Smart Piezoelectric-Based Wearable System for Calorie Intake Estimation Using Machine Learning

Eating an appropriate food volume, maintaining the required calorie count, and making good nutritional choices are key factors for reducing risk of obesity, which has many consequences such as Osteoarthritis (OA) that affects patient’s knee. In this paper, we present a wearable sensor in form necklace embedded with piezoelectric sensor, detects skin movement from lower trachea while eating. contrast to previous state-of-the-art sensor-based system used spectral features, our fully exploits temporal amplitude-varying signals optimal thus classifies foods more accurately. Through evaluation frame length position swallowing frame, found best performance was 30 samples (1.5 s), located towards end frame. This demonstrates chewing sequence carries important information classification. Additionally, new approach weight solid can be estimated swallow count calculated their weight. Our based on smartphone app helps users live healthily by providing them real-time feedback about ingested types, count.