南京中医药大学附属南京市中医院,江苏省南京市秦淮区大明路157号,210022
fsyy00237@njucm.edu.cn
收稿:2025-09-17,
修回:2026-01-16,
纸质出版:2026-04-02
移动端阅览
邓茹月,丁康,朱雨欣等.基于大语言模型与潜在类别模型的混合痔中医证型分布特征及危险因素分析——一项多中心真实世界研究[J].中医杂志,2026,67(07):755-763.
DENG Ruyue,DING Kang,ZHU Yuxin,et al.Syndrome Patterns Distribution and Risk Factors of Mixed Hemorrhoids in Traditional Chinese Medicine: A Multicenter Real-world Study Using Large Language Models and Latent Class Analysis[J].Journal of Traditional Chinese Medicine,2026,67(07):755-763.
邓茹月,丁康,朱雨欣等.基于大语言模型与潜在类别模型的混合痔中医证型分布特征及危险因素分析——一项多中心真实世界研究[J].中医杂志,2026,67(07):755-763. DOI: 10.13288/j.11-2166/r.2026.07.010.
DENG Ruyue,DING Kang,ZHU Yuxin,et al.Syndrome Patterns Distribution and Risk Factors of Mixed Hemorrhoids in Traditional Chinese Medicine: A Multicenter Real-world Study Using Large Language Models and Latent Class Analysis[J].Journal of Traditional Chinese Medicine,2026,67(07):755-763. DOI: 10.13288/j.11-2166/r.2026.07.010.
目的
2
基于多中心真实世界诊疗数据,构建混合痔中医证型标准化分类模型,揭示其证型分布规律及核心危险因素,为促进中医辨证规范化与实施精准辨证论治干预提供循证依据。
方法
2
采用多中心横断面研究设计,纳入2023年9月1日至2024年12月31日江苏省8所医院的13 283例混合痔患者。整合DeepSeek-R1-Distill-Qwen-7B及LLaMA-3.3大语言模型与潜在类别分析方法,对中医证候信息进行无监督学习与潜在类别建模。通过单因素分析筛选潜在危险因素,并进一步运用Logistic回归分析确定各证型的独立危险因素。
结果
2
模型各项指标在不同病历段落类型中表现稳定可靠:门诊病历中既往史(F1=99.7%)、现病史(F1=94.9%)、专科检查(F1=90.7%);住院病历中既往史(F1=98.2%)、现病史(F1=91.2%)、专科检查(F1=90.3%)、出院情况(F1=90.6%)。潜在类别模型识别出混合痔四种核心中医证型:脾虚气陷证(915例,6.89%)、湿热下注证(10 820例,81.46%)、气滞血瘀证(1252例,9.43%)及风伤肠络证(296例,2.22%);潜在类别概率分别为
0.069、0.815、0.094、0.022。Logistic回归分析显示,性别、年龄、病程、高血压病、糖尿病、高脂血症、便秘、吸烟史、饮酒史是不同证型分化的独立危险因素(
P
<0.05)。疗效验证评估发现,脾虚气陷证与气滞血瘀证患者的治愈率高于湿热下注证(校正后
P
<0.05),其余证型间差异无统计学意义。
结论
2
湿热下注证为混合痔的主要证型,性别、年龄、病程、高血压病、糖尿病、高脂血症、便秘、吸烟史、饮酒史是不同证型分化的独立危险因素。
Objective
2
To develop a standardized classification model for traditional Chinese medicine (TCM) syndrome patterns of mixed hemorrhoids using multi-center real-world data, and unveil their distribution patterns and core risk factors, thereby providing evidence-based support for standardizing TCM syndrome differentiation and implementing precision interventions.
Methods
2
A multi-center cross-sectional study was conducted, enrolling 13 283 mixed hemorrhoid patients from eight hospitals in Jiangsu Province between September 1st, 2023 and December 31st, 2024. DeepSeek-R1-Distill-Qwen-7B and LLaMA-3.3 large language models (LLM) were integrated with latent class analysis (LCA) to perform unsupervised learning and latent class modeling of TCM symptomatology. Potential risk factors were screened via univariate analysis, followed by logistic regression to identify independent risk factors for each syndrome pattern.
