搜索结果：Comprehensive reviews in food science and food safety

The demand for protein-based liquid foods is increasing due to growing awareness of the impact of diet on human health. This trend has prompted the food industry to explore minimal processing technologies that ensure both safety and clean-label appeal. This review presents a comprehensive assessment of selected innovative nonthermal technologies-based on high pressure, electromagnetic, acoustic, plasma fields, and membrane filtration principles-to process protein-based liquid foods. Key engineering considerations for designing process conditions suitable for protein systems are discussed. The review also examines the effects of these technologies on microbiological safety and quality attributes, including structural (particle size and microstructure), functional (solubility, rheology, emulsification, and foaming properties), and nutritional aspects (digestibility and allergenicity), along with possible underlying mechanisms. Findings highlight the importance of uniform application of the lethal agent (e.g., pressure, temperature, and electrical field) and thermal effects within the processed volume to validate microbial safety. Product-specific factors such as composition including fat and protein, pH, and water activity must also be carefully considered. Evidence suggests that nonthermal technologies can induce diverse structural and conformational changes in proteins, thereby altering their interactions with other food components and leading to variable impacts on quality attributes such as viscosity and emulsion stability. Increasing thermal intensity in combination with nonthermal agents generally degrade product quality. Future research should aim to optimize nonthermal processing parameters for a variety of protein-based foods by integrating both process and product factors to ensure microbial safety and enhanced product quality. The strategic application of nonthermal technologies-alone or in combination with mild thermal treatments-offers significant potential for developing sustainable, high-quality, and tailor-made protein-based food products.

The California almond industry underwent a remarkable transformation in food safety management and culture following outbreaks of salmonellosis associated with the consumption of raw almonds in 2000-2001 and 2003-2004. However, limited studies have examined these changes from a longitudinal perspective. This study documents the transformation of food safety management in the California almond industry over an 18-year period, explores indicators of change in food safety culture, identifies the key factors driving these changes, and examines the determinants of industry-wide technology adoption. A multifaceted approach was used, consisting of document analysis and semi-structured interviews. This study provides a detailed review of the almond industry's responses to the outbreaks, highlighting the industry commodity board's proactive leadership in crisis management, collaborative research efforts, risk assessment, and the development of a mandatory Salmonella-control program to mitigate the risks associated with raw almonds. These measures significantly strengthened food safety management systems across the industry. The industry has also shown a shift in mentality toward food safety over time, evidenced by increased prioritization of food safety, stronger management commitment, and reduced resistance to change. A conceptual framework integrating institutional theory and diffusion of innovation theory is proposed to illustrate how external and internal institutional pressures, along with intervention characteristics, influenced the almond industry's adoption of Salmonella-control interventions. The study offers valuable lessons on proactive, industry-driven food safety improvements and self-regulation in enhancing food safety outcomes.

The pervasive threat of microbial contamination and the escalating crisis of antimicrobial resistance necessitate the development of novel, sustainable food preservation strategies. Antimicrobial peptides (AMPs), especially those sourced from food-grade microbes, are emerging as feasible substitutes for conventional chemical preservatives. They provide substantial benefits, including biodegradability, excellent biocompatibility, and a beneficial foundation for later safety evaluations and risk management. This review critically synthesizes current knowledge on AMPs sourced from edible fungi, fermented fungi, and probiotics. It systematically reviews AMP biosynthetic pathways and sources, with an emphasis on structure-activity relationships to link structural features to antimicrobial activity and safety. It further analyzes mechanisms of action across two major modes, membrane targeting and intracellular targeting, and surveys AI-assisted de novo design strategies. Advanced preparation and screening workflows are summarized. Finally, it discusses progress and limitations in food systems and emerging applications in active and intelligent packaging. Key insights reveal that food-grade microbial AMPs are predominantly cationic and amphipathic, with their activity fine-tuned by molecular weight, amino acid composition, and secondary structure. Incorporating these peptides into nanofiber membranes, nanoparticle delivery systems, and biosensors can mitigate the constraints associated with their direct application as antimicrobial agents in food. This method efficiently prolongs food shelf life and facilitates real-time quality assessment. However, challenges such as batch-to-batch variability leading to inconsistent activity, cost-effective production, and the need for standardized safety assessment remain. Thus, future research should focus on the synergistic role of multiomics, AI-assisted design, and precision fermentation in propelling the field toward sustainable and intelligent food packaging solutions.

