中药质量标志物（quality marker，Q-Marker）是指存在于中药中，与中药的质量、药效、安全性密切相关，能够反映中药品质特征、区分中药真伪优劣、追溯中药来源，且可被精准检测的化学物质或特征组合，是中药质量控制与评价的核心指标。传统中药质量控制多依赖经验判断与单一指标成分，存在筛选效率低、主观性强、难以兼顾“化学特异性-药效关联性-整体协调性”的局限，难以适配中药“多成分、多靶点、整体调节”的核心特点。随着人工智能（artificial intelligence，AI）技术的快速发展，机器学习、深度学习、知识图谱等技术凭借强大的多源数据融合、高维特征挖掘与非线性关联建模能力，逐步渗透到中药Q-Marker研究的全流程，实现了从“经验驱动”向“数据驱动”的转变，显著提升了Q-Marker筛选的精准度、效率与科学性。系统梳理AI驱动的中药Q-Marker研究的多源数据基础、AI预处理技术、核心算法、研究范式，分析当前研究面临的挑战并展望未来发展方向，为中药Q-Marker的智能化研究与标准化应用提供参考。

Quality marker (Q-Marker) of traditional Chinese medicine (TCM) refers to a chemical substance or characteristic combination that exists in TCM, which is closely related to its quality, efficacy and safety, can reflect its quality characteristics, distinguish genuine and inferior products, trace its origin, and be accurately detected. It serves as the core indicator for TCM quality control and evaluation. Traditional TCM quality control mostly relies on empirical judgment and single-component chemical analysis, which suffers from low efficiency, strong subjectivity, and difficulty in balancing chemical specificity, efficacy correlation and overall coordination. Such limitations make it incompatible with the core characteristics of TCM featuring multi-components, multi-targets and holistic regulation. With the rapid development of artificial intelligence (AI), technologies including machine learning, deep learning and knowledge graphs have been gradually applied in the whole process of TCM Q-Marker research due to their powerful capabilities in multi-source data fusion, high-dimensional feature mining and nonlinear correlation modeling, realizing the transformation from “experience-driven” to “data-driven” and significantly improving the accuracy, efficiency and scientificity of Q-Marker screening. This paper systematically reviews the multi-source data basis, AI preprocessing techniques, core algorithms and research paradigms of AI-driven Q-Marker research in TCM, analyzes the current challenges and discusses future development directions, so as to provide references for the intelligent research and standardized application of TCM Q-Marker.

R28;TP18

国家自然科学基金面上项目（81973625）；河南省科技研发计划联合基金（242301420114，252301420095）；河南省科技攻关项目（232102311221，262102311235）；河南省中医药科学研究专项（2024ZY2011）；河南省医学科技攻关项目（LHGJ20250412）