Abstract: As an important part of food quality management, the detection of food hazard factors and safety risk prediction have always been a research hotspot. In order to better carry out related work in the future, this article reviews research progress on conventional methods for the detection of food hazard factors, and introduces new methods for the detection of food hazard factors such as nanozymes, clustered regularly interspaced short palindromic repeats/CRISPR-associated proteins (CRISPR/Cas) sensors and nuclear magnetic resonance (NMR) spectroscopy and the application of new materials such as magnetic covalent organic frameworks in this field. Then, it summarizes the application of subjective weighting, machine learning and deep learning in food safety risk prediction. Finally, the advantages and disadvantages of each method are analyzed, and future research directions are discussed.

Key words: food safety; food detection; nanozyme; nuclear magnetic resonance; risk prediction; deep learning