Results
2
The model's performance indicators were stable and reliable across different clinical data types,i.e. in the outpatient records, past medical history (F1=99.7%), current medical history (F1=94.9%), and specialist examination (F1=90.7%); in inpatient records, past medical history (F1=98.2%), current medical history (F1=91.2%), specialist examination (F1=90.3%), and discharge status (F1=90.6%). Latent class mode-ling identified four core TCM syndrome patterns including spleen deficiency and
qi
sinking syndrome (915 cases, 6.89%), damp-heat pouring downward syndrome (10 820 cases, 81.46%),
qi
stagnation and blood stasis syndrome (1252
cases, 9.43%), and wind injuring intestinal collaterals syndrome (296 cases, 2.22%), with respective latent class probabilities of 0.069, 0.815, 0.094, and 0.022. Logistic regression demonstrated that gender, age, disease duration, hypertension, diabetes, hyperlipidemia, constipation, smoking history, and alcohol consumption were independent risk factors for pattern differentiation (
P
<0.05). The efficacy validation evaluation revealed that the cure rates for patients with spleen deficiency and
qi
sinking syndrome and
qi
stagnation and blood stasis syndrome were higher than those for patients with damp-heat pouring downward syndrome (adjusted
P
<0.05), with no statistically significant differences among other syndrome patterns.
Conclusion
2
Damp-heat pouring downward syndrome is the predominant syndrome in mixed hemorrhoids. Gender, age, disease duration, hypertension, diabetes, hyperlipi-demia, constipation, smoking history, and alcohol consumption are independent risk factors for the differentiation of syndrome types.
江维 , 张虹玺 , 隋楠 , 等 . 中国城市居民常见肛肠疾病流行病学调查 [J]. 中国公共卫生 , 2016 , 32 ( 10 ): 1293 - 1296 .
XIONG YY , LIU H . Application intention and influen-cing factors of traditional Chinese medicine in patients with mixed hemorrhoids: a cross-sectional study in China [J]. Patient Prefer Adher , 2025 , 19 ( 1 ): 1673 - 1680 .
郭呈华 , 车晓玉 , 林芝 , 等 . 中国某健康体检人群痔的流行病学特征 [J]. 北京大学学报(医学版) , 2024 , 56 ( 5 ): 815 - 819 .
SUN XW , XU JY , ZHU CZ , et al . Analysis of factors impacting postoperative pain and quality of life in patients with mixed hemorrhoids: a retrospective study [J]. World J Gastrointest Surg , 2024 , 16 ( 3 ): 731 - 739 .
中国中西医结合学会大肠肛门病专业委员会 . 中国痔病诊疗指南(2020) [J]. 结直肠肛门外科 , 2020 , 26 ( 5 ): 519 - 533 .
杨涛 , 朱学芳 . 中医辨证智能化研究现状及发展趋势 [J]. 南京中医药大学学报 , 2021 , 37 ( 4 ): 597 - 601 .
王琪 , 胡良平 , 高颖 , 等 . 潜在类别分析在缺血性中风病患者证候数据中的应用 [J]. 中国卫生统计 , 2013 , 30 ( 5 ): 638 - 640 .
李伟珂 , 伊明洋 , 倪园园 , 等 . 基于潜在类别结合隐结构模型的慢性阻塞性肺疾病肺气虚证分级量化诊断研究 [J]. 中医杂志 , 2025 , 66 ( 7 ): 710 - 716 .
陈子佳 , 彭文茜 , 张德政 , 等 . 大语言模型在中医药领域的应用、挑战与前景 [J]. 协和医学杂志 , 2025 , 16 ( 1 ): 83 - 89 .
李欣桐 , 马素芬 , 张丰聪 , 等 . 中医药领域大语言模型的研究进展与应用前景 [J]. 南京中医药大学学报 , 2024 , 40 ( 12 ): 1393 - 1403 .
陈子佳 , 韩宇 , 刘欣 , 等 . 基于大模型的肺癌表皮生长因子受体突变患者生存预测模型构建与验证 [J]. 世界中医药 , 2025 , 20 ( 8 ): 1379 - 1390 .
