Assessment of an AI tool to improve dietary intake
Research type
Research Study
Full title
Proof of Principle Evaluation of the AI-derived CoDiet Tool- A Randomised Controlled Trial
IRAS ID
353857
Contact name
Gary Frost
Contact email
Duration of Study in the UK
1 years, 5 months, 4 days
Research summary
Increasing evidence suggests that there is a link between dietary intake and risk for non-communicable diseases, such as obesity and heart disease. To be able to reduce the burden of these conditions, it is necessary to provide personalised dietary advice to individuals based on their current dietary intake, and on various general health parameters. The aim of this study is to improve dietary intake in study participants through a randomised controlled study. One arm is a personalised dietary advice (PDA) using an artificial intelligence (AI) modelling system based on findings from a previous study (REC Ref: 23/PR/1109) in our research group, the CoDiet study (Combatting Diet-related Non-communicable Diseases through Enhanced Surveillance) while the other arm is a General Dietary Advice (GDA) based on WHO (World Health Organisation) dietary guidelines. Eligible volunteers (aged 18-65, body mass index >25 kg/m2 and waist circumference >102 cm in males and 88 cm in females) will wear a camera attached to glasses frames for one week which will record the food and drinks they consume, a physical activity monitor and provide urine, stool and blood samples and provide body measurements during one study visit at the NIHR Imperial Clinical Research Facility. This information will feed into the AI modelling system which will provide dietary guidance to the volunteers on the PDA arm. The volunteers on both arms will receive regular prompts on how to improve their dietary habits. This study will last 12 weeks during which the volunteers will wear the camera for three one-week periods and attend three study visits during which their dietary intake and its impact on their general health will be assessed. The impact on dietary intake and health outcomes will be compared between the two intervention arms.
REC name
London - Westminster Research Ethics Committee
REC reference
25/LO/0626
Date of REC Opinion
18 Sep 2025
REC opinion
Further Information Favourable Opinion