Food Composition Databases: Does It Matter to Human Health?

Integrating Vision-Language Models for Accelerated High-Throughput Nutrition Screening

Addressing the critical need for swift and precise nutritional profiling in healthcare and in food industry, this study pioneers the integration of vision-language models (VLMs) with chemical analysis techniques. A cutting-edge VLM is unveiled, utilizing the expansive UMDFood-90k database, to significantly improve the speed and accuracy of nutrient estimation processes. Demonstrating a macro-AUCROC of 0.921 for lipid quantification, the model exhibits less than 10% variance compared to traditional chemical analyses for over 82% of the analyzed food items. This innovative approach not only accelerates nutritional screening by 36.9% when tested amongst students but also sets a new benchmark in the precision of nutritional data compilation. This research marks a substantial leap forward in food science, employing a blend of advanced computational models and chemical validation to offer a rapid, high-throughput solution for nutritional analysis.

AI nutrition recommendation using a deep generative model and ChatGPT

In recent years, major advances in artificial intelligence (AI) have led to the development of powerful AI systems for use in the field of nutrition in order to enhance personalized dietary recommendations and improve overall health and well-being. However, the lack of guidelines from nutritional experts has raised questions on the accuracy and trustworthiness of the nutritional advice provided by such AI systems. This paper aims to address this issue by introducing a novel AI-based nutrition recommendation method that leverages the speed and explainability of a deep generative network and the use of novel sophisticated loss functions to align the network with established nutritional guidelines. The use of a variational autoencoder to robustly model the anthropometric measurements and medical condition of users in a descriptive latent space, as well as the use of an optimizer to adjust meal quantities based on users' energy requirements enable the proposed method to generate highly accurate, nutritious and personalized weekly meal plans. Coupled with the ability of ChatGPT to provide an unparalleled pool of meals from various cuisines, the proposed method can achieve increased meal variety, accuracy and generalization capabilities. Extensive experiments on 3000 virtual user profiles and 84000 daily meal plans, as well as 1000 real profiles and 7000 daily meal plans, demonstrate the exceptional accuracy of the proposed diet recommendation method in generating weekly meal plans that are appropriate for the users in terms of energy intake and nutritional requirements, as well as the easiness with which it can be integrated into future diet recommendation systems.

FooDOxS: a database of oxidized sterols content in foods

Dietary oxidized sterols (DOxS) are cholesterol-like molecules known to exert pro-inflammatory, pro-oxidant, and pro-apoptotic effects, among others. We present the FooDOxS database, a comprehensive compilation of DOxS content in over 1680 food items from 120 publications across 25 countries, augmented by data generated by our group. This database reports DOxS content in foods classified under the NOVA and What We Eat in America (WWEIA) systems, allowing a comprehensive and statistically robust summary of DOxS content in foods. Notably, we evaluated the efficacy of using NOVA and WWEIA classifications in capturing DOxS variations across food categories. Our findings provide insights into the strengths and limitations of these classification systems, enhancing their utility for assessing dietary components. This research contributes to the understanding of DOxS in food processing and suggests refinements for classification systems, holding promise for improved food safety and public health assessments.

The effects of peppermint (Mentha piperita L.) and chicory (Cichorium intybus L.) in comparison with a prebiotic on productive performance, blood constituents, immunity and intestinal microflora in broiler chickens

