Many livestock and aquaculture feeds compete for resources with food production. Increasing the use of food system by-products and residues as feed could reduce this competition. We gathered data on global food system material flows for crop, livestock and aquaculture production, focusing on feed use and the availability of by-products and residues. We then analysed the potential of replacing food-competing feedstuff—here cereals, whole fish, vegetable oils and pulses that account for 15% of total feed use—with food system by-products and residues. Considering the nutritional requirements of food-producing animals, including farmed aquatic species, this replacement could increase the current global food supply by up to 13% (10–16%) in terms of kcal and 15% (12–19%) in terms of protein content. Increasing the use of food system by-products as feed has considerable potential, particularly when combined with other measures, in the much-needed transition towards circular food systems. Optimizing biomass use by reducing food–feed competition is paramount to achieving sustainable food systems. This study assesses global food systems in terms of livestock and aquaculture feed use and the availability of food system by-products and residues to quantify the potential for replacing food-grade feeds with food system by-products.
BackgroundHigh trans-fatty acid (TFA) intake is a major modifiable risk factor for cardiovascular disease (CVD), especially in older adults. This study aimed to assess global trends and health inequalities in CVD burden attributable to high TFA intake from 1990 to 2021 and project future patterns through 2036.MethodsUsing data from the Global Burden of Disease (GBD) Study 2021, we analyzed age-standardized mortality rates (ASMR), disability-adjusted life years (ASDR), and inequality indicators across 204 countries and territories. Age-Period-Cohort (APC) models and Bayesian projections were applied to estimate future trends.ResultsGlobally, ASMR and ASDR attributable to high TFA intake declined by 69 and 68%, respectively, from 1990 to 2021. The most significant reductions were observed in high-SDI regions, where comprehensive TFA bans and public health policies were implemented. In contrast, the absolute burden remains high in low- and middle-SDI countries due to limited policy enforcement and dietary interventions. Socioeconomic inequalities narrowed over time, but vulnerable populations still face elevated risks. Projections indicate a continued global decline in CVD burden attributable to TFA through 2036, though widening uncertainties reflect demographic and policy challenges.ConclusionWhile global progress in reducing TFA-related CVD burden is evident, persistent disparities and emerging risks in low-resource settings underscore the need for global elimination of industrial TFA, strengthened health systems, and targeted strategies to protect high-risk groups.
Liam Bakar, Zachary Englhardt, Vidya Srinivas
et al.
Food logging, both self-directed and prescribed, plays a critical role in uncovering correlations between diet, medical, fitness, and health outcomes. Through conversations with nutritional experts and individuals who practice dietary tracking, we find current logging methods, such as handwritten and app-based journaling, are inflexible and result in low adherence and potentially inaccurate nutritional summaries. These findings, corroborated by prior literature, emphasize the urgent need for improved food logging methods. In response, we propose SnappyMeal, an AI-powered dietary tracking system that leverages multimodal inputs to enable users to more flexibly log their food intake. SnappyMeal introduces goal-dependent follow-up questions to intelligently seek missing context from the user and information retrieval from user grocery receipts and nutritional databases to improve accuracy. We evaluate SnappyMeal through publicly available nutrition benchmarks and a multi-user, 3-week, in-the-wild deployment capturing over 500 logged food instances. Users strongly praised the multiple available input methods and reported a strong perceived accuracy. These insights suggest that multimodal AI systems can be leveraged to significantly improve dietary tracking flexibility and context-awareness, laying the groundwork for a new class of intelligent self-tracking applications.
Emanuele Calò, Giordano De Marzo, Vito D. P. Servedio
Ecosystems face intensifying threats from climate change, overexploitation, and other human pressures, emphasizing the urgent need to identify keystone species and vulnerable ones. While established network-based measures often rely on a single metric to quantify a species' relevance, they overlook how organisms can be both carbon providers and consumers, thus playing a dual role in food webs. Here, we introduce a novel approach that assigns each species two complementary scores--an importance index quantifying their centrality as carbon source and a predatory index capturing their vulnerability. We show that species with high importance index are more likely to trigger co-extinctions upon removal, while high-robustness index species typically endure until later stages of collapse, in line with their broader prey ranges. On the other hand, low robustness index species are the most vulnerable and susceptible to extinctions. Tested on multiple food webs, our method outperforms traditional degree-based analyses and competes effectively with eigenvector-based approaches, while also providing additional insights. This scalable and data-driven approach, relying solely on interaction data, provides a cost-effective tool that complements expert classifications for prioritizing conservation efforts.