Green oil-in-water (O/W) nanoemulsions incorporating essential oils (EOs) and food-derived compounds are gaining prominence as biopesticidal platforms that address the growing demand for sustainable agriculture, food safety, and reduced chemical inputs. These colloidal delivery systems enhance the solubility, stability, and bioefficacy of hydrophobic phytochemicals, offering an environmentally friendly alternative to conventional agrochemicals. This comprehensive review critically examines recent advances in the formulation and application of green O/W nanoemulsions for crop protection. We synthesize evidence on their pesticidal activity against a wide range of pests and phytopathogens relevant to food crop systems, including aphids, insects, fungi, bacteria, and weeds. Emphasis is placed on food-grade and biodegradable formulation components, such as biosurfactants and natural emulsifiers, as well as their implications for toxicological safety, environmental risk, and scalability. Nanoemulsions have been reported to enhance pest control relative to conventional formulations; however, the reported efficacy varies depending on the formulation composition, target organism, and application conditions. Their full potential remains underexplored in terms of field application, phytotoxicity, and impacts on nontarget organisms. Important gaps persist in addressing underrepresented targets such as plant viruses and nematodes. By integrating concepts of green chemistry, nanotechnology, and food system resilience, this review provides a forward-looking perspective on the role of green nanoemulsions in sustainable crop management. Their development aligns with the Sustainable Development Goals and offers promising solutions for integrated pest management, organic agriculture, and preharvest food safety, reinforcing the transition toward safer and more resilient food systems.

Food safety and quality are increasingly undermined by adulterants, contaminants, pathogens, toxins, and spoilage processes, necessitating rapid, sensitive, cost-effective, and environmentally sustainable detection systems. Traditional analytical methods typically depend on synthetic dyes, heavy metal salts, strong acids, and resource-intensive processes, limiting sustainability and on-site applicability. In line with green analytical chemistry and clean-label technologies, naturally derived sensing materials have gained increasing attention. Curcumin, a natural polyphenolic compound from Curcuma longa, has emerged as a safe, renewable, low-toxicity, and multifunctional alternative for agro-food safety monitoring. Its pH responsiveness, fluorescence behavior, redox activity, metal-chelation capacity, chemical reactivity, and strong binding affinity enable its function as a ligand, indicator, and active sensing agent. This review critically examines recent advances in curcumin-based detection strategies for food adulterants, pathogens, toxins, microbial spoilage, and quality deterioration, with a focus on its interaction mechanisms, colorimetric and fluorescent indicators, biosensors, smart packaging, and shelf-life monitoring. Additionally, the potential of curcumin-based indicators for detecting pesticide residues, toxic gases, and chemical residues, beyond the food sector, is also discussed. Although curcumin-based sensing systems represent a promising and sustainable approach for advancing next-generation food safety monitoring, further studies are required to improve sensor integration, stability, and selectivity to allow large-scale adoption.

