Evaluation of a Smart Fork to Decelerate Eating Rate

Capturing Eating Behavior from Video Analysis: A Systematic Review

Current methods to detect eating behavior events (i.e., bites, chews, and swallows) lack objective measurements, standard procedures, and automation. The video recordings of eating episodes provide a non-invasive and scalable source for automation. Here, we reviewed the current methods to automatically detect eating behavior events from video recordings. According to PRISMA guidelines, publications from 2010-2021 in PubMed, Scopus, ScienceDirect, and Google Scholar were screened through title and abstract, leading to the identification of 277 publications. We screened the full text of 52 publications and included 13 for analysis. We classified the methods in five distinct categories based on their similarities and analyzed their accuracy. Facial landmarks can count bites, chews, and food liking automatically (accuracy: 90%, 60%, 25%). Deep neural networks can detect bites and gesture intake (accuracy: 91%, 86%). The active appearance model can detect chewing (accuracy: 93%), and optical flow can count chews (accuracy: 88%). Video fluoroscopy can track swallows but is currently not suitable beyond clinical settings. The optimal method for automated counts of bites and chews is facial landmarks, although further improvements are required. Future methods should accurately predict bites, chews, and swallows using inexpensive hardware and limited computational capacity. Automatic eating behavior analysis will allow the study of eating behavior and real-time interventions to promote healthy eating behaviors.

Current Perspectives on Food Oral Processing

Food oral processing (FOP) is a fast-emerging research area in the food science discipline. Since its first introduction about a decade ago, a large amount of literature has been published in this area, forming new frontiers and leading to new research opportunities. This review aims to summarize FOP research progress from current perspectives. Food texture, food flavor (aroma and taste), bolus swallowing, and eating behavior are covered in this review. The discussion of each topic is organized into three parts: a short background introduction, reflections on current research findings and achievements, and future directions and implications on food design. Physical, physiological, and psychological principles are the main concerns of discussion for each topic. The last part of the review shares views on the research challenges and outlooks of future FOP research. It is hoped that the review not only helps readers comprehend what has been achieved in the past decade but also, more importantly, identify where the knowledge gaps are and in which direction the FOP research will go.

Mobile Health Apps in Pediatric Obesity Treatment: Process Outcomes From a Feasibility Study of a Multicomponent Intervention

BACKGROUND

Multicomponent family interventions underline current best practice in childhood obesity treatment. Mobile health (mHealth) adjuncts that address eating and physical activity behaviors have shown promise in clinical studies.

OBJECTIVE

This study aimed to describe process methods for applying an mHealth intervention to reduce the rate of eating and monitor physical activity among children with obesity.

METHODS

The study protocol was designed to incorporate 2 mHealth apps as an adjunct to usual care treatment for obesity. Children and adolescents (aged 9-16 years) with obesity (BMI ≥98th centile) were recruited in person from a weight management service at a tertiary health care center in the Republic of Ireland. Eligible participants and their parents received information leaflets, and informed consent and assent were signed. Participants completed 2 weeks of baseline testing, including behavioral and quality of life questionnaires, anthropometry, rate of eating by Mandolean, and physical activity level using a smart watch and the myBigO smartphone app. Thereafter, participants were randomized to the (1) intervention (usual clinical care+Mandolean training to reduce the rate of eating) or (2) control (usual clinical care) groups. Gender and age group (9.0-12.9 years and 13.0-16.9 years) stratifications were applied. At the end of a 4-week treatment period, participants repeated the 2-week testing period. Process evaluation measures included recruitment, study retention, fidelity parameters, acceptability, and user satisfaction.

RESULTS

A total of 20 participants were enrolled in the study. A web-based randomization system assigned 8 participants to the intervention group and 12 participants to the control group. Attrition rates were higher among the participants in the intervention group (5/8, 63%) than those in the control group (3/12, 25%). Intervention participants undertook a median of 1.0 training meal using Mandolean (25th centile 0, 75th centile 9.3), which represented 19.2% of planned intervention exposure. Only 50% (9/18) of participants with smart watches logged physical activity data. Significant differences in psychosocial profile were observed at baseline between the groups. The Child Behavior Checklist (CBCL) mean total score was 71.7 (SD 3.1) in the intervention group vs 57.6 (SD 6.6) in the control group, t-test P<.001, and also different among those who completed the planned protocol compared with those who withdrew early (CBCL mean total score 59.0, SD 9.3, vs 67.9, SD 5.6, respectively; t-test P=.04).