Food banks are crucial for alleviating food insecurity, but their effectiveness hinges on accurately forecasting highly volatile in-kind donations to ensure equitable and efficient resource distribution. Traditional forecasting models often fail to maintain consistent accuracy due to unpredictable fluctuations and concept drift driven by seasonal variations and natural disasters such as hurricanes in the Southeastern U.S. and wildfires in the West Coast. To address these challenges, we propose FoodRL, a novel reinforcement learning (RL) based metalearning framework that clusters and dynamically weights diverse forecasting models based on recent performance and contextual information. Evaluated on multi-year data from two structurally distinct U.S. food banks-one large regional West Coast food bank affected by wildfires and another state-level East Coast food bank consistently impacted by hurricanes, FoodRL consistently outperforms baseline methods, particularly during periods of disruption or decline. By delivering more reliable and adaptive forecasts, FoodRL can facilitate the redistribution of food equivalent to 1.7 million additional meals annually, demonstrating its significant potential for social impact as well as adaptive ensemble learning for humanitarian supply chains.
This study investigated the effects of galactooligosaccharide (GOS) purity (80 % and 95 %) and heating time (60 and 150 min) on characterization and bioactivities of Maillard reaction products (MRPs) formed by conjugating tilapia scale collagen peptides with GOS at 90 °C. Spectral analyses, degree of grafting, and furosine content showed that lower-purity GOS exhibited higher glycation reactivity. Amino acid composition and advanced glycation end products analyses showed that arginine was the primary amino acid involved in glycation, with abundant formation of methylglyoxal-derived hydroimidazolone 1 (MG-H1). Digested MRPs maintained strong antioxidant activity, particularly in the lower-purity GOS group. Meanwhile, MRPs heated for 150 min enhanced L. casei, L. pentosus, and B. bifidum growth, while those heated for 60 min favored B. longum. MG-H1 level was positively correlated with antioxidant activity and B. bifidum growth (p < 0.01). This study highlights the broad application potential of glycated bioactive peptides with lower-purity GOS.
Nutrition. Foods and food supply, Food processing and manufacture
Short food supply chains (SFSCs) are increasingly garnering attention in food systems research, owing to their rising popularity among consumers, producers and policy-makers in the last few decades. Written with the aim to identify research gaps for the Horizon Europe research and innovation programme, this literature review provides a state of play of the definition and characterisation of SFSCs, and of their sustainability. Drawing on hypotheses about SFSC sustainability elaborated in an expert network in France, this review summarises a wide range of papers from various disciplines in the SFSC literature, written in English or French, while specifically highlighting the empirical results derived from European projects. Though the literature tends to generally agree on the social benefits of SFSCs, their economic and environmental impacts typically elicit more heterogeneous outcomes, while their health/nutrition and governance dimensions remain under-explored. Based on this review, recommendations for a future research and innovation programme are outlined, addressing the contribution of SFSCs to agrifood system transition and resilience in the current context of the Covid-19 crisis and of the Green New Deal objectives.
Starch-based materials have viscoelasticity, viscous film-forming, dough pseudoplasticity, and rheological properties, which possess the structural characteristics (crystal structure, double helix structure, and layered structure) suitable for three-dimensional (3D) food printing inks. 3D food printing technology has significant advantages in customizing personalized and precise nutrition, expanding the range of ingredients, designing unique food appearances, and simplifying the food supply chain. Precision nutrition aims to consider individual nutritional needs and individual differences, which include special food product design and personalized precise nutrition, thus expanding future food resources, then simplifying the food supply chain, and attracting extensive attention in food industry. Different types of starch-based materials with different structures and rheological properties meet different 3D food printing technology requirements. Starch-based materials suitable for 3D food printing technology can accurately deliver and release active substances or drugs. These active substances or drugs have certain regulatory effects on the gut microbiome and diabetes, so as to maintain personalized and accurate nutrition.
