Yaodi Zhu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yaodi Zhu has a total of 64 co-authors. Here are the top co-authors by number of shared publications:
Lijun Zhao
College of Food Science and Technology, Henan Agricultural University, Zhengzhou 450002, PR China; International Joint Laboratory of Meat Processing and Safety in Henan Province, Henan Agricultural University, Zhengzhou 450002, PR China
5 shared publicationsGaiming Zhao
College of Food Science and Technology, Henan Agricultural University, Zhengzhou, PR China
4 shared publicationsMiaoyun Li
College of Food Science and Technology, Henan Agricultural University, Zhengzhou 450002, PR China; International Joint Laboratory of Meat Processing and Safety in Henan Province, Henan Agricultural University, Zhengzhou 450002, PR China
4 shared publicationsDong Liang
College of Food Science and Technology, Henan Agricultural University, Zhengzhou 450002, PR China; International Joint Laboratory of Meat Processing and Safety in Henan Province, Henan Agricultural University, Zhengzhou 450002, PR China
3 shared publicationsLingxia Sun
College of Food Science and Technology, Henan Agricultural University, Zhengzhou 450002, PR China; International Joint Laboratory of Meat Processing and Safety in Henan Province, Henan Agricultural University, Zhengzhou 450002, PR China
3 shared publicationsQiancheng Tu
College of Food Science and Technology, Henan Agricultural University, Zhengzhou 450002, PR China; International Joint Laboratory of Meat Processing and Safety in Henan Province, Henan Agricultural University, Zhengzhou 450002, PR China
3 shared publications
Latest Publications
- Open Access
Advances in Sensing Techniques and Deep Learning for Food Quality Detection: Opportunities, Challenges, and Perspectives from Image Recognition to Multimodal Data Fusion
Food quality and safety inspection increasingly requires advanced solutions because conventional methods are challenged by the complexity of modern food systems. Effective inspection must address product quality, authenticity, and safety, together with process-related risks arising during production, storage, transportation, and distribution. However, existing analytical approaches and deep learning studies often focus on individual food attributes or sensing modalities, providing limited guidance for selecting appropriate models and sensing strategies for specific inspection objectives. This review systematically compares CNNs, RNNs/LSTMs, Transformers, GNNs, GANs, and hybrid architectures, as well as transfer learning, self-supervised learning, contrastive learning, few-shot learning, lightweight networks, edge computing, and multimodal fusion. Beyond predictive performance, we evaluate dataset size and representativeness, sample- and batch-level validation, external validation, data-leakage risks, interpretability, computational requirements, and the maturity of food-specific evidence. The principal contribution is an application-oriented framework linking inspection objectives and food matrices with sensing modalities, model architectures, validation evidence, and deployment conditions. Although Transformer-, GNN-, GAN-, multimodal-, and few-shot-learning-based approaches show substantial potential, many remain at developing, emerging, or prototype stages in food-specific applications. For high-risk targets, including toxicants, allergens, adulterants, and foodborne pathogens, deep learning systems should primarily support rapid screening and decision-making, while safety-critical results require confirmation using validated reference methods. Overall, this review provides a systematic perspective on deep learning for food quality and safety inspection and identifies key priorities for future research and industrial deployment.
Open AccessAdvances in Sensing Techniques and Deep Learning for Food Quality Detection: Opportunities, Challenges, and Perspectives from Image Recognition to Multimodal Data Fusion
Food quality and safety inspection increasingly requires advanced solutions because conventional methods are challenged by the complexity of modern food systems. Effective inspection must address product quality, authenticity, and safety, together with process-related risks arising during production, storage, transportation, and distribution. However, existing analytical approaches and deep learning studies often focus on individual food attributes or sensing modalities, providing limited guidance for selecting appropriate models and sensing strategies for specific inspection objectives. This review systematically compares CNNs, RNNs/LSTMs, Transformers, GNNs, GANs, and hybrid architectures, as well as transfer learning, self-supervised learning, contrastive learning, few-shot learning, lightweight networks, edge computing, and multimodal fusion. Beyond predictive performance, we evaluate dataset size and representativeness, sample- and batch-level validation, external validation, data-leakage risks, interpretability, computational requirements, and the maturity of food-specific evidence. The principal contribution is an application-oriented framework linking inspection objectives and food matrices with sensing modalities, model architectures, validation evidence, and deployment conditions. Although Transformer-, GNN-, GAN-, multimodal-, and few-shot-learning-based approaches show substantial potential, many remain at developing, emerging, or prototype stages in food-specific applications. For high-risk targets, including toxicants, allergens, adulterants, and foodborne pathogens, deep learning systems should primarily support rapid screening and decision-making, while safety-critical results require confirmation using validated reference methods. Overall, this review provides a systematic perspective on deep learning for food quality and safety inspection and identifies key priorities for future research and industrial deployment.