A total of 320 one-day-old broiler chickens were used in a 42-day feeding trial to evaluate the effects of peppermint (Mentha piperita L.) and chicory (Cichorium intybus L.) in comparison with a prebiotic on-growth performance, blood constitutes, immunity and intestinal microflora. The dietary treatments were as follows: basal diet (control); control + prebiotic (Fermacto™); control + 0.1% peppermint; control + 0.1% chicory, respectively. A significant (p < 0.05) body weight gain and feed intake was found at 21 and 42 days of growth period in broilers fed diet supplemented with 0.1% chicory compared with other groups. Feeding of prebiotic or chicory led to higher (p < 0.05) feed intake. Chickens fed control diet had higher (p < 0.05) abdominal fat compared with the other groups. Serum blood constituents indicated that broilers fed prebiotic or supplemented with peppermint or chicory had reduced (p < 0.05) levels of cholesterol, triglycerides and low-density lipoprotein than control group. Immunity-related parameters showed that chicken fed chicory had lower (p < 0.05) heterophil-to-lymphocyte ratio compared with the other groups. Intestinal microflora revealed that chickens fed prebiotic or herbals had higher count of Lactobacillus and lower E. coli than control. Thus, it can be concluded that broiler dietary supplementation with prebiotic or chicory can improve performance supporting positively health status.

An overview of food lipids toward food lipidomics

Increasing evidence regarding lipids' beneficial effects on human health has changed the common perception of consumers and dietary officials about the role(s) of food lipids in a healthy diet. However, lipids are a wide group of molecules with specific nutritional and bioactive properties. To understand their true nutritional and functional value, robust methods are needed for accurate identification and quantification. Specific analytical strategies are crucial to target specific classes, especially the ones present in trace amounts. Finding a unique and comprehensive methodology to cover the full lipidome of each foodstuff is still a challenge. This review presents an overview of the lipids nutritionally relevant in foods and new trends in food lipid analysis for each type/class of lipids. Food lipid classes are described following the LipidMaps classification, fatty acids, endocannabinoids, waxes, C8 compounds, glycerophospholipids, glycerolipids (i.e., glycolipids, betaine lipids, and triglycerides), sphingolipids, sterols, sercosterols (vitamin D), isoprenoids (i.e., carotenoids and retinoids (vitamin A)), quinones (i.e., coenzyme Q, vitamin K, and vitamin E), terpenes, oxidized lipids, and oxylipin are highlighted. The uniqueness of each food group: oil-, protein-, and starch-rich, as well as marine foods, fruits, and vegetables (water-rich) regarding its lipid composition, is included. The effect of cooking, food processing, and storage, in addition to the importance of lipidomics in food quality and authenticity, are also discussed. A critical review of challenges and future trends of the analytical approaches and computational methods in global food lipidomics as the basis to increase consumer awareness of the significant role of lipids in food quality and food security worldwide is presented.

Food Composition Databases (FCDBs): A Bibliometric Analysis

Food composition databases (FCDBs) are important tools that provide information on the nutritional content of foods. Previously, it was largely unclear what nutritional contents and which FCDBs were involved in highly cited papers. The bibliometric study aimed to identify the most productive authors, institutions, and journals. The chemicals/chemical compounds with high averaged citations and FCDBs used by highly cited papers were identified. In July 2023, the online database Web of Science Core Collection (WoSCC) was queried to identify papers related to FCDBs. A total of 803 papers were identified and analyzed. The first paper indexed in WoSCC was published in 1992 by Pennington, which described the usefulness of FCDB for researchers to identify core foods for their own studies. In that paper, the FCDB described was the USDA 1987-88 NFCS (the United States Department of Agriculture 1987-88 Nationwide Food Consumption Survey). The most productive author was Dr. Paul M. Finglas, the Head of the Food Databanks National Capability at the Quadram Institute (Norwich, UK) and the Managing Director of EuroFIR. His most cited paper among this dataset was about the development of an online Irish food composition database together with EuroFIR. The most productive institutions were the USDA and the World Health Organization (WHO) instead of universities. Flavonoid was the most recurring chemical class among the highly cited ones. The anti-oxidative properties and protective effects against heart disease and cancer of flavonoids might be some of the reasons for their popularity in research. Among the highly cited papers, the most heavily used FCDBs were the USDA database for the flavonoid content of selected foods, Fineli, the USDA National Nutrient Database for Standard Reference (USNDB), EuroFIR eBASIS-Bioactive Substances in Food Information Systems, and Phenol-Explorer. High-quality national and international FCDBs should be promoted and made more accessible to the research and public communities to promote better nutrition and public health on a global scale.