BACKGROUND Hospitals have a responsibility to support human health, and given the link between human and environmental health, hospitals should consider their environmental impacts. Hospital foodservices can negatively affect the environment at every stage of the food supply chain (production/procurement, distribution, preparation, consumption, and waste management/disposal). OBJECTIVE To systematically identify and synthesize the following across the hospital patient food/nutrition supply chain: environmental and associated economic impacts of foodservice; outcomes of strategies that aim to improve the environmental sustainability of foodservice; and perspectives of patients, staff, and stakeholders on environmental impacts of foodservice and strategies that aim to improve the environmental sustainability of foodservice. METHODS Eight electronic databases (ie, Cumulative Index to Nursing and Allied Health Literature Plus, Embase via Ovid, Global Health, National Health Service Economic Evaluation Database, Ovid Medline, ProQuest Environmental Science Collection, Scopus, and Web of Science) were searched from database inception to November 2018 for original research conducted across any stage of the hospital food supply chain (from production/procurement to waste management/disposal) that provides food/nutrition to patients, with no restrictions on language or study design. Titles/abstracts then full texts were screened independently by two authors. The Mixed Methods Appraisal Tool was used for quality appraisal for included studies. Data were synthesized narratively. RESULTS From 29,655 records identified, 80 studies met eligibility criteria. Results were categorized into production/procurement (n=12), distribution (n=0), preparation (n=6), consumption (n=49), waste management/disposal (n=8), and multiple food supply chain aspects (n=5). The environmental impact most widely explored was food waste, with many studies reporting on food waste quantities, and associated economic losses. Strategies focused on reducing food waste by increasing patients' intake through various foodservice models. Perspectives identified a shared vision for sustainable foodservices, although there are many practical barriers to achieving this. CONCLUSION The literature provides examples across the hospital food supply chain that demonstrate how environmental sustainability can be prioritized and evaluated and the opportunities for credentialed nutrition and dietetics practitioners to contribute. Future studies are warranted, particularly those measuring environmental impacts and testing the effects of sustainable strategies in the distribution, preparation, and waste management stages.
Food computing brings various perspectives to computer vision like vision-based food analysis for nutrition and health. As a fundamental task in food computing, food detection needs Zero-Shot Detection (ZSD) on novel unseen food objects to support real-world scenarios, such as intelligent kitchens and smart restaurants. Therefore, we first benchmark the task of Zero-Shot Food Detection (ZSFD) by introducing FOWA dataset with rich attribute annotations. Unlike ZSD, fine-grained problems in ZSFD like inter-class similarity make synthesized features inseparable. The complexity of food semantic attributes further makes it more difficult for current ZSD methods to distinguish various food categories. To address these problems, we propose a novel framework ZSFDet to tackle fine-grained problems by exploiting the interaction between complex attributes. Specifically, we model the correlation between food categories and attributes in ZSFDet by multi-source graphs to provide prior knowledge for distinguishing fine-grained features. Within ZSFDet, Knowledge-Enhanced Feature Synthesizer (KEFS) learns knowledge representation from multiple sources (e.g., ingredients correlation from knowledge graph) via the multi-source graph fusion. Conditioned on the fusion of semantic knowledge representation, the region feature diffusion model in KEFS can generate fine-grained features for training the effective zero-shot detector. Extensive evaluations demonstrate the superior performance of our method ZSFDet on FOWA and the widely-used food dataset UECFOOD-256, with significant improvements by 1.8% and 3.7% ZSD mAP compared with the strong baseline RRFS. Further experiments on PASCAL VOC and MS COCO prove that enhancement of the semantic knowledge can also improve the performance on general ZSD. Code and dataset are available at https://github.com/LanceZPF/KEFS.
Food image classification systems play a crucial role in health monitoring and diet tracking through image-based dietary assessment techniques. However, existing food recognition systems rely on static datasets characterized by a pre-defined fixed number of food classes. This contrasts drastically with the reality of food consumption, which features constantly changing data. Therefore, food image classification systems should adapt to and manage data that continuously evolves. This is where continual learning plays an important role. A challenge in continual learning is catastrophic forgetting, where ML models tend to discard old knowledge upon learning new information. While memory-replay algorithms have shown promise in mitigating this problem by storing old data as exemplars, they are hampered by the limited capacity of memory buffers, leading to an imbalance between new and previously learned data. To address this, our work explores the use of neural image compression to extend buffer size and enhance data diversity. We introduced the concept of continuously learning a neural compression model to adaptively improve the quality of compressed data and optimize the bitrates per pixel (bpp) to store more exemplars. Our extensive experiments, including evaluations on food-specific datasets including Food-101 and VFN-74, as well as the general dataset ImageNet-100, demonstrate improvements in classification accuracy. This progress is pivotal in advancing more realistic food recognition systems that are capable of adapting to continually evolving data. Moreover, the principles and methodologies we've developed hold promise for broader applications, extending their benefits to other domains of continual machine learning systems.
