Chronic and widespread diseases such as obesity, diabetes, and hypercholesterolemia require patients to monitor their food intake, and food journaling is currently the most common method for doing so. However, food journaling is subject to self-bias and recall errors, and is poorly adhered to by patients. In this paper, we propose an alternative by introducing EarBit, a wearable system that detects eating moments. We evaluate the performance of inertial, optical, and acoustic sensing modalities and focus on inertial sensing, by virtue of its recognition and usability performance. Using data collected in a simulated home setting with minimum restrictions on participants’ behavior, we build our models and evaluate them with an unconstrained outside-the-lab study. For both studies, we obtained video footage as ground truth for participants activities. Using leave-one-user-out validation, EarBit recognized all the eating episodes in the semi-controlled lab study, and achieved an accuracy of 90.1% and an F1-score of 90.9% in detecting chewing instances. In the unconstrained, outside-the-lab evaluation, EarBit obtained an accuracy of 93% and an F1-score of 80.1% in detecting chewing instances. It also accurately recognized all but one recorded eating episodes. These episodes ranged from a 2 minute snack to a 30 minute meal.

No abstract

Some small field-of-view (FOV) head worn displays (HWD), like Epson's Moverio BT-300, are mounted directly in the user's line of sight. In contrast, Google Glass is mounted “out of the way” and above the line of sight. Other displays like the Vuzix M100 or Optinvent ORA-1 allow the user to adjust the display position, and some users have expressed a desire for the display to be laterally displaced toward the ear, out of the main line of sight. How far toward the ear can a small FOV display be mounted and still be used comfortably? Using a 30-minute reading task and an emulated display with the FOV of a typical smart phone (9.2°x 16.3°), we study a user's perceived comfort level while reading at four horizontally displaced positions. We ask participants to rate their comfort every five minutes using a 5-point Likert scale knob (5 being most comfortable), for a total of seven measurements. Scores are summed over the seven measurements to form a summed comfort score. We find that 0° (Md = 34.0; p«0.001), 10° (Md = 33.5; p«c0.001), and 20° (Md = 33.5; p«c0.001) are more comfortable than 30° (Md = 29.5) and that 0° (p<0.01) and 10° (p<0.01) are more comfortable than 20°. Reading performance and workload measures were numerically similar across all conditions. Given the main results of the experiment, post-hoc analysis on other measurements such as preference and asthenopia, and participant comments, we suggest that small FOV displays should be mounted at lateral displacement angles of 20° and less for sustained use.

No abstract