In the context of dietary management, accurate monitoring of eating habits is receiving increased attention. Wearable sensors, combined with the connectivity and processing of modern smartphones, can be used to robustly extract objective and real-time measurements of human behavior. In particular, for the task of chewing detection, several approaches based on an in-ear microphone can be found in the literature, while other types of sensors have also been reported, such as strain sensors. In this paper, performed in the context of the SPLENDID project, we propose to combine an in-ear microphone with a photoplethysmography (PPG) sensor placed in the ear concha, in a new high accuracy and low sampling rate prototype chewing detection system. We propose a pipeline that initially processes each sensor signal separately, and then fuses both to perform the final detection. Features are extracted from each modality, and support vector machine (SVM) classifiers are used separately to perform snacking detection. Finally, we combine the SVM scores from both signals in a late-fusion scheme, which leads to increased eating detection accuracy. We evaluate the proposed eating monitoring system on a challenging, semifree living dataset of 14 subjects, which includes more than 60 h of audio and PPG signal recordings. Results show that fusing the audio and PPG signals significantly improves the effectiveness of eating event detection, achieving accuracy up to 0.938 and class-weighted accuracy up to 0.892.

Observational studies performed in Asian populations suggest that eating rate is related to BMI. This paper investigates the association between self-reported eating rate (SRER) and body mass index (BMI) in a Dutch population, after having validated SRER against actual eating rate. Two studies were performed; a validation and a cross-sectional study. In the validation study SRER (i.e., ‘slow’, ‘average’, or ‘fast’) was obtained from 57 participants (men/women = 16/41, age: mean ± SD = 22.6 ± 2.8 yrs., BMI: mean ± SD = 22.1 ± 2.8 kg/m2) and in these participants actual eating rate was measured for three food products. Using analysis of variance the association between SRER and actual eating rate was studied. The association between SRER and BMI was investigated in cross-sectional data from the NQplus cohort (i.e., 1473 Dutch adults; men/women = 741/732, age: mean ± SD = 54.6 ± 11.7 yrs., BMI: mean ± SD = 25.9 ± 4.0 kg/m2) using (multiple) linear regression analysis. In the validation study actual eating rate increased proportionally with SRER (for all three food products P < 0.01). In the cross-sectional study SRER was positively associated with BMI in both men and women (P = 0.03 and P < 0.001, respectively). Self-reported fast-eating women had a 1.13 kg/m2 (95% CI 0.43, 1.84) higher BMI compared to average-speed-eating women, after adjusting for confounders. This was not the case in men; self-reported fast-eating men had a 0.29 kg/m2 (95% CI -0.22, 0.80) higher BMI compared to average-speed-eating men, after adjusting for confounders. These studies show that self-reported eating rate reflects actual eating rate on a group-level, and that a high self-reported eating rate is associated with a higher BMI in this Dutch population.

Choosing foods that require more time to consume and have a low energy density might constitute an effective strategy to control energy intake, because of their satiating capacity. The current study assessed the eating rate of Dutch food, and investigated the associations between eating rate and other food properties. We also explored the opportunities for a diet with a low energy intake rate (kJ/min). Laboratory data on the eating rate of 240 foods—representing the whole Dutch diet—was obtained. The results show a wide variation in both eating rate (from 2 g/min for rice waffle to 641 g/min for apple juice) and energy intake rate (from 0 kJ/min (0 kcal/min) for water to 1766 kJ/min (422 kcal/min) for chocolate milk). Eating rate was lower when foods were more solid. Moreover, eating rate was positively associated with water content and inversely with energy density. Energy intake rate differed substantially between and within food groups, demonstrating that the available foods provide opportunities for selecting alternatives with a lower energy intake rate. These findings offer guidance when selecting foods to reduce energy intake.

Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: The available methods for monitoring food intake—which for a great part rely on self-report—often provide biased and incomplete data. Currently, no good technological solutions are available. Hence, the SPLENDID eating detection sensor (i.e., an ear-worn device) was developed to enable complete and objective measurements of eating events. The technical performance of this device has been described before. To date, literature is lacking a description of how such a device is perceived and experienced by potential users. Objective: To describe how potential users perceived and experienced the SPLENDID eating detection sensor during the different stages of its development. Methods: Potential users evaluated the eating detection sensor at different stages of its development: 1) At the start twelve health professionals (e.g., dieticians or personal trainers) were interviewed and a focus group was held with five potential end-users (m/w = 0/5, age: 22±2 yrs., BMI: 22.5±1.9 kg/m2) to find out about their thoughts on the concept of the eating detection sensor. 2) The first prototypes of the eating detection sensor were tested in a laboratory setting where 23 participants (m/w = 13/10, age: 23±3 yrs., BMI: 22.6±3 kg/m2) reported their experiences. 3) The first wearable version of the eating detection sensor was tested in a semi-controlled study where 22 young, overweight adults (m/w = 3/19, age: 23±2 yrs., BMI: 28.0±2.3 kg/m) used the sensor on two days and reported their experiences. 4) The final version of the sensor was tested in a 4-week feasibility study by 20 young, overweight adults (m/w = 4/16, age: 25±2 yrs., BMI: 28.8±2.8 kg/m2) who reported their experiences. Results: Most individuals were enthusiastic about the concept across the different stages of development. It, however, was stressed multiple times that it is critical that the device is discrete and comfortable to wear for a longer period of time. The experiences of the potential users with the eating detection sensor show that some participants would like the sensor to be less noticeable, but that the device needs to be further improved on wearer comfort especially. In final study the eating detection sensor received a grade of 3.7 for wearer comfort, on a scale of 1 to 10. Moreover, experienced discomfort was the main reason for wearing the eating detection sensor less than two hours a day. The participants reported to have used the eating detection sensor on, on average, 19 out of 28 days. Conclusions: The SPLENDID eating detection sensor is a promising new device that can facilitate the collection of reliable food intake data, as shown by its technical potential, which has been described before. Potential users are enthusiastic, but in order to be successful the wearer comfort and discreteness of the device need to be improved. Clinical Trial: n.a.

Monitoring of human eating behaviour has been attracting interest over the last few years, as a means to a healthy lifestyle, but also due to its association with serious health conditions, such as eating disorders and obesity. Use of self-reports and other non-automated means of monitoring have been found to be unreliable, compared to the use of wearable sensors. Various modalities have been reported, such as acoustic signal from ear-worn microphones, or signal from wearable strain sensors. In this work, we introduce a new sensor for the task of chewing detection, based on a novel photoplethysmography (PPG) sensor placed on the outer earlobe to perform the task. We also present a processing pipeline that includes two chewing detection algorithms from literature and one new algorithm, to process the captured PPG signal, and present their effectiveness. Experiments are performed on an annotated dataset recorded from 21 individuals, including more than 10 hours of eating and non-eating activities. Results show that the PPG sensor can be successfully used to support dietary monitoring.