- 17 March 2026
- Published by Elsevier in Food Research International
- Vol. 233 (Pt 1) , 118960
From structural characteristics to metabolic characteristics: Structural morphological differences and proteomic analysis of Clostridium perfringens and its spores
- 17 March 2026
- Published by Elsevier in Food Research International
- Vol. 233 (Pt 1) , 118960
From structural characteristics to metabolic characteristics: Structural morphological differences and proteomic analysis of Clostridium perfringens and its spores
- 16 February 2026
- Published by Springer Nature in Archives of Microbiology
- Vol. 208 (4) , 195
Advances in understanding the Clostridial spore inner membrane
Foodborne Clostridium spores represent a significant group of pathogens that threaten food safety, posing a unique control challenge due to their extreme resistance to conventional processing and high pathogenicity upon germination. The spore inner membrane (SIM), a critical structure enveloping the spore core, serves as the central functional interface governing resistance, dormancy maintenance, and germination regulation. Notably, the composition and organization of the Clostridial SIM exhibit distinct features compared to the well-studied Bacillus model, though some mechanistic understandings are extrapolated from the latter. Critically, the SIM lipids in dormant spores exist in a tightly packed gel phase, which drastically reduces membrane fluidity and constitutes the fundamental physical basis for its exceptional low-permeability barrier. Current debates center on whether key germination receptors are individually embedded within the SIM or assemble into a higher-order “germinosome” complex that may span or sit atop the membrane to coordinate signal perception. This article systematically reviews the research progress on the composition, structure, and functional characteristics of the SIM in foodborne Clostridial spores. It elaborates on the low-permeability barrier and signaling hub formed by its unique phospholipids (e.g., cardiolipin, phosphatidylglycerol) and functional proteins (e.g., Ger family receptors, SpoVA channels). The core roles of the SIM in resisting heat and chemical stress, maintaining core homeostasis, and responding to germination signals are analyzed. Furthermore, key research techniques in this field (e.g., electron microscopy, solid-state NMR, lipidomics / proteomics, gene editing) and their applications are summarized. Finally, current challenges and bottlenecks are outlined, including difficulties in dynamic analysis of the SIM, understanding interspecies mechanistic differences, and developing targeted control applications. This review aims to deepen the understanding of the resistance and germination mechanisms of Clostridial spores and to provide a insight and a potential theoretical basis for developing more targeted spore-control strategies targeting the SIM. Graphical Abstract
- 16 February 2026
- Published by Springer Nature in Archives of Microbiology
- Vol. 208 (4) , 195
Advances in understanding the Clostridial spore inner membrane
Foodborne Clostridium spores represent a significant group of pathogens that threaten food safety, posing a unique control challenge due to their extreme resistance to conventional processing and high pathogenicity upon germination. The spore inner membrane (SIM), a critical structure enveloping the spore core, serves as the central functional interface governing resistance, dormancy maintenance, and germination regulation. Notably, the composition and organization of the Clostridial SIM exhibit distinct features compared to the well-studied Bacillus model, though some mechanistic understandings are extrapolated from the latter. Critically, the SIM lipids in dormant spores exist in a tightly packed gel phase, which drastically reduces membrane fluidity and constitutes the fundamental physical basis for its exceptional low-permeability barrier. Current debates center on whether key germination receptors are individually embedded within the SIM or assemble into a higher-order “germinosome” complex that may span or sit atop the membrane to coordinate signal perception. This article systematically reviews the research progress on the composition, structure, and functional characteristics of the SIM in foodborne Clostridial spores. It elaborates on the low-permeability barrier and signaling hub formed by its unique phospholipids (e.g., cardiolipin, phosphatidylglycerol) and functional proteins (e.g., Ger family receptors, SpoVA channels). The core roles of the SIM in resisting heat and chemical stress, maintaining core homeostasis, and responding to germination signals are analyzed. Furthermore, key research techniques in this field (e.g., electron microscopy, solid-state NMR, lipidomics / proteomics, gene editing) and their applications are summarized. Finally, current challenges and bottlenecks are outlined, including difficulties in dynamic analysis of the SIM, understanding interspecies mechanistic differences, and developing targeted control applications. This review aims to deepen the understanding of the resistance and germination mechanisms of Clostridial spores and to provide a insight and a potential theoretical basis for developing more targeted spore-control strategies targeting the SIM. Graphical Abstract
