江苏省中医院/南京中医药大学附属医院,江苏省南京市秦淮区汉中路155号,210029
zhulei5100@njucm.edu.cn
收稿:2025-09-23,
修回:2025-12-26,
录用:2025-12-18,
纸质出版:2026-04-17
移动端阅览
沈照峰,张梦媛,朱磊等.基于多模态人工智能技术的中西医结合炎症性肠病大数据平台及专病队列的构建[J].中医杂志,2026,67(08):888-895.
SHEN Zhaofeng,ZHANG Mengyuan,ZHU Lei,et al.Construction of a Multimodal AI-Based Integrated Traditional Chinese and Western Medicine Big Data Platform for Inflammatory Bowel Disease with Specialized Disease Cohort[J].Journal of Traditional Chinese Medicine,2026,67(08):888-895.
沈照峰,张梦媛,朱磊等.基于多模态人工智能技术的中西医结合炎症性肠病大数据平台及专病队列的构建[J].中医杂志,2026,67(08):888-895. DOI: 10.13288/j.11-2166/r.2026.08.014.
SHEN Zhaofeng,ZHANG Mengyuan,ZHU Lei,et al.Construction of a Multimodal AI-Based Integrated Traditional Chinese and Western Medicine Big Data Platform for Inflammatory Bowel Disease with Specialized Disease Cohort[J].Journal of Traditional Chinese Medicine,2026,67(08):888-895. DOI: 10.13288/j.11-2166/r.2026.08.014.
目的
2
构建基于多模态人工智能(AI)技术的中西医结合大数据平台与专病队列,为开展高质量中医药循证研究提供方法学支撑与数据基础。
方法
2
以炎症性肠病(IBD)为研究对象,参考国内外专病队列数据库构建模式,依托江苏省中医优势病种诊疗大数据平台,制定中医优势病种研究核心数据集,整合江苏省11家三级甲等中医院多源异构数据,搭建中西医结合IBD医疗大数据平台;应用自然语言处理(NLP)、机器学习等AI技术,对真实世界数据进行深度治理(包括抽取、转换、脱敏、加密、清洗、映射、人工标注),实现专病数据的标准化、结构化、归一化;建立动态患者筛选入组、数据采集与随访管理体系,形成IBD专病队列;采用光学字符识别(OCR)结合霍夫变换直线提取法实现肠镜报告文本识别与提取,利用卷积神经网络(CNN)模型对舌象、肠镜图片进行特征分析与量化。
结果
2
初步制定了中西医结合IBD专病数据元标准,完成了36 890条IBD全量诊疗记录数据治理,实现了中西医数据的标准化映射、整合与质控,构建了多模态IBD大数据平台(14 424人次);建立了包含1781例IBD患者的专病队列;开发了AI辅助的中医预测模型。
结论
2
形成了一套可推广的中西医结合多模态IBD大数据平台及专病队列建设技术框架,实现了中西医真实世界数据的深度融合,可为中医药循证研究提供高质量的数据支撑。
Objective
2
To develop an integrated traditional Chinese and western medicine big data platform with specialized disease cohort based on multimodal real-world data (RWD), providing methodological support and data foundation for high-quality evidence-based traditional Chinese medicine (TCM) research.
Methods
2
Following the domestic and international construction models for specialized cohort database, a core dataset for TCM-dominant disease, that is inflammatory bowel disease (IBD), was developed relying on the Jiangsu TCM Dominant Diseases Diagnosis & Treatment Data Platform, which integrated multi-source heterogeneous data from 11 tertiary grade-A TCM hospitals in Jiangsu province, and an integrated traditional Chinese and western medicine big data platform for IBD was constructed thereafter. Leveraging artificial intelligence technologies including natural language processing (NLP) and machine learning, RWD is deeply governed through extraction, transformation, desensitization, encryption, cleaning, mapping, and manual annotation to achieve the standardization, structuring, and normalization of patient-specific data. A dynamic management system was developed for patient screening, data collection, and follow-up to form an IBD-specific cohort. Furthermore, optical character recognition (OCR) combined with the Hough transform line extraction method was used to recognize and extract texts from colonoscopy reports, and convolutional neural network (CNN) models were employed for feature analysis and quantification of tongue images and colonoscopy phothos.
Results
2
A preliminary standard for the integrated traditional Chinese and western medicine IBD-specific disease metadata was developed, which processed 36
890 complete IBD diagnosis and treatment records, achieving standardized mapping, integration and quality control of TCM and western medicine data. A multi-modal IBD big data platform was established with 14,424 cases, and a specialized disease cohort was constructed including the full lifecycle management of 1781 IBD patients. Additionally, an AI-assisted TCM prediction model has been developed.
