T/XZBX 0072-2025 Quality control procedures for intelligent algorithms for food microbiological detection based on big data

Scope

Food microbiological detection is a core step in ensuring food safety and public health. Traditional methods mainly rely on manual operations in laboratories and standardized cultivation and detection processes. However, with the increasing complexity of the food industry chain, the rapid growth in the number of detection samples, and the ever-increasing demands for timeliness in risk monitoring and response, traditional detection methods face bottlenecks in terms of accuracy, efficiency, and coverage. In recent years, with the development of big data technology and the rapid iteration of artificial intelligence algorithms, intelligent microbiological identification and detection assistance systems have gradually been applied in the field of food safety. Intelligent algorithms can identify, judge, predict, and verify the results of the microbiological detection process through deep learning of multi-source information such as historical detection data, image data, and species characteristics, effectively improving detection speed and consistency. However, due to sample heterogeneity, poor transparency of algorithm models, and uneven quality of training data, the controllability, stability, and evaluation mechanisms of intelligent detection algorithms themselves have not yet established unified standards. To ensure the reproducibility, accuracy, and review traceability of big data-based food microbiological detection intelligent algorithms in actual detection processes, this document is formulated. This document clarifies the key technical points and standard processes of intelligent detection algorithms in quality control from aspects such as data collection and preprocessing, algorithm modeling and training, quality indicator setting and evaluation, and system verification and continuous improvement, and is applicable to relevant units engaged in the research and development, deployment, operation, and supervision of food microbiological intelligent detection sys

Development Information

Related Standards