An Innovative Machine Learning Approach to Predict the Dietary Fiber Content of Packaged Foods

Abstract

Underconsumption of dietary fiber is prevalent worldwide and is associated with multiple adverse health conditions. Despite the importance of fiber, the labeling of fiber content on packaged foods and beverages is voluntary in most countries, making it challenging for consumers and policy makers to monitor fiber consumption. Here, we developed a machine learning approach for automated and systematic prediction of fiber content using nutrient information commonly available on packaged products. An Australian packaged food dataset with known fiber content information was divided into training (n = 8986) and test datasets (n = 2455). Utilization of a k-nearest neighbors machine learning algorithm explained a greater proportion of variance in fiber content than an existing manual fiber prediction approach (R2 = 0.84 vs. R2 = 0.68). Our findings highlight the opportunity to use machine learning to efficiently predict the fiber content of packaged products on a large scale.

Socio-economic difference in purchases of ultra-processed foods in Australia: an analysis of a nationally representative household grocery purchasing panel

“…This database contains nutrition information for more than 80,000 packaged foods and beverages that have been available for sale in Australia since January 2013. Most of the data (~ 60% of all products) are captured by trained data collectors through in-store surveys at five large Australian supermarkets owned by Aldi, Coles, Harris Farm, Independent Grocers of Australia (IGA) and Woolworths in the Sydney metropolitan area [ 29 ]. Images of the pack of each food and beverage product are captured (front of pack, nutrient declaration, ingredients list, manufacturer details), using a bespoke smartphone application.…”

“…The product name, brand name, package size (g) and nutrient content per 100 g/mL and per serve are then extracted [ 23 , 30 ]. The database also contains data that are (i) crowdsourced using the FoodSwitch smartphone application (~ 30%) and (ii) provided directly by the food industry (~ 10%) [ 29 ].…”

The fourth industrial revolution in the food industry—part II: Emerging food trends

“…Machine learning is the core of AI and data science (Jordan & Mitchell, 2015) and has found its way into various food-related applications, including functional foods and fortification. Machine learning allows a computer system to develop an algorithm that can map input information, such as details about packaged foods and beverages, and to predict a specified output (e.g., fiber content) based on commonly available nutrient information (Davies et al, 2021). The integration of AI into the discovery and development of functional food ingredients can lead to a safer and more sustainable food chain achieving safe and costeffective solutions for improved human and animal health (Doherty et al, 2021).…”

Predicting nutrient profiles in food after processing

“…USDA investigators predicted the content of 3 label nutrients (carbohydrates, protein and sodium) in processed foods from the ingredient list, using the Branded Foods datatype in Food Data Central (FDC) [20]. Several projects predicted nutrient contents from the composition data; nutrient content was predicted for the missing values in food composition data [21], lactose content was predicted in dietary recall database [22], fiber content was predicted for commercially processed foods [23].…”

“…USDA investigators predicted the content of 3 label nutrients (carbohydrates, protein and sodium) in processed foods from the ingredient list, using the Branded Foods datatype in Food Data Central (FDC) [20]. Several projects predicted nutrient contents from the composition data; nutrient content was predicted for the missing values in food composition data [21], lactose content was predicted in dietary recall database [22], fiber content was predicted for commercially processed foods [23]. Availability of datasets with high quality data for training and testing is essential, and databases such as BitterDB [24], BTP640 [25], FlavorDB [26],FooDB [27], SuperSweet [28], Fenaroli [29],GoodScents [30], FDC [31] as well as specifically curated datasets of hyperspectral images may contribute to this end.…”