In response to the increasing demand for efficient and non-invasive methods to estimate food weight, this paper presents a vision-based approach utilizing 2D images. The study employs a dataset of 2380 images comprising fourteen different food types in various portions, orientations, and containers. The proposed methodology integrates deep learning and computer vision techniques, specifically employing Faster R-CNN for food detection and MobileNetV3 for weight estimation. The detection model achieved a mean average precision (mAP) of 83.41\%, an average Intersection over Union (IoU) of 91.82\%, and a classification accuracy of 100\%. For weight estimation, the model demonstrated a root mean squared error (RMSE) of 6.3204, a mean absolute percentage error (MAPE) of 0.0640\%, and an R-squared value of 98.65\%. The study underscores the potential applications of this technology in healthcare for nutrition counseling, fitness and wellness for dietary intake assessment, and smart food storage solutions to reduce waste. The results indicate that the combination of Faster R-CNN and MobileNetV3 provides a robust framework for accurate food weight estimation from 2D images, showcasing the synergy of computer vision and deep learning in practical applications.
Mario Vendrell Calatayud, Mario Vendrell Calatayud, Xueqi Li
et al.
IntroductionOlive oil is a widely recognized and appreciated food commodity, its quality and health benefits can be compromised when the oil goes through oxidative processes that may occur during production and storage. This study aimed to investigate the effects of the olive genotype on polar phenolic content after seven months of storage.MethodsOil produced from eight different olive cultivars (Leccino, Leccio del Corno, Moraiolo, Frantoio, Bianchera, Pendolino, Maurino, and Caninese) grown in southern Tuscany, Italy, were subjected to chemical analysis such as free fatty acids, peroxide value, K232 and K268, phenolics and UPLC-DAD at the beginning of the trial (Control) and seven months later (Stored).Results and ConclusionsFree fatty acids, peroxide values, K232 and K268, significantly increased, suggesting heightened hydrolysis and oxidation after storage. A cultivar effect was observed, with Leccino, Moraiolo, and Pendolino showing less susceptibility to oxidation (low differences between Control and Stored). In contrast, others (Bianchera and Caninese) are more affected (higher differences between Control and Stored). Phenolics analysis supports this observation, revealing that samples with higher resistance to oxidation exhibit elevated levels of hydroxytyrosol, tyrosol, vanillic acid, caffeic acid, p-coumaric acid, and ferulic acid. Principal Component Analysis highlights that Bianchera and Caninese cultivars correlate with rutin, tyrosol, and pinoresinol. As this research delves into the intricate relationship between genotype diversity, phenolic composition, and oxidative stability, a nuanced understanding emerges, shedding light on how different cultivars may present varying compositions and concentrations of phenols, ultimately influencing the oil’s resistance to the oxidation that occurred during storage.
Tomohiro Furuki, Tomohiro Nobeyama, Shunji Suetaka
et al.
Reentrant condensation (RC) is a protein behavior in which the protein solution shifts between the one- and two-phase state more than twice by increasing a single parameter. Although RC would be a candidate mechanism for the physicochemical design of food additives, no realistic model has been established under diverse contaminants like food materials. Here, we found that a mixture of cola and milk yielded RC. At pH 3.2–3.6, cola induced milk condensation at 30–40%, while lower or higher concentrations of cola did not. Furthermore, we reduced this cola/milk system to two pure components, casein in milk and polyphosphate (polyP) in cola, and investigated the characteristics of casein concentration and zeta potential. This was the first experimental demonstration of RC occurrence in a multicomponent system. The well-characterized cola/milk system would explore both the universal nature of proteins and the industrial application of RC.
Nutrition. Foods and food supply, Food processing and manufacture
Food security and the utilization of natural resources in a sustainable manner are vital to the expansion of China's agricultural system. The relationship between environmental pressure and dietary structure has influenced the quantity and spatial distribution of China's food supply and demand, but it has not been evaluated. Our research centered on the security of China's food nutrition–resources–food (NRF) system, considering the inherent relationship between food security, nutritional health, and resource security. The following are the study's findings: (1) The Chinese population is rapidly changing from a diet focused on grains to a more diverse diet. Between 1990 and 2019, the dietary quality and nutritional status of Chinese individuals have vastly improved. In terms of nutrient levels, discrepancies between urban and rural resident persist, with urban residents consuming a diet that is closer to the ideal structure. However, the structure of rural residents' food consumption is diversifying, and the gap between urban and rural residents is gradually narrowing. (2) From 2000 to 2019, the pressure, status, and response indices of China's NRF system all show an upward trend, and the security of the NRF system has steadily grown. The magnitude of change in the response index exceeded that of the state index, which exceeded that of the pressure index. This indicates that the increase in the pressure and state indices of the NRF system was primarily attributable to the effectiveness of policy efforts.
