Bedri Abdelkareem, Li Richard, Haynes Malcolm, Kosaraju Raj Prateek, Grover Ishaan, Prioleau Temiloluwa, Beh Min Yan, Goel Mayank, Starner Thad, Abowd Gregory
Carnegie Mellon University, Carnegie Mellon, 5000 Forbes Avenue, Pittsburgh, PA 15213, US.
Georgia Institute of Technology, 85 5th St NW, Atlanta, GA 30308, US.
Proc ACM Interact Mob Wearable Ubiquitous Technol. 2017 Sep;1(3). doi: 10.1145/3130902.
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 F-score of 90.9% in detecting chewing instances. In the unconstrained, outside-the-lab evaluation, EarBit obtained an accuracy of 93% and an F-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.
肥胖、糖尿病和高胆固醇血症等慢性广泛性疾病要求患者监测食物摄入量，而饮食记录是目前最常用的方法。然而，饮食记录容易出现自我偏差和回忆误差，且患者的依从性较差。在本文中，我们提出了一种替代方法，即引入EarBit，这是一种可检测进食时刻的可穿戴系统。我们评估了惯性、光学和声学传感模式的性能，并凭借其识别和可用性性能重点关注惯性传感。利用在模拟家庭环境中收集的数据，对参与者的行为限制最小，我们构建了模型，并通过无约束的实验室外研究对其进行评估。对于这两项研究，我们获取了视频片段作为参与者活动的地面真值。使用留一法验证，EarBit在半控制实验室研究中识别出了所有进食事件，在检测咀嚼实例时准确率达到90.1%，F值达到90.9%。在无约束的实验室外评估中，EarBit在检测咀嚼实例时准确率达到93%，F值达到80.1%。它还准确识别出了除一次记录的进食事件外的所有事件。这些事件从2分钟的小吃到30分钟的正餐不等。
