Small-scale wireless sensors for automated dietary monitoring




Chun, Keum San

Journal Title

Journal ISSN

Volume Title



With advances in wearable and mobile computing technology, a significant body of work has been dedicated to the field of automated dietary monitoring (ADM). The ability to detect eating activities in naturalistic environments steadily improved over the past decade. Nonetheless, mainstream adoption of the technologies has been hampered by obtrusive form factors (e.g., neck band, headphones, chest band) and high false positive rates in naturalistic settings. This dissertation addresses such problems through small-scale (<15cm²) sensing systems. Specifically, I demonstrate that the small-scale sensors can realize simple and effective ADM systems through on-body localized and targeted remote sensing in minimally obtrusive form factors. The work discussed in the present dissertation encompasses three form factors (necklace, adhesive patch, and mouthpiece) and five sensing modalities (proximity, temperature, acceleration, humidity, and gas sensing) across five studies I conducted with a total of 90 human subjects. In this dissertation, I show small-scale wireless sensors are capable of monitoring dietary activities in practical manner through localized and targeted sensing. The small-scale wireless sensors investigated in this dissertation include: FlashBite, IntoXense, Sticki v1, and Sticki v2. In the discussion of FlashBite and IntoXense that take necklace form factor, I demonstrate that mastication and alcohol intake activities can be inferred using targeted remote sensing. And in subsequent discussion of ADM approaches with Sticki v1 and Sticki v2, I demonstrate that localized sensing in adhesive patch form factor can effectively balance the practicality and eating detection performance. Furthermore, I show that localized sensing in the intraoral space with Sticki v2 embedded into a mouthpiece can provide useful information for inferring food-related information.


LCSH Subject Headings