TY - JOUR

T1 - Predicting Food Intake From Food Reward and Biometric Responses to Food Cues in Adults With Normal Weight Using Machine Learning

AU - Pedersen, Hanne

AU - Diaz, Lars J

AU - Clemmensen, Kim K. B.

AU - Jensen, Marie M

AU - Jørgensen, Marit E.

AU - Finlayson, Graham

AU - Quist, Jonas S

AU - Vistisen, Dorte

AU - Færch, Kristine

N1 - © The Author(s) 2022. Published by Oxford University Press on behalf of the American Society for Nutrition.

PY - 2022

Y1 - 2022

N2 - BACKGROUND: Eating behaviours are determined by a complex interplay between behavioural and physiological signalling occurring prior to, during and after eating.OBJECTIVES: The aim was to explore how selected behavioural and physiological variables separately and grouped together predicted intake of eight different foods.METHODS: One hundred adults with normal weight performed a food preference task combined with biometric measurements (the Steno Biometric Food Preference Task) in the fasting state. The task measured food reward as well as biometric (eye tracking, electrodermal activity, and facial expressions) response to images of foods varying in fat content and taste. Energy intake of the same eight foods as assessed in the preference task was subsequently assessed from an ad libitum buffet. A mixed effects random forest approach was applied to explore how individual and combined measures of food reward and biometric responses predicted energy intake of the eight single foods. The performance of the different prediction models was compared to the predictions from a linear model including only an intercept (naïve model) using bootstrap cross-validation.RESULTS: Participants had a median [IQR] intake of 369 [126; 472] kJ per food. Combined or separate measures of food reward or biometric responses did not predict energy intake better than the naïve model.CONCLUSION: We did not find that the reward or biometric responses to food cues assessed in a clinical setting were useful in predicting energy intake of single foods. However, this study provides a framework in the field of behavioural nutrition for applying machine learning with focus on individual predictions. This is necessary on the road towards personalized nutrition and provides great potential for handling complex data with multiple variables. Trial registration: The study is registered at ClinicalTrials.gov (Identifier: NCT03986619).

AB - BACKGROUND: Eating behaviours are determined by a complex interplay between behavioural and physiological signalling occurring prior to, during and after eating.OBJECTIVES: The aim was to explore how selected behavioural and physiological variables separately and grouped together predicted intake of eight different foods.METHODS: One hundred adults with normal weight performed a food preference task combined with biometric measurements (the Steno Biometric Food Preference Task) in the fasting state. The task measured food reward as well as biometric (eye tracking, electrodermal activity, and facial expressions) response to images of foods varying in fat content and taste. Energy intake of the same eight foods as assessed in the preference task was subsequently assessed from an ad libitum buffet. A mixed effects random forest approach was applied to explore how individual and combined measures of food reward and biometric responses predicted energy intake of the eight single foods. The performance of the different prediction models was compared to the predictions from a linear model including only an intercept (naïve model) using bootstrap cross-validation.RESULTS: Participants had a median [IQR] intake of 369 [126; 472] kJ per food. Combined or separate measures of food reward or biometric responses did not predict energy intake better than the naïve model.CONCLUSION: We did not find that the reward or biometric responses to food cues assessed in a clinical setting were useful in predicting energy intake of single foods. However, this study provides a framework in the field of behavioural nutrition for applying machine learning with focus on individual predictions. This is necessary on the road towards personalized nutrition and provides great potential for handling complex data with multiple variables. Trial registration: The study is registered at ClinicalTrials.gov (Identifier: NCT03986619).

U2 - 10.1093/jn/nxac053

DO - 10.1093/jn/nxac053

M3 - Journal article

C2 - 35325189

VL - 152

SP - 1574

EP - 1581

JO - Journal of Nutrition

JF - Journal of Nutrition

SN - 0022-3166

IS - 6

ER -