Conclusion
2
The research has established a replicable technical framework for building a multimodal integrated traditional Chinese and western medicine big data platform and disease cohort for IBD, which achieves the deep integration of RWD from both TCM and western medicine, highligh-ting TCM's unique advantages, and providing high-quality data support for future evidence-based studies in TCM.
DART R J , ZLATAREVA I , VANTOUROUT P , et al . Conserved γδ T cell selection by BTNL proteins limits progression of human inflammatory bowel disease [J]. Science , 2023 , 381 ( 6663 ): eadh0301 .
叶放 , 周学平 , 程海波 , 等 . 基于整体观理解和创新中医病机辨证新理论 [J]. 南京中医药大学学报 , 2017 , 33 ( 5 ): 436 - 439 .
陈赛娟 . 发展重大疾病专病数据库意义重大 [J]. 科技创新与品牌 , 2022 ( 3 ): 20 .
IACUCCI M , MAEDA Y , SANTACROCE G , et al . AI-enabled 'endo-histo-omics': breaking down intestinal barriers in IBD [J]. Nat Rev Gastroenterol Hepatol , 2025 , 22 ( 6 ): 362 - 363 .
吴敏敏 , 王鑫钰 , 王伟炳 . 医学人工智能的技术发展和场景应用 [J]. 复旦学报(医学版) , 2025 , 52 ( 3 ): 470 - 474 .
HOLTEN K I A , BERNKLEV T , OPHEIM R , et al . Fatigue in patients with newly diagnosed inflammatory bowel disease: results from a prospective inception cohort, the IBSEN Ⅲ study [J]. J Crohns Colitis , 2023 , 17 ( 11 ): 1781 - 1790 .
RAFFALS L E , SAHA S , BEWTRA M , et al . The deve-lopment and initial findings of a study of a prospective adult research cohort with inflammatory bowel disease (SPARC IBD) [J]. Inflamm Bowel Dis , 2022 , 28 ( 2 ): 192 - 199 .
DARBANI B , BRODERSEN T , LILJENSE A , et al . Cohort profile: Copenhagen hospital biobank-chronic inflammatory disease-inflammatory bowel disease (CHB-CID:IBD) genetic cohort [J]. Eur J Epidemiol , 2025 , 40 ( 6 ): 721 - 734 .
冯雨萱 , 郑锐 , 于广军 . 我国临床专病数据库建设情况概述 [J]. 上海医药 , 2024 , 45 ( 9 ): 10 - 13,18 .
吴晓萍 , 郑涛 , 房静远 , 等 . 炎症性肠病队列数据库的建设和应用 [J]. 胃肠病学 , 2023 , 28 ( 4 ): 237 - 242 .
REN Y , JIA Y L , LIANG W X , et al . Best practice for developing integrative Chinese-western medicine databases using electronic health records [J]. World J Integr Tradit West Med , 2025 , 11 ( 3 ): 174 - 186 .
VON ELM E , ALTMAN D G , EGGER M , et al . The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for repor-ting observational studies [J]. Ann Intern Med , 2007 , 147 ( 8 ): 573 - 577 .
HARRISON J E , WEBER S , JAKOB R , et al . ICD-11: an international classification of diseases for the twenty-first century [J]. BMC Med Inform Decis Mak , 2021 , 21 ( Suppl 6 ): 206 .
中华人民共和国国家卫生健康委员会 . 卫生健康信息数据集元数据标准 [EB/OL].( 2023-04-01 )[ 2025-11-20 ]. https://www.nhc.gov.cn/cms-search/downFiles/b4e61d83c6014b04a556810f04fce3a8.pdf https://www.nhc.gov.cn/cms-search/downFiles/b4e61d83c6014b04a556810f04fce3a8.pdf .
国家市场监督管理总局 , 中国国家标准化管理委员会 . 信息安全技术 健康医疗数据安全指南 [EB/OL].( 2020-12-14 )[ 2025-11-20 ]. https://openstd.samr.gov.cn/bzgk/gb/newGbInfo?hcno=239351905E7B62A7DF537856738247CE https://openstd.samr.gov.cn/bzgk/gb/newGbInfo?hcno=239351905E7B62A7DF537856738247CE .
中华医学会消化病学分会炎症性肠病学组 , 中国炎症性肠病诊疗质量控制评估中心 . 中国溃疡性结肠炎诊治指南(2023年·西安) [J]. 胃肠病学 , 2024 , 29 ( 3 ): 145 - 173 .