Abstract Addressing food loss and waste (FLW) globally is critical for both improving food security and mitigating environmental pollution. While there are numerous studies addressing FLW in terms of nutrition, food security, food safety, public health and the economy, there is only a small body of life cycle assessment (LCA) research aimed at understanding impacts from FLW. We conducted a literature review of LCA studies focused on FLW in the food supply chain (FSC) to ascertain the state of the science and identify the research gaps. We identified 22 original research articles that met our search criteria and spanned the four stages of LCA. Regarding the goal and scope, there were a dearth of studies focused on the top of the waste hierarchy (prevention). Further, we identified a research gap in studies that accounted for avoided production from food waste management in the overall LCA and distinguished between avoidable and unavoidable waste streams. LCA studies to date largely used a mass-basis as the functional unit and were limited in terms of spatial and temporal specificity. Within the life cycle inventory, most of the studies were conducted in Europe and only one study in the US. In addition, some of the studies lack data transparency. The life cycle impact assessment phase showed that most of the studies only assess global warming potential with fewer studies evaluating energy, water demand and human toxicity. Lastly, within life cycle interpretation more than half of the studies focus on at least one of the three types of uncertainties supporting more informed policy decision making.
Tianyu Jiang, Sonia Vinogradova, Nathan Stringham
et al.
Our research explores the use of natural language processing (NLP) methods to automatically classify entities for the purpose of knowledge graph population and integration with food system ontologies. We have created NLP models that can automatically classify organizations with respect to categories associated with environmental issues as well as Standard Industrial Classification (SIC) codes, which are used by the U.S. government to characterize business activities. As input, the NLP models are provided with text snippets retrieved by the Google search engine for each organization, which serves as a textual description of the organization that is used for learning. Our experimental results show that NLP models can achieve reasonably good performance for these two classification tasks, and they rely on a general framework that could be applied to many other classification problems as well. We believe that NLP models represent a promising approach for automatically harvesting information to populate knowledge graphs and aligning the information with existing ontologies through shared categories and concepts.
With the growth in capabilities of generative models, there has been growing interest in using photo-realistic renders of common 3D food items to improve downstream tasks such as food printing, nutrition prediction, or management of food wastage. Despite 3D modelling capabilities being more accessible than ever due to the success of NeRF based view-synthesis, such rendering methods still struggle to correctly capture thin food objects, often generating meshes with significant holes. In this study, we present an optimized strategy for enabling improved rendering of thin 3D food models, and demonstrate qualitative improvements in rendering quality. Our method generates the 3D model mesh via a proposed thin-object-optimized differentiable reconstruction method and tailors the strategy at both the data collection and training stages to better handle thin objects. While simple, we find that this technique can be employed for quick and highly consistent capturing of thin 3D objects.
Vision-language models (VLMs) have shown impressive performance in substantial downstream multi-modal tasks. However, only comparing the fine-tuned performance on downstream tasks leads to the poor interpretability of VLMs, which is adverse to their future improvement. Several prior works have identified this issue and used various probing methods under a zero-shot setting to detect VLMs' limitations, but they all examine VLMs using general datasets instead of specialized ones. In practical applications, VLMs are usually applied to specific scenarios, such as e-commerce and news fields, so the generalization of VLMs in specific domains should be given more attention. In this paper, we comprehensively investigate the capabilities of popular VLMs in a specific field, the food domain. To this end, we build a food caption dataset, Food-500 Cap, which contains 24,700 food images with 494 categories. Each image is accompanied by a detailed caption, including fine-grained attributes of food, such as the ingredient, shape, and color. We also provide a culinary culture taxonomy that classifies each food category based on its geographic origin in order to better analyze the performance differences of VLM in different regions. Experiments on our proposed datasets demonstrate that popular VLMs underperform in the food domain compared with their performance in the general domain. Furthermore, our research reveals severe bias in VLMs' ability to handle food items from different geographic regions. We adopt diverse probing methods and evaluate nine VLMs belonging to different architectures to verify the aforementioned observations. We hope that our study will bring researchers' attention to VLM's limitations when applying them to the domain of food or culinary cultures, and spur further investigations to address this issue.