When Machine Learning and Deep Learning Come to the Big Data in Food Chemistry

Since the first food database was released over one hundred years ago, food databases have become more diversified, including food composition databases, food flavor databases, and food chemical compound databases. These databases provide detailed information about the nutritional compositions, flavor molecules, and chemical properties of various food compounds. As artificial intelligence (AI) is becoming popular in every field, AI methods can also be applied to food industry research and molecular chemistry. Machine learning and deep learning are valuable tools for analyzing big data sources such as food databases. Studies investigating food compositions, flavors, and chemical compounds with AI concepts and learning methods have emerged in the past few years. This review illustrates several well-known food databases, focusing on their primary contents, interfaces, and other essential features. We also introduce some of the most common machine learning and deep learning methods. Furthermore, a few studies related to food databases are given as examples, demonstrating their applications in food pairing, food-drug interactions, and molecular modeling. Based on the results of these applications, it is expected that the combination of food databases and AI will play an essential role in food science and food chemistry.

Natural language processing and machine learning approaches for food categorization and nutrition quality prediction compared with traditional methods

BACKGROUND

Food categorization and nutrient profiling are labor intensive, time consuming, and costly tasks, given the number of products and labels in large food composition databases and the dynamic food supply.

OBJECTIVES

This study used a pretrained language model and supervised machine learning to automate food category classification and nutrition quality score prediction based on manually coded and validated data, and compared prediction results with models using bag-of-words and structured nutrition facts as inputs for predictions.

METHODS

Food product information from University of Toronto Food Label Information and Price Database 2017 (n = 17,448) and University of Toronto Food Label Information and Price Database 2020 (n = 74,445) databases were used. Health Canada's Table of Reference Amounts (TRA) (24 categories and 172 subcategories) was used for food categorization and the Food Standards of Australia and New Zealand (FSANZ) nutrient profiling system was used for nutrition quality score evaluation. TRA categories and FSANZ scores were manually coded and validated by trained nutrition researchers. A modified pretrained sentence-Bidirectional Encoder Representations from Transformers model was used to encode unstructured text from food labels into lower-dimensional vector representations, followed by supervised machine learning algorithms (i.e., elastic net, k-Nearest Neighbors, and XGBoost) for multiclass classification and regression tasks.

RESULTS

Pretrained language model representations utilized by the XGBoost multiclass classification algorithm reached overall accuracy scores of 0.98 and 0.96 in predicting food TRA major and subcategories, outperforming bag-of-words methods. For FSANZ score prediction, our proposed method reached a similar prediction accuracy (R²: 0.87 and MSE: 14.4) compared with bag-of-words methods (R²: 0.72-0.84; MSE: 30.3-17.6), whereas structured nutrition facts machine learning model performed the best (R²: 0.98; MSE: 2.5). The pretrained language model had a higher generalizable ability on the external test datasets than bag-of-words methods.

CONCLUSIONS

Our automation achieved high accuracy in classifying food categories and predicting nutrition quality scores using text information found on food labels. This approach is effective and generalizable in a dynamic food environment, where large amounts of food label data can be obtained from websites.

Mediterranean Diet: The Role of Phenolic Compounds from Aromatic Plant Foods

Today's global food system aggravates climate change while failing in meeting SDG2 and more. Yet, some sustainable food cultures, such as the Mediterranean Diet (MD), are simultaneously safe, healthy, and rooted in biodiversity. Their wide range of fruits, herbs, and vegetables convey many bioactive compounds, often associated with colour, texture, and aroma. Phenolic compounds are largely responsible for such features of MD's foods. These plant secondary metabolites all share in vitro bioactivities (e.g., antioxidants), and some are evidenced in vivo (e.g., plant sterols lower cholesterol levels in blood). The present work examines the role of polyphenols in the MD, with respect to human and planetary health. Since the commercial interest in polyphenols is increasing, a strategy for the sustainable exploitation of Mediterranean plants is essential in preserving species at risk while valuing local cultivars (e.g., through the geographical indication mechanism). Finally, the linkage of food habits with cultural landscapes, a cornerstone of the MD, should enable awareness-raising about seasonality, endemism, and other natural constraints to ensure the sustainable exploitation of Mediterranean plants.