Ergot alkaloids (EAs), toxic secondary metabolites produced by Claviceps purpurea, pose food and feed safety concerns for cereal grains, particularly rye and wheat. While EAs are most frequently associated with rye, their occurrence in other cereals has been increasingly reported across diverse regions. This review critically synthesizes current knowledge on the global occurrence, analytical detection, regulatory control, and processing behavior of EAs, with a particular focus on factors that limit data reliability and risk management. A defining challenge in EA monitoring is the highly heterogeneous distribution of ergot sclerotia and alkaloids within bulk grain, which introduces substantial sampling uncertainty that often exceeds analytical variability. Recent advances in comminution, spatial sampling models, and composite sampling strategies have improved representativeness, yet remain underemphasized relative to instrumental developments. Analytical methods have progressed from nonspecific HPLC approaches to high-throughput LC-MS/MS and LC-HRMS platforms, enabling sensitive, multi-analyte detection and epimer-specific quantification; however, persistent limitations, including matrix effects, limited availability of reference materials, and insufficient characterization of conjugated or matrix-bound EAs, continue to constrain comparability across studies. Regulatory approaches vary globally, with the European Union adopting chemical limits for total EAs, while other regions rely primarily on sclerotia-based grading. Climate change and evolving agricultural practices further complicate exposure assessment and mitigation. By integrating sampling theory, analytical chemistry, processing effects, and regulatory frameworks, this review provides a unified perspective on EA risk management and identifies priorities for future research, including harmonized sampling strategies and high-throughput analytical platforms capable of capturing both free and conjugated EA forms.

The projected global protein deficit, expected to exceed 250 million tons by 2050, coincides with fruit-processing residues that represent nearly 30%-40% of processed biomass, highlighting the need for integrated solutions that simultaneously address nutritional security and food-waste valorization. This review critically synthesizes current evidence on single-cell protein (SCP) production from fruit-processing waste, examining how substrate composition, microbial metabolism, and downstream processing determine nutritional quality, techno-functional performance, safety, and industrial feasibility. Reported systems consistently produce 45%-65% crude protein, with essential amino-acid profiles comparable to soybean meal and fishmeal. Comparative analysis indicates that carbon-to-nitrogen ratio, pretreatment severity, oxygen transfer, and nucleic-acid reduction strategies, rather than microorganism type alone, are the dominant factors controlling yield and composition. Techno-functional properties, including water-holding capacity (2.1-4.8 g/g), emulsifying activity (35%-62%), and foaming capacity (28%-55%), are strongly influenced by cell-wall structure and protein recovery approaches, while bioactive potential is increasingly enhanced through omics-guided strain optimization. Economic assessments identify downstream processing and biomass drying as the major cost drivers, accounting for more than 40% of total production costs, and reveal that feedstock compositional variability is the principal source of inconsistent productivity across studies. Safety evaluations highlight persistent knowledge gaps related to mycotoxin contamination, allergenicity, and acceptable nucleic-acid levels, which currently limit regulatory acceptance despite favorable environmental outcomes demonstrated by life-cycle analyses. Unlike earlier reviews that primarily catalog substrates and microorganisms, this work integrates nutritional, functional, safety, techno-economic, and technological-readiness perspectives, emphasizing the need for process integration, predictive consortium control, and standardized safety frameworks to enable the large-scale adoption of fruit-waste-derived SCP in sustainable food and feed systems.

Aquatic foods are essential sources of protein and micronutrients and play a critical role in global nutrition, trade, and livelihoods. However, their safety and sustainability are frequently compromised by microbial contamination and biofilm formation during farming, processing, storage, and retail. Biofilms persist on moist surfaces, resist conventional cleaning practices, and contribute to spoilage, cross-contamination, and economic loss. This article reviews emerging applications of artificial intelligence and Industry 4.0 technologies for biofilm prevention and control in aquaculture and seafood systems. Particular emphasis is placed on the use of continuous water quality sensing, imaging platforms for early detection and cleaning verification, genomic and omics tools for microbial trait-level insight, and digital twin frameworks for virtual simulation of sanitation strategies. Recent advances demonstrate that sensor telemetry can predict biofilm-favorable conditions, imaging can verify removal in real time, and genomic data can identify persistence traits and tolerance mechanisms. When integrated, these approaches enable facility-specific digital twins that anticipate surface-specific risks and recommend optimized interventions before implementation. The convergence of AI, sensor networks, imaging, and omics offers a shift from reactive to proactive biofilm management in aquatic food systems. Positioned within the transition to Industry 5.0, these innovations support earlier detection, targeted interventions, and measurable improvements in food safety, quality, sustainability, and resilience, while aligning production systems with human-centric goals.