- 2 February 2026
- Published by Springer Nature in Food Analytical Methods
- Vol. 19 (2) , 96
Europium (III)- Modified Silver Nanoparticles as Fluorescence -SERS Dual-Mode Probe for Specific Identification and Digital Detection of Foodborne Spores
- 2 February 2026
- Published by Springer Nature in Food Analytical Methods
- Vol. 19 (2) , 96
Europium (III)- Modified Silver Nanoparticles as Fluorescence -SERS Dual-Mode Probe for Specific Identification and Digital Detection of Foodborne Spores
- Open Access
Effects of Crushing, Vacuum Nano-Collision, and Steam Explosion on the Flavor and Physical Properties of Solid Spices
Spices play a crucial role in shaping the characteristic flavor of marinated meat products. This study systematically compared the effects of physical crushing, vacuum nano-collision, and steam explosion on the physical and flavor characteristics of star anise and cinnamon. The vacuum nano-collision treatment effectively reduced particle size to below 15 nm, promoting faster flavor release and improving both moisture retention and solubility. Hydrocarbons, alcohols, and aldehydes were identified as the dominant volatile compounds. Among the non-volatile components, crushed cinnamon contained the highest shikimic acid concentration (1510.1 ± 25.45 μg/kg), while star anise treated with vacuum nano-collision reached the highest level of shikimic acid (893.10 ± 23.99 μg/kg). However, the main active components of these two spices did not show significant differences between the two treatment methods. Steam explosion treatment resulted in the lowest levels of both volatile and non-volatile compounds. Flavor profiling and electronic tongue analyses further revealed that the flavor characteristics of the crushed and nano-collision groups were similar, but distinctly different from those obtained with steam explosion. Overall, these results provide new insights into the development of efficient spice processing technologies and offer practical guidance for optimizing flavor quality in marinated meat products.
Open AccessEffects of Crushing, Vacuum Nano-Collision, and Steam Explosion on the Flavor and Physical Properties of Solid Spices
Spices play a crucial role in shaping the characteristic flavor of marinated meat products. This study systematically compared the effects of physical crushing, vacuum nano-collision, and steam explosion on the physical and flavor characteristics of star anise and cinnamon. The vacuum nano-collision treatment effectively reduced particle size to below 15 nm, promoting faster flavor release and improving both moisture retention and solubility. Hydrocarbons, alcohols, and aldehydes were identified as the dominant volatile compounds. Among the non-volatile components, crushed cinnamon contained the highest shikimic acid concentration (1510.1 ± 25.45 μg/kg), while star anise treated with vacuum nano-collision reached the highest level of shikimic acid (893.10 ± 23.99 μg/kg). However, the main active components of these two spices did not show significant differences between the two treatment methods. Steam explosion treatment resulted in the lowest levels of both volatile and non-volatile compounds. Flavor profiling and electronic tongue analyses further revealed that the flavor characteristics of the crushed and nano-collision groups were similar, but distinctly different from those obtained with steam explosion. Overall, these results provide new insights into the development of efficient spice processing technologies and offer practical guidance for optimizing flavor quality in marinated meat products.
- 29 March 2025
- Published by Springer Nature in Analytical and Bioanalytical Chemistry
- Vol. 417 (11) , 2281-2292
Rapid and accurate identification of foodborne bacteria: a combined approach using confocal Raman micro-spectroscopy and explainable machine learning
This study proposes a rapid identification method for foodborne pathogens by combining Raman spectroscopy with explainable machine learning. Spectral data of nine common foodborne pathogens are collected using a laser confocal Raman spectrometer, and their characteristic Raman peaks are identified and analyzed. Key spectral features are extracted using competitive adaptive reweighted sampling (CARS) and the successive projections algorithm (SPA), while t-distributed stochastic neighbor embedding (t-SNE) is employed for visualization. Subsequently, classification models, including support vector machine (SVM) and random forest (RF), are developed, and the optimal model is selected based on classification accuracy (ACC), with the RF model achieving a test accuracy of 98.91%. To enhance the interpretability of the model, Shapley Additive exPlanations (SHAP) analysis is applied to evaluate the contribution of each spectral feature to the classification results, identifying critical Raman shifts significantly influencing pathogen classification. The results demonstrate that CARS-SPA feature selection not only improves the accuracy and efficiency of the classification model but also enhances its transparency and reliability. This study optimizes the workflow for food safety testing, reduces the risk of foodborne diseases, and provides robust technical support for public health and safety.