智从从 , 李雪 , 程一乘 , 等 . 痔中西医结合诊疗指南(2025版) [J]. 中国全科医学 , 2025 , 28 ( 26 ): 3217 - 3228 .
赵雪 , 张海 , 王东波 . 大语言模型评测研究现状、应用、问题与趋势分析 [J]. 情报学报 , 2025 , 44 ( 8 ): 1058 - 1074 .
谢珉宁 , 徐璇 , 张磊 , 等 . 混合痔中医辨证分型与流行病学特点的相关性研究 [J]. 江苏中医药 , 2016 , 48 ( 11 ): 38 - 40 .
MCCUTCHEON AL . Basic concepts and procedures in single- and multiple-group latent class analysis [M]. Cambridge : Cambridge University Press , 2002 : 56 - 86 .
SCOTTO RN , BAER JC . Latent class analysis: a method for capturing heterogeneity [J]. Soc Work Res , 2012 , 36 ( 1 ): 61 - 69 .
曾宪华 , 肖琳 , 张岩波 . 潜在类别分析原理及实例分析 [J]. 中国卫生统计 , 2013 , 30 ( 6 ): 815 - 817 .
乔敬华 , 何佳伟 , 周军惠 . 基于流行病学调查的农村社区居民痔病中医药防治对策探讨 [J]. 上海中医药杂志 , 2019 , 53 ( 6 ): 14 - 19 .
蓝可彰 , 潘海涛 , 罗建琼 , 等 . 广西忻城县546例混合痔患者的临床分析 [J]. 蛇志 , 2017 , 29 ( 1 ): 17 - 18 .
刘建乔 , 李宇栋 , 李雪 , 等 . 139例粪便潜血阳性混合痔患者结肠镜检查结果与中医证型的回顾性分析 [J]. 北京中医药 , 2022 , 41 ( 2 ): 207 - 209 .
蔡华花 , 郑梅娟 , 曾凤荣 , 等 . 120例闽北地区混合痔中医证素规律研究 [J]. 中国卫生标准管理 , 2021 , 12 ( 15 ): 110 - 113 .
康健 , 余乐来 , 龚光明 . 500例混合痔患者的中医体质研究 [J]. 湖北中医杂志 , 2015 , 37 ( 9 ): 20 - 21 .
朱谦让 , 戴月 , 谢玮 , 等 . 江苏省居民膳食结构与营养素摄入状况评价 [J]. 江苏预防医学 , 2017 , 28 ( 3 ): 259 - 261,265 .
AVSAR AF , KESKIN HL . Haemorrhoids during pregnancy [J]. J Obstet Gynaecol , 2010 , 30 ( 3 ): 231 - 237 .
梁建民 . 女性肛肠病的临床发病特点及治疗分析(附600例报告) [J]. 中国肛肠病杂志 , 2007 , 27 ( 7 ): 2 .
徐红 , 黄卫国 , 郭淳 . 1795例常规体检者中痔病患病情况分析 [J]. 中国普外基础与临床杂志 , 2011 , 18 ( 8 ): 884 - 886 .
HONG J , KIM I , SONG J , et al . Socio-demographic factors and lifestyle associated with symptomatic hemorrhoids: big data analysis using the National Health insurance Service-National Health screening cohort(NHIS-HEALS) database in Korea [J]. Asian J Surg , 2022 , 45 ( 1 ): 353 - 359 .
DELCÒ F , SONNENBERG A . Associations between hemorrhoids and other diagnoses [J]. Dis Colon Rectum , 1998 , 41 ( 12 ): 1534 - 1541 .
HONG YS , JUNG KU , RAMPAL S , et al . Risk factors for hemorrhoidal disease among healthy young and middle-aged Korean adults [J]. Sci Rep , 2022 , 12 ( 1 ): 129 .
杜勇军 , 黄德铨 , 杨洁 , 等 . 混合痔危险因素的病例对照研究 [J]. 重庆医学 , 2017 , 46 ( 32 ): 4553 - 4555 .
张皛 , 王剑 . 赣南地区痔病的危险因素 [J]. 中国老年学杂志 , 2015 , 35 ( 22 ): 6554 - 6555 .
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