Eggshell membrane (ESM) is a poultry egg-processing by-product increasingly recognized as a valuable food-relevant bioresource and a potential food-grade functional ingredient due to its rich composition (collagen, keratin, glycosaminoglycans) and inherent bioactivities. This comprehensive review examines the structural composition and key functional properties of ESM, including antioxidant, anti-inflammatory, and antimicrobial activities. Recent advances in ESM separation and modification techniques are systematically evaluated, spanning physical (mechanical separation, ultrasonication), chemical (acid/base treatments), and biological (enzymatic hydrolysis) methods, with particular emphasis on green, sustainable processing approaches that preserve functionality. Interdisciplinary applications of ESM are critically discussed across the food, health, and packaging sectors. For example, ESM is being explored as a functional ingredient in nutraceuticals and functional foods, as a functional ingredient in nutraceuticals and functional foods (including dietary supplements), and as an active component in biodegradable/edible packaging materials; selected non-food biomedical uses are briefly noted for context. This review also addresses current challenges in scaling up ESM utilization-including improving extraction efficiency, ensuring quality consistency, and meeting regulatory requirements-and highlights future directions for precision modification and multi-sector collaboration. Overall, the integrative analysis provides scientific insights and technical guidance for the sustainable high-value utilization of ESM across multiple disciplines.

Flavor-induced sensory satisfaction is critical for food acceptance and market success. However, traditional sensory evaluation methods, relying heavily on subjective assessments, often fail to accurately reflect real-time, objective neural processing underlying complex multisensory flavor experiences. This limitation highlights the need for innovative methods that objectively quantify how flavors are perceived and integrated within the brain. In this review, we first examine the neural pathways underlying flavor perception, focusing on how gustatory, olfactory, and oral somatosensory inputs interact with reward and hedonic networks to form integrated flavor experience. Building on this foundation, we then outline the latest strategies for developing flavor-oriented brain-computer interface (flavor-BCI), summarizing key features of various neuroimaging techniques and associated technical implementation workflows. Finally, we assess emerging applications of flavor-BCI in sensory assessment and consumer decision-making and identify opportunities and challenges for future food design and product development. Flavor perception begins with parallel encoding of chemical stimuli in the primary gustatory and olfactory cortices and in trigeminal pathways. These signals are subsequently integrated in higher order regions, forming a distributed neural network across cortical, limbic, and subcortical structures that support flavor recognition, hedonic appraisal, and motivated eating. Flavor-BCI systems record neural activity from these regions using electrophysiology or neuroimaging and apply advanced algorithms to decode neural representations, translating them into objective sensory outputs. Relative to traditional evaluations, this approach enables real-time, precise quantification of flavor experience. Flavor-BCI thus offers promising avenues for intelligent sensory evaluation and novel human-machine interactions.

In the context of the circular economy and the increasing demand for safe and sustainable packaging, this work addresses the safety assessment of food contact materials (FCMs) derived from agro-industrial by-products. Despite growing interest in these bio-based materials, the literature still lacks a structured safety-assessment framework able to account for substrate-related contaminants, microbial processing, and downstream impurities. The novelty of this work lies in applying the European Food Safety Authority (EFSA) Technical Report perspective on natural mixtures to two representative case studies: bacterial cellulose (BC) produced by Komagataeibacter spp., as a promising microbial biopolymer for food packaging application, and nisin produced by Lactococcus lactis subsp. lactis, as an antimicrobial peptide, to functionalize the packaging material. This study aims to evaluate whether the EFSA-oriented framework can support the identification of potential substances of concern across the production chain when cellulose is produced starting from agro-industrial waste or nisin is applied. For this, a systematic literature review (SLR) was conducted to investigate potential substances of concern from agro-industrial substrates through fermentation to the final activated materials. The findings highlight the need to characterize natural carbon sources, including pesticide residues, consider the qualified presumption of safety (QPS) status of production microorganisms, and assess metabolites and fermentation by-products. The behavior of these substances during processing and their potential migration into food are critical aspects. A preliminary safety assessment at early development stages is therefore essential to guide material design and regulatory compliance. Overall, this study provides a practical framework to support researchers, developers, and risk assessors in identifying safety concerns and improving the regulatory readiness of innovative bio-based FCMs.

