A comparison of piezoelectric-based inertial sensing and audio-based detection of swallows

Abstract

Prior research has shown a correlation between poor dietary habits and countless negative health outcomes such as heart disease, diabetes, and certain cancers. Automatic monitoring of food intake in an unobtrusive, wearable form-factor can encourage healthy dietary choices by enabling individuals to regulate their eating habits. This paper presents an objective comparison of two methods for digital dietary intake monitoring: piezoelectric swallow sensing by means of a smart necklace which monitors vibrations in the neck, and audio-based detection using a throat microphone. Data was collected from twenty subjects with ages ranging from 22 to 40 as they consumed a variety of foods using both devices. In experiment 1, we distinguished sandwich, chips, and water. In experiment 2, we distinguished nuts, chocolate, and a meat patty. F-Measures for the audio based approach were 91.3% and 88.5%, respectively. In the piezo-based approach, F-measures were significantly lower, at 75.3% and 79.4%, , respectively, though computational overhead and power dissipation were significantly less due to lower sampling rates associated with the piezoelectric approach.

Food Intake Detection and Classification Using a Necklace-Type Piezoelectric Wearable Sensor System

“…A single sensor was employed in our proposed approach to detect food intake and classify food categories by capturing the neck skin movement during chewing and swallow events rather than two sensors [15], [22] and three sensors [14]. Some studies [6], [13], [19], [23], [41] used a single sensor for food intake detection and food recognition. However, their food categories were small for food recognition [6], [13], [19], [23], and food intake detection accuracy of [41] was lower than the food intake detection accuracy of our system.…”

“…There are various methods of measuring and recording daily dietary information. Different researchers have developed automated non-invasive food monitoring methods using different sensors such as accelerometers [29], microphones [4], [5], [7]- [9], [30], [31], cameras [35], gyroscopes [36]- [39], proximity sensor [17], textile pressure sensors [21], strain gauges [40], piezoelectric sensors [6], [13], [15], [16], [19], [22], [23], [26], [41], orientation sensors [10], [42], [43], electromyography [18], electroglottography [44], and wrist band [45].…”

“…The performance of piezoelectric and microphone sensors for swallowing detection was compared in [6], and used separately.…”

“…Moreover, they attached the sensor with temporalis muscle with tape which might cause discomfort and lower the accuracy in long-term usage if the sensor contact gets loose. Kalantarian et al presented an objective comparison of a piezoelectric sensor and a microphone for automated dietary intake monitoring systems [6]. They were able to classify food more accurately with microphone than with piezoelectric sensor.…”

Wearable Food Intake Monitoring Technologies: A Comprehensive Review

“…Stumbo [15] provides a detailed review of visual (camera-based) approaches of food intake monitoring. Kalantarian et al [16] honed in on piezoelectric-and audio-based approaches for food intake monitoring. However, these surveys fall short of discussing the trade-offs of each sensing approach needed to make an informed decision, including sensing algorithms and comfort assessment.…”