clinicians can use our detection technology to collect improved data from their patients. This allows for improved data collection compared to traditional food logging. The technology can help provide users with tools for controlling weight loss and serve as a method for intervention using ecological momentary assessments (EMAs).
Perspectives
The large dataset we have collected has given us new perspectives and shown us new challenges in the field that we would not have come across if we had a smaller dataset. We hope the lessons we learned from this massive data collection are useful to other researchers.
Surya Sharma
Clemson University
Read the Original
This page is a summary of: The Impact of Walking and Resting on Wrist Motion for Automated Detection of Meals, ACM Transactions on Computing for Healthcare, December 2020, ACM (Association for Computing Machinery),
DOI: 10.1145/3407623.
You can read the full text:
Resources
Contributors
The following have contributed to this page