CONCLUSIONS

A high early attrition rate was a key barrier to full study implementation. Perceived task burden in combination with behavioral issues may have contributed to attrition. Low exposure to the experimental intervention was explained by poor acceptability of Mandolean as a home-based tool for treatment. Self-monitoring using myBigO and the smartwatch was acceptable among this cohort. Further technical and usability studies are needed to improve adherence in our patient group in the tertiary setting.

How Important Is Eating Rate in the Physiological Response to Food Intake, Control of Body Weight, and Glycemia?

The link between eating rate and energy intake has long been a matter of extensive research. A better understanding of the effect of food intake speed on body weight and glycemia in the long term could serve as a means to prevent weight gain and/or dysglycemia. Whether a fast eating rate plays an important role in increased energy intake and body weight depends on various factors related to the studied food such as texture, viscosity and taste, but seems to be also influenced by the habitual characteristics of the studied subjects as well. Hunger and satiety quantified via test meals in acute experiments with subsequent energy intake measurements and their association with anorexigenic and orexigenic regulating peptides provide further insight to the complicated pathogenesis of obesity. The present review examines data from the abundant literature on the subject of eating rate, and highlights the main findings in people with normal weight, obesity, and type 2 diabetes, with the aim of clarifying the association between rate of food intake and hunger, satiety, glycemia, and energy intake in the short and long term.

Sensory Interactive Table (SIT)-Development of a Measurement Instrument to Support Healthy Eating in a Social Dining Setting

This paper presents the Sensory Interactive Table (SIT): an instrumented, interactive dining table. Through the use of load cells and LEDs that are embedded in the table surface, SIT allows us to study: (1) the eating behaviors of people in a social setting, (2) the social interactions around the eating behaviors of people in a social setting, and (3) the continuous cycle of feedback through LEDs on people’s eating behavior and their response to this feedback in real time, to ultimately create an effective dietary support system. This paper presents the hard- and software specifications of the system, and it shows the potential of the system to capture mass-related dimensions in real time and with high accuracy and spatial resolution.

Computational Commensality: From Theories to Computational Models for Social Food Preparation and Consumption in HCI

Food and eating are inherently social activities taking place, for example, around the dining table at home, in restaurants, or in public spaces. Enjoying eating with others, often referred to as “commensality,” positively affects mealtime in terms of, among other factors, food intake, food choice, and food satisfaction. In this paper we discuss the concept of “Computational Commensality,” that is, technology which computationally addresses various social aspects of food and eating. In the past few years, Human-Computer Interaction started to address how interactive technologies can improve mealtimes. However, the main focus has been made so far on improving the individual's experience, rather than considering the inherently social nature of food consumption. In this survey, we first present research from the field of social psychology on the social relevance of Food- and Eating-related Activities (F&EA). Then, we review existing computational models and technologies that can contribute, in the near future, to achieving Computational Commensality. We also discuss the related research challenges and indicate future applications of such new technology that can potentially improve F&EA from the commensality perspective.

Effects of eating with an augmented fork with vibrotactile feedback on eating rate and body weight: a randomized controlled trial

Background

Eating rate is a basic determinant of appetite regulation: people who eat more slowly feel sated earlier and eat less. A high eating rate contributes to overeating and potentially to weight gain. Previous studies showed that an augmented fork that delivers real-time feedback on eating rate is a potentially effective intervention to decrease eating rate in naturalistic settings. This study assessed the impact of using the augmented fork during a 15-week period on eating rate and body weight.

Methods

In a parallel randomized controlled trial, 141 participants with overweight (age: 49.2 ± 12.3 y; BMI: 31.5 ± 4.48 kg/m2) were randomized to intervention groups (VFC, n = 51 or VFC+, n = 44) or control group (NFC, n = 46). First, we measured bite rate and success ratio on five consecutive days with the augmented fork without feedback (T1). The intervention groups (VFC, VFC+) then used the same fork, but now received vibrotactile feedback when they ate more than one bite per 10 s. Participants in VFC+ had additional access to a web portal with visual feedback. In the control group (NFC), participants ate with the fork without either feedback. The intervention period lasted four weeks, followed by a week of measurements only (T2) and another measurement week after eight weeks (T3). Body weight was assessed at T1, T2, and T3.

