【目的】 基于个性化知识图谱分析和数据挖掘技术，探讨广东省名中医司徒仪教授辨治子宫内膜异位症（endometriosis， EMs）规律，为EMs的中医诊疗提供参考。【方法】 基于EMs诊治的文献研究基础，采用人机结合方法建立名医理法方药本体 知识树；采集司徒仪教授于2013年至2019年在广东省中医院门诊诊治的EMs患者的病历，对采集的病案信息进行规范化处 理后，形成知识集，构建EMs的中医药诊疗知识图谱，分析疾病特征与用药规律。【结果】 EMs患者的核心症状群为痛经、慢 性盆腔痛、不孕，证候以血瘀为核心，气滞、肾虚、气虚为主要病机。频繁项集分析得到的核心药物组合有“延胡索-香 附-白芍”“枸杞子-桑寄生-茯苓-山药”“枳壳-益母草”等。基于知识元聚类和神经网络算法而构建的EMs词云图显示，丹 参-赤芍、延胡索-郁金、桂枝-牡丹皮-赤芍-茯苓、丹参-赤芍-三棱-莪术、川芎-延胡索、桑寄生-菟丝子等为司徒仪教授 治疗 EMs的常用药物组合。有关中药、症状、证型之间的层次聚类分析和关联网络构建结果显示，证型、症状、用药间存 在复杂的内在联系。【结论】 司徒仪教授治疗EMs以活血化瘀、补肾为主要治疗方法，同时顺应月经周期，攻补兼施。基于语义 网络的知识图谱分析与数据挖掘有助于将司徒仪教授的临床诊治经验的隐性知识显性化，可为EMs的中医辨治提供借鉴。

Objective To explore the treatment for endometriosis（EMs）by Professor SITU Yi， a Guangdong famous traditional Chinese medicine physician，thus to provide reference for the diagnosis and treatment of EMs in traditional Chinese medicine based on personalized knowledge graph analysis and data mining technology. Methods Based on the literature research achievements of diagnosis and treatment of EMs，the ontology knowledge tree of famous physician’s principles，methods，prescriptions and medicines was established by user-computer integration method. The medical records of EMs patients treated by Professor SITU Yi in the outpatient department of Guangdong Provincial Hospital of Chinese Medicine from 2013 to 2019 were collected. After standardization of the data of medical records，the knowledge set was defined，the knowledge map of TCM diagnosis and treatment of EMs was constructed，and the symptom characteristics and medication rules for EMs were analyzed. Results The core symptoms of EMs patients were dysmenorrhea， chronic pelvic pain and infertility， with the predominated syndrome of blood stasis. The pathogenesis of EMs patients was mainly due to qi stagnation，kidney deficiency， and qi deficiency. Frequent itemset analysis yielded the core drug groups of“Corydalis Rhizoma，Cyperi Rhizoma， Paeoniae Radix Alba”“Lycii Fructus，Taxilli Herba，Poria，Dioscoreae Rhizoma” and “Aurantii Fructus， Leonuri Herba”. The EMs word cloud map constructed on the basis of knowledge element clustering and neural network algorithm showed that the drug combinations of “Salviae Miltiorrhizae Radix et Rhizoma，Paeoniae Radix Rubra”“Corydalis Rhizoma，Curcumae Radix”“Cinnamomi Ramulus，Moutan Cortex，Paeoniae Radix Rubra， Poria”“Salviae Miltiorrhizae Radix et Rhizoma， Paeoniae Radix Rubra， Sparganii Rhizoma， Curcumae Rhizoma”“Chuanxiong Rhizoma，Corydalis Rhizoma” and “Taxilli Herba，Cuscutae Semen” were commonly used for the treatment of EMs by Professor SITU Yi. The results of hierarchical clustering analysis and association network construction for Chinese medicinals，symptoms and syndromes showed that there were complex inner link among the syndromes， symptoms and medication. Conclusion For the treatment of EMs， Professor SITU Yi usually adopt the methods of activating blood to remove stasis and tonifying kidney，and also performs the periodic therapy and simultaneous application of purging and nourishing therapeutics. Knowledge graph analysis and data mining based on semantic web are helpful to unveil the tacit knowledge of clinical diagnosis and treatment experience of Professor SITU Yi，which can provide reference for the differentiation and treatment of EMs with Chinese medicine.

R271.971

广东省自然科学基金项目（编号：2023A1515011109）；广东省基础与应用基础研究基金省市联合基金（粤穗联合基金）-青年基金 项目（编号：2022A1515110697）；广东省中医药局科研基金项目（编号：20231018）；广州市科技局-青年基金项目（编号： 2023A04J1130）；广东省哲学社会科学规划项目（编号：GD20LMZTS05）；广东省名老中医药专家传承工作室建设项目（国中医 药人教函［2018］134 号）