中华医学会消化病学分会炎症性肠病学组 , 中国炎症性肠病诊疗质量控制评估中心 . 中国克罗恩病诊治指南(2023年·广州) [J]. 胃肠病学 , 2024 , 29 ( 4 ): 200 - 235 .
中华中医药学会脾胃病分会 . 溃疡性结肠炎中医诊疗专家共识(2023) [J]. 中华中医药杂志 , 2024 , 39 ( 1 ): 288 - 296 .
中华中医药学会脾胃病分会 . 克罗恩病中医诊疗专家共识意见(2024) [J]. 中医杂志 , 2024 , 65 ( 16 ): 1734 - 1740 .
章萌 , 沈鹏 , 刘志科 , 等 . 观测性医疗结果合作组织通用数据模型在中国的应用研究现状 [J]. 中华流行病学杂志 , 2025 , 46 ( 5 ): 907 - 913 .
朱燕波 . 《中医体质量表》应用中的问题及其使用规范 [J]. 中华中医药杂志 , 2022 , 37 ( 9 ): 5066 - 5070 .
TURNER D , RICCIUUTO A , LEWIS A , et al . STRIDE-Ⅱ: an update on the selecting therapeutic targets in inflammatory bowel disease (STRIDE) initiative of the international organization for the study of IBD (IOIBD): determining therapeutic goals for treat-to-target strategies in IBD [J]. Gastroenterology , 2021 , 160 ( 5 ): 1570 - 1583 .
董志威 , 王钧岚 , 谢韬 , 等 . 基于优劣解距离算法联合熵值法的针刺治疗前列腺癌术后尿失禁的循证决策研究 [J]. 中医杂志 , 2024 , 65 ( 23 ): 2434 - 2441 .
国家市场监督管理总局 , 国家标准化管理委员会 . 健康信息学 数据交换标准 HL 7临床文档架构(版本 2 )[EB/OL].( 2023-03-17 )[ 2025-12-20 ]. https://openstd.samr.gov.cn/bzgk/gb/newGbInfo?hcno=A358014699D2D48295C9C529FBB70F3A https://openstd.samr.gov.cn/bzgk/gb/newGbInfo?hcno=A358014699D2D48295C9C529FBB70F3A .
张林 , 赵英杰 , 陈兴 . 观测指标标识符逻辑命名与编码系统(LOINC)数据库介绍 [J]. 河北省科学院学报 , 2004 ( 4 ): 66 - 68,71 .
AHMADI N , PENG Y , WOLFIEN M , et al . OMOP CDM can facilitate data-driven studies for cancer prediction: a systematic review [J]. Int J Mol Sci , 2022 , 23 ( 19 ): 11834 .
CHENG A C , FACILE R , OWEN J , et al . Making clinical data acquisition standards harmonization (CDASH) electronic case report forms available on the REDCap shared data instrument library [J]. J Soc Clin Data Manag , 2022 , 2 ( 3 ): 7 .
GARZA M , DEL FIOL G , TENENBAUM J , et al . Eva-luating common data models for use with a longitudinal community registry [J]. J Biomed Inform , 2016 , 64 ( 12 ): 333 - 341 .
刘海 , 张瞩熹 , 任雯 , 等 . 面向异构数据源的分布式集成工具研究与设计 [J]. 计算机应用研究 , 2020 , 37 ( S1 ): 204 - 206 .
王传立 , 李金才 , 吴移桃 . Struts与Web Service集成模式研究与应用 [J]. 计算机工程与科学 , 2011 , 33 ( 8 ): 181 - 184 .
姜维宁 , 李晓明 , 栾兵 , 等 . 基于提高临床科室管理能力的医院信息集成平台的建立及应用效果 [J]. 中国卫生统计 , 2023 , 40 ( 6 ): 869 - 871,874 .
STANLEY J , RABOT E , REDDY S , et al . Large language models deconstruct the clinical intuition behind diagnosing autism [J]. Cell , 2025 , 188 ( 8 ): 2235 - 2248 .
张炫 , 杨艳 . 基于改进随机霍夫变换的直线提取算法 [J]. 信息技术 , 2017 ( 12 ): 6 - 9 .
GAO C , ZHU Q , WANG P , et al . Structured multi-modal attentions for TextVQA [J]. IEEE Trans Pattern Anal Mach Intell , 2022 , 44 ( 12 ): 9603 - 9614 .
MATSUMOTO T , HISAMATSU T , ESAKI M , et al . Guidelines for endoscopic diagnosis and treatment of inflammatory bowel diseases [J]. Dig Endosc , 2025 , 37 ( 4 ): 319 - 351 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010102007558号