Results

Participants in VFC and VFC+ had a lower bite rate (p < .01) and higher success ratio (p < .0001) than those in NFC at T2. This effect persisted at T3. In both intervention groups participants lost more weight than those in the control group at T2 (p < .02), with no rebound at T3.

Conclusions

The findings of this study indicate that an augmented fork with vibrotactile feedback is a viable tool to reduce eating rate in naturalistic settings. Further investigation may confirm that the augmented fork could support long-term weight loss strategies.

Trial registration

The research reported in this manuscript was registered on 4 November 2015 in the Netherlands Trial Register with number NL5432 (https://www.trialregister.nl/trial/5432).

Digital Commensality: Eating and Drinking in the Company of Technology

Commensality is a key aspect of social dining. However, previous research has identified a number of pros and cons associated with the incorporation of digital technology into eating and drinking episodes. For instance, those who are distracted by digital technology may eat/drink more (that is, they may overconsume) as a result of their failure to attend to the food-related sensations that are thought to cue the termination of eating. Similarly, it has often been suggested that the use of mobile devices at mealtimes can disrupt the more commensal aspects of dining/drinking (at least among those who are physically present together). At the same time, however, looking to the future, it seems clear that digital technologies also hold the promise of delivering opportunities for enhanced multisensory experiential dining. For instance, they might be used to match the auditory, visual, or audiovisual entertainment to the eating/drinking episode (e.g., think only about watching a Bollywood movie while eating a home-delivery Indian meal, say). Indeed, given the growing societal problems associated with people dining by themselves, there are a number of routes by which digital technologies may increasingly help to connect the solo diner with physically co-located, remote, or even virtual dining partners. In this review of the literature, our focus is specifically on the role of technology in inhibiting/facilitating the more pleasurable social aspects of dining, what one might call "digital commensality." The focus is primarily on Westernized adults with reasonable access to, and familiarity with, digital technologies.

Food Intake during School Lunch Is Better Explained by Objectively Measured Eating Behaviors than by Subjectively Rated Food Taste and Fullness: A Cross-Sectional Study

School lunches contribute significantly to students’ food intake (FI) and are important to their long-term health. Objective quantification of FI is needed in this context. The primary aim of this study was to investigate how much eating rate (g/min), number of food additions, number of spoonfuls, change in fullness, food taste, body mass index (BMI), and sex explain variations in school lunch FI. The secondary aim was to assess the reliability of repeated FI measures. One hundred and three (60 females) students (15–18 years old) were monitored while eating lunch in their normal school canteen environment, following their usual school schedules. A subgroup of students (n = 50) participated in a repeated lunch (~3 months later). Linear regression was used to explain variations in FI. The reliability of repeated FI measurements was assessed by change in mean, coefficient of variation (CV), and intraclass correlation (ICC). The regression model was significant and explained 76.6% of the variation in FI. Eating rate was the strongest explanatory variable, followed by spoonfuls, sex, food additions, food taste, BMI, and change in fullness. All explanatory variables were significant in the model except BMI and change in fullness. No systematic bias was observed in FI (−7.5 g (95% CI = −43.1–28 g)) while individual students changed their FI from −417 to +349 g in the repeated meal (CV 26.1% (95% CI = 21.4–33.5%), ICC 0.74 (95% CI = 0.58–0.84)). The results highlight the importance of objective eating behaviors for explaining FI in a school lunch setting. Furthermore, our methods show promise for large-scale quantification of objectively measured FI and eating behaviors in schools.

Self-reported eating rate is associated with weight status in a Dutch population: a validation study and a cross-sectional study

BACKGROUND

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.

METHODS

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.

RESULTS

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.

CONCLUSIONS

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.

Fast or slow-foods? Describing natural variations in oral processing characteristics across a wide range of Asian foods

The structural properties of foods have a functional role to play in oral processing behaviours and sensory perception, and also impact on the meal size and the experience of fullness.