In the food industry, freezing heat transfer models have emerged as indispensable tools for optimizing freezing processes by accurately simulating temperature distribution and phase change dynamics, thereby enhancing both freezing efficiency and product quality. This paper systematically reviews the fundamental theories of heat and mass transfer models in food freezing, with a focus on three critical influencing factors: Food composition (moisture, protein, fat), which governs thermophysical properties and component interactions; Porous structure, which modulates thermal conductivity and water migration through porosity and connectivity; and geometric parameters (shape/size), which determine heat transfer uniformity via surface-to-volume effects. The study evaluates the applicability and limitations of analytical and numerical approaches, demonstrating that while analytical models enable rapid freezing time estimation, their simplicity restricts adaptability to complex food systems. In contrast, numerical models-including the finite difference method (FDM), finite element method (FEM), and finite volume method (FVM)-excel in handling phase change latent heat and unsteady heat transfer, providing reliable predictions of temperature distribution during freezing. By synthesizing theoretical insights and practical applications, this review highlights the pivotal role of advanced modeling techniques in improving process control and preserving food quality, while also identifying future research directions for further optimization of freezing technologies.

Artificial intelligence (AI) has been increasingly applied to address challenges in food packaging, including food waste, sustainability, and real-time quality assurance. However, existing studies are often confined to specific applications, with limited integration across different stages of the packaging life cycle and insufficient linkage between material performance, functionality, and system-level outcomes. This review systematically analyzes peer-reviewed studies retrieved from the Web of Science Core Collection (2021-2025), selected based on their relevance to AI applications in food packaging, including material performance, safety, and life cycle management. A life cycle-oriented framework is proposed, linking major AI paradigms (supervised, unsupervised, reinforcement, deep learning, and hybrid models) to six key domains: material design, production optimization, food quality prediction, safety assurance, smart labeling and traceability, and recycling. Within this framework, AI supports data-driven prediction, monitoring, and decision-making, whereas hybrid models improve robustness in complex, multifactor systems. Despite challenges related to data quality, model generalization, and regulatory acceptance, AI-driven packaging systems may support a transition from passive containment toward more adaptive and data-informed solutions that improve efficiency, sustainability, and consumer trust.

Low-moisture foods (LMFs) are prone to microbial contamination. Thermal treatments, although effective, could compromise LMFs quality. Hence, interest in alternative decontamination strategies such as light-based treatments has increased. This manuscript aims to provide insights into using light-based decontamination technologies in LMFs processing by reviewing the 1) microbial contamination issues affecting LMFs, 2) application, mechanisms of action, and process factors affecting light-based decontamination efficacy, and 3) potential challenges and future directions on establishing light-based technologies as a standard LMFs decontamination step. Pathogenic microorganism contamination of LMFs has been well documented in the existing literature. The use of LEDs, UV, and pulsed light (PL) in diverse LMFs categories produced reductions ranging from 0.7 to 4.0. These reductions were facilitated by induced inactivation mechanisms including ROS generation, cell membrane damage, DNA damage, and localized heating. The decontamination efficacy of LEDs, UV, and PL treatments was also affected by factors such as target microorganism, processing conditions, and inherent food matrix characteristics of treated LMFs. The review findings also found critical gaps that need to be addressed such as 1) lack of standardized experiment/validation methodologies, 2) inadequate exposure of LMFs due to shadowing/obstruction, 3) sub-optimal process and light delivery engineering, 4) lack of process validations on scaled-up/commercial equipment, and 5) insufficient techno-economic assessments on cost-effectiveness and throughput capability. The review findings indicate that light-based technologies can be a viable alternative decontamination step for LMFs although they have critical limitations to be addressed before being fully usable in LMFs processing.