Food composition databases in the era of Big Data: Vegetable oils as a case study

Understanding the population's dietary patterns and their impacts on health requires many different sources of information. The development of reliable food composition databases is a key step in this pursuit. With them, nutrition and health care professionals can provide better public health advice and guide society toward achieving a better and healthier life. Unfortunately, these databases are full of caveats. Focusing on the specific case of vegetable oils, we analyzed the possible obsolescence of the information and the differences or inconsistencies among databases. We show that in many cases, the information is limited, incompletely documented, old or unreliable. More importantly, despite the many efforts carried out in the last decades, there is still much work to be done. As such, institutions should develop long-standing programs that can ensure the quality of the information on what we eat in the long term. In the face of climate change and complex societal challenges in an interconnected world, the full diversity of the food system needs to be recognized and more efforts should be put toward achieving a data-driven food system.

Practical Strategies for Improving Sustainability and Scale-up of Noncommunicable Disease-related Public Health Interventions: Lessons from the Better Health Program in Southeast Asia

Introduction

The Better Health Program has been addressing key health system issues in the prevention and control of noncommunicable diseases (NCDs) in Malaysia, Thailand, Vietnam, and the Philippines. As the program comes to an end, the sustainability and scaling-up of issues have assumed importance.

Objectives

The objective is to assess how well sustainability and scale-up strategies have been integrated into the design and implementation of a 3-year multicountry technical program; to explore enablers and barriers of sustainability and scaling up; and to identify practical strategies that can improve sustainability and scale-up of Better Health Program interventions.

Methods

We applied a staged approach to explore barriers and enablers and to identify practical strategies to improve sustainability and scale-up of four NCD interventions: community-based obesity prevention, front-of-pack labeling, local learning networks (LLNs), and NCD surveillance. We extracted evidence from peer-reviewed literature and local documents. We also conducted in-depth interviews with the implementation teams and key stakeholders. We conducted a thematic synthesis of the resulting information to identify practical strategies that improve sustainability and scale-up of the four interventions.

Results

Strong engagement of stakeholders at higher levels of the health system was identified as the main enabler, while limited funding and commitment from local governments were identified as a key barrier to sustainability and scale-up. Strengthening the social and institutional anchors of community health volunteers, enhancing evidence-based advocacy for front-of-pack labeling, trailblazing the LLN innovation, and securing the commitment of local governments in the implementation of NCD surveillance were among the key strategies for improving sustainability and scale-up of Better Health Program interventions in Malaysia, Thailand, Philippines, and Vietnam, respectively.

Conclusions

This study identified practical strategies for improving sustainability and scale-up of NCD-related interventions. Implementation of the strategies that had high priority and feasibility will improve the sustainability of critical elements of the program in the respective countries.

Food Habits: Insights from Food Diaries via Computational Recurrence Measures

Humans are creatures of habit, and hence one would expect habitual components in our diet. However, there is scant research characterizing habitual behavior in food consumption quantitatively. Longitudinal food diaries contributed by app users are a promising resource to study habitual behavior in food selection. We developed computational measures that leverage recurrence in food choices to describe the habitual component. The relative frequency and span of individual food choices are computed and used to identify recurrent choices. We proposed metrics to quantify the recurrence at both food-item and meal levels. We obtained the following insights by employing our measures on a public dataset of food diaries from MyFitnessPal users. Food-item recurrence is higher than meal recurrence. While food-item recurrence increases with the average number of food-items chosen per meal, meal recurrence decreases. Recurrence is the strongest at breakfast, weakest at dinner, and higher on weekdays than on weekends. Individuals with relatively high recurrence on weekdays also have relatively high recurrence on weekends. Our quantitatively observed trends are intuitive and aligned with common notions surrounding habitual food consumption. As a potential impact of the research, profiling habitual behaviors using the proposed recurrent consumption measures may reveal unique opportunities for accessible and sustainable dietary interventions.

