中药质量影响其临床应用的安全性和有效性，随着中药产业的快速发展，传统质量控制方法难以满足复杂多变的中药质量检测需求。近年来，人工智能（artificial intelligence，AI）凭借其强大的数据处理和模式识别能力，能够建立预测模型以高效处理中药多源异构数据（如光谱、色谱、图像及文本信息），实现质量指标的智能预测与异常检测，为中药质量控制提供了新的思路和方法。通过对AI在中药质量中的应用进行综述，系统阐述了机器学习及深度学习的算法在中药质量控制研究中的研究进展，分析了目前面临的问题及挑战，以期进一步提升中药质量控制的智能化水平，为中药产业的高质量发展提供有力支持。

The quality of traditional Chinese medicine (TCM) directly affects the safety and effectiveness of its clinical application. With the rapid development of the TCM industry, conventional quality control methods have become inadequate to meet the complex and evolving requirements of TCM quality testing. In recent years, artificial intelligence (AI), characterized by its powerful data processing and pattern recognition capabilities, has enabled the establishment of predictive models for efficiently handling multi-source heterogeneous data in TCM (such as spectra, chromatograms, images, and text information). This enables intelligent prediction and anomaly detection of quality indicators, and provides novel approaches and methodologies for TCM quality control. This paper systematically reviews the application progress of machine learning and deep learning algorithms in TCM quality research, analyzes the current problems and challenges, and aims to further improve the intelligence level of TCM quality control, providing robust support for the high-quality development of TCM industry.

R28

国家重点研发计划项目（2022YFC3501601，2023YFC3504101）