The migration of intentionally and non-intentionally added substances (IAS/NIAS) from food packaging into foodstuffs presents a significant challenge to consumer health and food safety. Accurate and comprehensive identification of these chemical migrants is therefore paramount. This review systematically summarizes recent advances in the analytical workflows used to identify these migrants. We critically evaluate the latest developments in both gas chromatography coupled to mass spectrometry (GC-MS) and liquid chromatography coupled to high-resolution mass spectrometry (LC-HRMS). Special attention is given to cutting-edge techniques, such as comprehensive two-dimensional gas chromatography (GC × GC) for enhanced separation of complex mixtures, high-resolution filtering (HRF) for leveraging the dual advantages of gas chromatography coupled to high-resolution mass spectrometry (GC-HRMS) accurate mass measurements and conventional low-resolution spectral matching, and ion mobility spectrometry (IMS) for its unique ability to resolve isomers. Concurrently, we provide an in-depth critique of the evolving data analysis strategies, from conventional targeted analysis to the more comprehensive suspect and nontargeted screening approaches. The principles, advantages, and limitations of each workflow are discussed in the context of their application to food packaging materials. Then, the review dissects major bottlenecks, notably the scarcity of reference standards and comprehensive mass spectral libraries, which hinder confident identification. Looking forward, we highlight promising future directions, emphasizing that the synergistic integration of open-access mass spectral databases, adoption of novel analytical techniques, and machine learning-based molecular property prediction will facilitate the identification of IAS and NIAS in food packaging. In addition, integrating chemical analysis with bioassays will enable the prioritization of high-hazard chemicals, ultimately improving the safety evaluation of food packaging.

Lipid droplets (LDs) are central organelles governing lipid metabolism, and their dynamics (including formation, expansion, and degradation) are closely linked to systemic lipid homeostasis. Tea has been demonstrated to alleviate lipid metabolic disorders, potentially through its regulatory on LDs. Tea enhances the abundance of Lachnoclostridium, promoting the production of lysophosphatidylethanolamine (LPE), lysophosphatidylcholine (LPC), and phosphatidylinositol (PtdIns). Concurrently, tea elevates the levels of butyric acid by enriching Blautia, Allobaculum, and Bifidobacterium, thereby contributing to elevated phosphatidylcholine (PC) synthesis. Tea-derived tryptophan is metabolized by Clostridiaceae, Bacteroidales S24-7 group, and Lactobacillaceae into indole-3-propionic acid, which activates the aryl hydrocarbon receptor and upregulates Pemt via the insulin/protein kinase B signaling pathway, facilitating phosphatidylethanolamine (PE)-to-PC conversion. In parallel, tea promotes Prevotella, Bacteroides, and Muribaculaceae, while suppressing Desulfovibrio, favoring a reduction in PE levels. Collectively, these alterations reshape the phospholipid composition of the LD monolayer, characterized by increased proportions of LPE, LPC, PtdIns, and PC and reduced PE. This shift elevates the Γ2/γ (Γ, line tension; γ, surface tension) ratio, limits LD fusion, and increases LD-specific surface area, thereby enhancing lipolytic enzyme access to stored triglycerides. The primary aim of this review is to systematically clarify the molecular mechanisms by which tea regulates LD dynamics and to examine its translational potential in the food sector. We summarize the molecular basis of LD dynamics, how tea modulates the gut microbiota and LD phospholipid composition, the signaling pathways involved in LD homeostasis, and applications of these mechanisms in food, including optimizing nanoemulsions and achieving targeted LD regulation through enhanced tea bioactivity.