- 29 March 2025
- Published by Springer Nature in Analytical and Bioanalytical Chemistry
- Vol. 417 (11) , 2281-2292
Rapid and accurate identification of foodborne bacteria: a combined approach using confocal Raman micro-spectroscopy and explainable machine learning
This study proposes a rapid identification method for foodborne pathogens by combining Raman spectroscopy with explainable machine learning. Spectral data of nine common foodborne pathogens are collected using a laser confocal Raman spectrometer, and their characteristic Raman peaks are identified and analyzed. Key spectral features are extracted using competitive adaptive reweighted sampling (CARS) and the successive projections algorithm (SPA), while t-distributed stochastic neighbor embedding (t-SNE) is employed for visualization. Subsequently, classification models, including support vector machine (SVM) and random forest (RF), are developed, and the optimal model is selected based on classification accuracy (ACC), with the RF model achieving a test accuracy of 98.91%. To enhance the interpretability of the model, Shapley Additive exPlanations (SHAP) analysis is applied to evaluate the contribution of each spectral feature to the classification results, identifying critical Raman shifts significantly influencing pathogen classification. The results demonstrate that CARS-SPA feature selection not only improves the accuracy and efficiency of the classification model but also enhances its transparency and reliability. This study optimizes the workflow for food safety testing, reduces the risk of foodborne diseases, and provides robust technical support for public health and safety.
- 13 November 2024
- Published by Elsevier in International Journal of Biological Macromolecules
- Vol. 283 (Pt 1) , 137640
Improving emulsion stability: The role of flaxseed gum in stabilizing bovine bone protein hydrolysates emulsions
- 13 November 2024
- Published by Elsevier in International Journal of Biological Macromolecules
- Vol. 283 (Pt 1) , 137640
Improving emulsion stability: The role of flaxseed gum in stabilizing bovine bone protein hydrolysates emulsions
- Open Access
Effect of Molecular Weight on the Structural and Emulsifying Characteristics of Bovine Bone Protein Hydrolysate
The emulsifying capacity of bovine bone protein extracted using high-pressure hot water (HBBP) has been determined to be good. Nevertheless, given that HBBP is a blend of peptides with a broad range of molecular weights, the distinction in emulsifying capacity between polypeptide components with high and low molecular weights is unclear. Therefore, in this study, HBBP was separated into three molecular weight components of 10–30 kDa (HBBP 1), 5–10 kDa (HBBP 2), and <5 kDa (HBBP 3) via ultrafiltration, and the differences in their structures and emulsifying properties were investigated. The polypeptide with the highest molecular weight displayed the lowest endogenous fluorescence intensity, the least solubility in an aqueous solution, and the highest surface hydrophobicity index. Analysis using laser confocal Raman spectroscopy showed that with an increase in polypeptide molecular weight, the α-helix and β-sheet contents in the secondary structure of the polypeptide molecule increased significantly. Particle size, rheological characteristics, and laser confocal microscopy were used to characterize the emulsion made from peptides of various molecular weights. High-molecular-weight peptides were able to provide a more robust spatial repulsion and thicker interfacial coating in the emulsion, which would make the emulsion more stable. The above results showed that the high-molecular-weight polypeptide in HBBP effectively improved the emulsion stability when forming an emulsion. This study increased the rate at which bovine bone was utilized and provided a theoretical foundation for the use of bovine bone protein as an emulsifier in the food sector.