Development of a nutrition management software based on selected Middle Eastern and Mediterranean dishes to support personalized diet and weight management

The nutrient composition of 50 commonly consumed Jordanian food dishes was determined to support the development of a novel nutrition management system designed to assist with dietary intake assessment and diet management. Composite dishes were selected by interviewing households located in the northern region of Jordan. For each dish, five different recipes were collected from experienced chefs and the typical recipe was formulated based on the average weights of ingredients and net weight of the dish. Proximate composition as well as vitamin and mineral contents were determined and related to ingredient composition and cooking conditions. The newly created food composition database was used to develop a user-centric nutrition management software tailored to reflect the characteristics of the Jordanian diet with representative items from this diet. This novel nutrition management system is customizable, enabling users to build daily meal plans in accordance with personalized dietary needs and goals.

Development of an Unified Food Composition Database for the European Project "Stance4Health"

The European Commission funded project Stance4Health (S4H) aims to develop a complete personalised nutrition service. In order to succeed, sources of information on nutritional composition and other characteristics of foods need to be as comprehensive as possible. Food composition tables or databases (FCT/FCDB) are the most commonly used tools for this purpose. The aim of this study is to describe the harmonisation efforts carried out to obtain the Stance4Health FCDB. A total of 10 FCT/FCDB were selected from different countries and organizations. Data were classified using FoodEx2 and INFOODS tagnames to harmonise the information. Hazard analysis and critical control points analysis was applied as the quality control method. Data were processed by spreadsheets and MySQL. S4H’s FCDB is composed of 880 elements, including nutrients and bioactive compounds. A total of 2648 unified foods were used to complete the missing values of the national FCDB used. Recipes and dishes were estimated following EuroFIR standards via linked tables. S4H’s FCDB will be part of the smartphone app developed in the framework of the Stance4Health European project, which will be used in different personalized nutrition intervention studies. S4H FCDB has great perspectives, being one of the most complete in terms of number of harmonized foods, nutrients and bioactive compounds included.

Identifying Nutrient Patterns in South African Foods to Support National Nutrition Guidelines and Policies

Food composition databases (FCDBs) provide the nutritional content of foods and are essential for developing nutrition guidance and effective intervention programs to improve nutrition of a population. In public and nutritional health research studies, FCDBs are used in the estimation of nutrient intake profiles at the population levels. However, such studies investigating nutrient co-occurrence and profile patterns within the African context are very rare. This study aimed to identify nutrient co-occurrence patterns within the South African FCDB (SAFCDB). A principal component analysis (PCA) was applied to 28 nutrients and 971 foods in the South African FCDB to determine compositionally similar food items. A second principal component analysis was applied to the food items for validation. Eight nutrient patterns (NPs) explaining 73.4% of the nutrient variation among foods were identified: (1) high magnesium and manganese; (2) high copper and vitamin B₁₂; (3) high animal protein, niacin, and vitamin B₆; (4) high fatty acids and vitamin E; (5) high calcium, phosphorous and sodium; (6) low moisture and high available carbohydrate; (7) high cholesterol and vitamin D; and (8) low zinc and high vitamin C. Similar food patterns (FPs) were identified from a PCA on food items, yielding subgroups such as dark-green, leafy vegetables and, orange-coloured fruit and vegetables. One food pattern was associated with high sodium levels and contained bread, processed meat and seafood, canned vegetables, and sauces. The data-driven nutrient and food patterns found in this study were consistent with and support the South African food-based dietary guidelines and the national salt regulations.