An organization's food safety culture (FSC) is increasingly being recognized as a key factor in its food safety performance. The reliable measurement of FSC is therefore crucial for identifying the strengths and weaknesses of food companies in food safety performance, enabling them to guide continuous improvement. While questionnaires are the most widely used tool for FSC assessment, existing instruments vary considerably in their scope, structure, and methodological rigor, which limits the consistency, comparability, and practical utility of these instruments. This review systematically examines all publicly available FSC questionnaires to consolidate the knowledge about their scope, content, and validation practices. Specifically, the review (1) outlines the characteristics of FSC studies and tracks the evolution of their questionnaires; (2) identifies and clusters the dimensions across instruments, highlighting commonalities and differences of the scope across questionnaires; (3) summarizes the items used within each clustered dimension to illustrate how FSC is operationalized; (4) examines all the validation methods employed; and (5) outlines opportunities for further development and application of FSC questionnaires. This article reveals significant diversity in theoretical frameworks, conceptual scope, and focus areas of detailed items, as well as notable gaps in validation practices. The findings provide a comprehensive overview of existing tools, offer practical guidance for researchers, industry practitioners, and regulators on selecting or developing robust questionnaires, and highlight the need for greater methodological rigor in the future development of FSC questionnaires.

Molecular hydrogen (H2) has emerged as a functional bioactive gas with exceptional diffusivity, selective radical-scavenging ability, and proven biosafety, offering new opportunities for next-generation food preservation technologies. Hydrogen-based reducing atmosphere packaging (H2-RAP), which diverges from conventional modified atmosphere packaging (MAP) that passively suppresses spoilage through CO2, O2, and N2, provides on-demand oxidative protection by directly regulating redox balance within food systems. This review critically evaluates the limitations of conventional MAP and develops a mechanistically discriminative framework for H2-RAP linking oxidative control to redox potential-linked spoilage trajectories. It summarizes advances in headspace-matrix exchange, dissolution/partitioning and interfacial exchange, effective diffusivity (Deff), and retention/decay kinetics, together with controlled-delivery architectures. The comparative analysis of different food categories demonstrates that H2-RAP effectively suppresses oxidation, microbial proliferation, and nutrient degradation at low and safe concentrations, thereby extending the shelf life of food products. Despite these advantages, the challenges of maintaining H2 retention, optimizing release kinetics, and establishing standardized evaluation and safety frameworks remain unresolved. Through the integration of knowledge from food chemistry, materials science, and sustainability science, this review provides a unified perspective on H2-based packaging systems and delineates key research priorities for the safe, scalable, and intelligent industrial implementation.

Food safety remains a critical factor in preventing contaminated and hazardous products from reaching consumers. The integration of artificial intelligence (AI) and its capacity to deal with vast datasets has significantly enhanced food safety protocols, and a substantial number of primary and secondary studies have emerged at the intersection of these two domains. Although several studies have addressed AI applications in food safety, no tertiary study has yet synthesized the collective insights from existing systematic reviews. To address this gap, this paper provides a comprehensive overview of the current state of AI applications in food safety through a systematic tertiary analysis of secondary studies. By systematically analyzing secondary studies, this research identifies key trends such as the food categories most frequently investigated, the data sources utilized, prevalent food safety hazards, the commonly adopted AI algorithms, and the challenges associated with their implementation within the field. The analysis revealed that dairy products received the greatest research attention, with sensing data serving as the primary data source. Neural networks emerged as the predominant AI approach. Furthermore, most applications focused on the detection of chemical food safety hazards rather than biological, physical, or general predictive modeling. Notably, this study highlights a lack of AI algorithms utilizing unstructured data, despite its growing relevance in the era of generative AI. Accordingly, future research directions are discussed, particularly the transformative potential of large language models (LLMs) in food safety monitoring and regulatory compliance.