Open AccessEffect of Molecular Weight on the Structural and Emulsifying Characteristics of Bovine Bone Protein Hydrolysate
The emulsifying capacity of bovine bone protein extracted using high-pressure hot water (HBBP) has been determined to be good. Nevertheless, given that HBBP is a blend of peptides with a broad range of molecular weights, the distinction in emulsifying capacity between polypeptide components with high and low molecular weights is unclear. Therefore, in this study, HBBP was separated into three molecular weight components of 10–30 kDa (HBBP 1), 5–10 kDa (HBBP 2), and <5 kDa (HBBP 3) via ultrafiltration, and the differences in their structures and emulsifying properties were investigated. The polypeptide with the highest molecular weight displayed the lowest endogenous fluorescence intensity, the least solubility in an aqueous solution, and the highest surface hydrophobicity index. Analysis using laser confocal Raman spectroscopy showed that with an increase in polypeptide molecular weight, the α-helix and β-sheet contents in the secondary structure of the polypeptide molecule increased significantly. Particle size, rheological characteristics, and laser confocal microscopy were used to characterize the emulsion made from peptides of various molecular weights. High-molecular-weight peptides were able to provide a more robust spatial repulsion and thicker interfacial coating in the emulsion, which would make the emulsion more stable. The above results showed that the high-molecular-weight polypeptide in HBBP effectively improved the emulsion stability when forming an emulsion. This study increased the rate at which bovine bone was utilized and provided a theoretical foundation for the use of bovine bone protein as an emulsifier in the food sector.
- Open Access
- 14 December 2023
- Published by Elsevier in Food Chemistry: X
- Vol. 21, 101069
Development of fluorescent GO-AgNPs-Eu3+ nanoparticles based paper visual sensor for foodborne spores detection
- 14 December 2023
- Published by Elsevier in Food Chemistry: X
- Vol. 21, 101069
Open AccessDevelopment of fluorescent GO-AgNPs-Eu3+ nanoparticles based paper visual sensor for foodborne spores detection
- Open Access
Effects of Different Bacteriostats on the Dynamic Germination of Clostridium perfringens Spores
Bacteriostats, as chemical substances that inhibit bacterial growth, are widely used in the sterilization process; however, their effects on spindle spores are unclear. In this study, the effects of bacteriostats, including nine commonly used food additives and four detergents, on the growth of Clostridium perfringens spores were investigated. The results showed that 0.07‰ ethylenediaminetetraacetate had a good inhibitory effect on C. perfringens spore growth, and the spore turbidity decreased by 4.8% after incubation for 60 min. Furthermore, 0.3‰ tea polyphenols, 0.8‰ D-isoascorbic acid, and 0.75‰ potassium sorbate promoted leakage of contents during spore germination. Among the four detergents, 5‰ glutaraldehyde solution presented the best inhibitory effect on the growth of C. perfringens spores, and the spore turbidity decreased by 5.6% after incubation for 60 min. Further analysis of the inactivation mechanism of spores by the bacteriostats was performed by comparing the leakage of UV-absorbing substances during germination. The results revealed that bacteriostats could not directly kill the spores, but could inactivate them by inhibiting germination or damaging the spore structure during germination, thus preventing the formation of bacterial vegetative bodies. These findings provide important information and reference for the mechanism underlying the effects of different bacteriostatic agents on spore growth.
Open AccessEffects of Different Bacteriostats on the Dynamic Germination of Clostridium perfringens Spores
Bacteriostats, as chemical substances that inhibit bacterial growth, are widely used in the sterilization process; however, their effects on spindle spores are unclear. In this study, the effects of bacteriostats, including nine commonly used food additives and four detergents, on the growth of Clostridium perfringens spores were investigated. The results showed that 0.07‰ ethylenediaminetetraacetate had a good inhibitory effect on C. perfringens spore growth, and the spore turbidity decreased by 4.8% after incubation for 60 min. Furthermore, 0.3‰ tea polyphenols, 0.8‰ D-isoascorbic acid, and 0.75‰ potassium sorbate promoted leakage of contents during spore germination. Among the four detergents, 5‰ glutaraldehyde solution presented the best inhibitory effect on the growth of C. perfringens spores, and the spore turbidity decreased by 5.6% after incubation for 60 min. Further analysis of the inactivation mechanism of spores by the bacteriostats was performed by comparing the leakage of UV-absorbing substances during germination. The results revealed that bacteriostats could not directly kill the spores, but could inactivate them by inhibiting germination or damaging the spore structure during germination, thus preventing the formation of bacterial vegetative bodies. These findings provide important information and reference for the mechanism underlying the effects of different bacteriostatic agents on spore growth.