2022/04/26 | Research | Artificial Intelligence

Review of dietary smartphone apps

Accurate monitoring of macronutrient and energy intake are important to manage diabetes mellitus and obesity, leading to improved glycemic control and weight reduction, respectively. So far, the potential of smartphone apps based on artificial intelligence has not been fully exploited to fulfill this monitoring task. In a review paper, the AI in Health and Nutrition lab has explored 22 image- and video-based apps designed for dietary assessment for various application fields and end-users.

Smartphone apps for dietary assessment need to fulfill user needs and expectations and have to be tested under real-world conditions to validate that they are as accurate as existing dual dietary assessment methods. (https://doi.org/10.1177/19322968221085026)

Current dietary assessment methods are time-consuming, costly, and due to manual input, prone to major errors in measurement. In addition, they do not provide real-time feedback. Smartphone apps, through advancements in AI, algorithms, and hardware are increasingly being introduced for dietary monitoring, using food or beverage multimedia data as input and generate almost instant nutrient content output.

The AI in Health and Nutrition lab at the ARTORG reviewed 22 different smartphone-based systems and divided these into three categories: laboratory (12), preclinical (7), and clinical (3). The team found a number of open research questions and technical challenges to be addressed and end users – including health care professionals and patients – need to be involved in the app design and development. Importantly, there is a clear need for these systems to be validated under unconstrained real-life conditions and for their performance to be compared against conventional methods for dietary assessment.