Food intake and chewing in women

Assessment of In-Meal Eating Behaviour using Fuzzy SVM

Certain patterns of eating behaviour during meal have been identified as risk factors for long-term abnormal eating development in healthy individuals and, eventually, can affect the body weight. To detect early signs of problematic eating behaviour, this paper proposes a novel method for building behaviour assessment models. The goal of the models is to predict whether the in-meal eating behaviour resembles patterns associated with obesity, eating disorders, or low-risk behaviours. The models are trained using meals recorded with a plate scale from a reference population and labels annotated by a domain expert. In addition, the domain expert assigned scores that characterise the degree of any exhibited abnormal patterns. To improve model effectiveness, we use the domain expert's scores to create training error regularisation weights that alter the importance of each training instance for its class during model training. The behaviour assessment models are based on the SVM algorithm and the fuzzy SVM algorithm for their instance-weighted variation. Experiments conducted on meals recorded from 120 individuals show that: (a) the proposed approach can produce effective models for eating behaviour classification (for individuals), or for ranking (for populations); and (b) the instance-weighted fuzzy SVM models achieve significant performance improvements, compared to the non-weighted, standard SVM models.

The Effect of Food Unit Sizes and Meal Serving Occasions on Eating Behaviour Characteristics: Within Person Randomised Crossover Studies on Healthy Women

Manipulating food properties and serving environment during a meal can significantly change food intake at group level. However, the evaluation of the usefulness of such manipulations requires an understanding of individual behavioural changes. Three studies were conducted to explore the effect of unit size and meal occasion on eating behaviour characteristics (food intake, meal duration, number of bites and chews). All studies used a randomised crossover design, with a one-week wash-out period, starting with a familiarisation meal, with the participation of healthy, normal weight females between the ages of 18–35 years. In Study 1 (n = 19) three cube sizes (0.5, 1.0 and 1.5 cm³) of vegetable hash and chicken were compared. In Study 2 (n = 18) mashed potatoes and mincemeat were compared to whole potatoes and meatballs. In Study 3 (n = 29) meals served at lunch time (11:00–13:00) were compared to identical meals served at dinner time (17:00–19:00). The largest food unit size lead to significantly increased meal duration in Study 2 (mean difference 0.9 min, 95% confidence interval (CI) 0.0–1.8), but not in Study 1 (mean difference 1 min, 95% CI 0.1–2.0). There was a significant increase in number of chews in the large unit size condition of both Study 1 (mean difference 88, 95% CI 12–158) and Study 2 (mean difference 95, 95% CI 12–179). Different serving occasions did not significantly change any of the eating behaviours measured. Except for number of bites in Study 2 (R² = 0.60), most individuals maintained their eating behaviour relative to the group across unit sizes and serving occasions conditions (R² > 0.75), which suggests single meal testing can provide information about the behavioural characteristics of individual eating styles under different conditions.

Automatic Analysis of Food Intake and Meal Microstructure Based on Continuous Weight Measurements

The structure of the cumulative food intake (CFI) curve has been associated with obesity and eating disorders. Scales that record the weight loss of a plate from which a subject eats food are used for capturing this curve; however, their measurements are contaminated by additive noise and are distorted by certain types of artifacts. This paper presents an algorithm for automatically processing continuous in-meal weight measurements in order to extract the clean CFI curve and in-meal eating indicators, such as total food intake and food intake rate. The algorithm relies on the representation of the weight-time series by a string of symbols that correspond to events such as bites or food additions. A context-free grammar is next used to model a meal as a sequence of such events. The selection of the most likely parse tree is finally used to determine the predicted eating sequence. The algorithm is evaluated on a dataset of 113 meals collected using the Mandometer, a scale that continuously samples plate weight during eating. We evaluate the effectiveness for seven indicators and for bite-instance detection. We compare our approach with three state-of-the-art algorithms, and achieve the lowest error rates for most indicators (24 g for total meal weight). The proposed algorithm extracts the parameters of the CFI curve automatically, eliminating the need for manual data processing, and thus facilitating large-scale studies of eating behavior.

Objective quantification of the food proximity effect on grapes, chocolate and cracker consumption in a Swedish high school. A temporal analysis

Close food proximity leads to increased short-term energy intake, potentially contributing to the long-term development of obesity. However, its precise effects on eating behaviour are still unclear, especially with food available for extended periods of time. This study involved two similar high school student groups (15–17 years old), which had ad libitum access to grapes, chocolates and crackers during an hour-long experimental session. In the distal condition the foods were placed 6 meters away from the students (n = 24), in contrast to the proximal condition (n = 17) were the food was placed near the students. The identification of the type and the quantification of the amount of each food selected, for each individual serving, was facilitated through use of food scales and video recording. In the proximal condition individuals served themselves grapes and crackers more often and consumed more chocolate than in the distal condition. In total, participants in the proximal condition ingested significantly more energy (726 kcal vs. 504 kcal; p = 0.029), without reporting higher fullness. Food proximity also affected the temporal distribution of servings, with the first five minutes of the sessions corresponding to 53.1% and 45.6% of the total energy intake for the distal and proximal conditions, respectively. After the first five minutes, the servings in the distal condition were strongly clustered in time, with many students getting food together. In the proximal condition however, students displayed an unstructured pattern of servings over time. In conclusion, this study strengthens past evidence regarding the important role of food proximity on individual energy intake and, for the first time, it associates continuous food proximity to the emergence of unstructured eating over time. These conclusions, expanded upon by future studies, could support the creation of meaningful intervention strategies based on spatially and temporally controlled food availability.

A parametric Probabilistic Context-Free Grammar for food intake analysis based on continuous meal weight measurements

Monitoring and modification of eating behaviour through continuous meal weight measurements has been successfully applied in clinical practice to treat obesity and eating disorders. For this purpose, the Mandometer, a plate scale, along with video recordings of subjects during the course of single meals, has been used to assist clinicians in measuring relevant food intake parameters. In this work, we present a novel algorithm for automatically constructing a subject's food intake curve using only the Mandometer weight measurements. This eliminates the need for direct clinical observation or video recordings, thus significantly reducing the manual effort required for analysis. The proposed algorithm aims at identifying specific meal related events (e.g. bites, food additions, artifacts), by applying an adaptive pre-processing stage using Delta coefficients, followed by event detection based on a parametric Probabilistic Context-Free Grammar on the derivative of the recorded sequence. Experimental results on a dataset of 114 meals from individuals suffering from obesity or eating disorders, as well as from individuals with normal BMI, demonstrate the effectiveness of the proposed approach.