Please wait a minute...
Advanced Search
数据分析与知识发现  2022, Vol. 6 Issue (1): 122-133     https://doi.org/10.11925/infotech.2096-3467.2021.0409
     研究论文 本期目录 | 过刊浏览 | 高级检索 |
融合多自然语言处理任务的中医辅助诊疗方案研究——以糖尿病为例*
张玉洁1,白如江1(),许海云2,韩靖3,赵梦梦1
1山东理工大学信息管理研究院 淄博 255000
2山东理工大学管理学院 淄博 255000
3山东中医药高等专科学校 烟台 264199
Assisted TCM Diagnosis and Treatment for Diabetes with Multi NLP Tasks
Zhang Yujie1,Bai Rujiang1(),Xu Haiyun2,Han Jing3,Zhao Mengmeng1
1Institute of Information Management, Shandong University of Technology, Zibo 255000, China
2School of Management, Shandong University of Technology, Zibo 255000, China
3Shandong College of Traditional Chinese Medicine, Yantai 264199, China
全文: PDF (1339 KB)   HTML ( 26
输出: BibTeX | EndNote (RIS)      
摘要 

【目的】 为中医诊断治疗提供更精准、更智能的辅助参考,减少中医诊疗过程中的不确定性和难以量化的问题。【方法】 以中医糖尿病医案数据为例,设计一种融合多自然语言处理任务的中医辅助诊疗方案,该方案融合情感识别、文本匹配等多自然语言处理任务序列,实现糖尿病的病情量化评估、症状信息匹配、症状自动摘要、病症类型判别和中药推荐,并进行实证研究。【结果】 以模糊综合评价法进行10轮测试评估,在4项评价指标的平均隶属度分别为0.1949、0.3140、0.2173、0.2738,最大隶属度结果显示该方法效果为良好。【局限】 临床医案数据稀缺,因此各子任务性能难以实现大幅度提升;未扩展到其他领域进行适应性检测。【结论】 该方法在实际使用过程中能够有效地帮助医生解决诊断的不确定性问题,较好地辅助医生形成明确的诊疗结论,具有较为新颖的学术价值和重要的应用价值。

服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
张玉洁
白如江
许海云
韩靖
赵梦梦
关键词 中医诊疗自然语言处理中医糖尿病辅助决策    
Abstract

[Objective] This paper tries to provide more accurate and intelligent auxiliary reference for the diagnosis and treatment of Traditional Chinese Medicine (TCM), aiming to reduce their uncertainty and difficult to quantify issues. [Methods] First, we collected medical records of TCM for diabetes. Then, we created an auxiliary diagnosis and treatment scheme integrating multiple NLP tasks, i.e., emotion recognition and text matching. Finally, we examined our new model with the quantitative assessment of diabetes, symptom information matching, automatic symptom summarization, disease type discrimination and TCM recommendation. [Results] We conducted ten rounds of tests with the fuzzy comprehensive evaluation method. The average membership degrees of the four evaluation indices were 0.194 9, 0.314 0, 0.217 3 and 0.273 8 respectively. The maximum membership degree indicated the effectiveness of the proposed method. [Limitations] Due to the scarcity of clinical medical records, it is difficult to improve the performance of each subtask significantly. More research is needed to examine the model with data from other fields. [Conclusions] This method can effectively help doctors reduce the uncertainty as well as evaluate diagnosis and treatment.

Key wordsTCM Diagnosis &    Treatment    NLP    Diabetes Mellitus in TCM    Assistant Decision Making
收稿日期: 2021-04-26      出版日期: 2022-02-22
ZTFLH:  TP391  
基金资助:*本文系山东省社会科学规划研究一般项目的研究成果之一(21CTQJ11)
通讯作者: 白如江,ORCID:0000-0003-3822-8484     E-mail: brj@sdut.edu.cn
引用本文:   
张玉洁, 白如江, 许海云, 韩靖, 赵梦梦. 融合多自然语言处理任务的中医辅助诊疗方案研究——以糖尿病为例*[J]. 数据分析与知识发现, 2022, 6(1): 122-133.
Zhang Yujie, Bai Rujiang, Xu Haiyun, Han Jing, Zhao Mengmeng. Assisted TCM Diagnosis and Treatment for Diabetes with Multi NLP Tasks. Data Analysis and Knowledge Discovery, 2022, 6(1): 122-133.
链接本文:  
https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/10.11925/infotech.2096-3467.2021.0409      或      https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/Y2022/V6/I1/122
主题 数量 学科 占比
数据挖掘 1 778 中医学 66.53%
用药规律 912 计算机软件及计算机应用 13.88%
用药规律研究 303 中药学 8.56%
数据挖掘技术 194 自动化技术 2.43%
名老中医 149 儿科学 1.74%
数据挖掘分析 129 医学教育与医学边缘学科 0.91%
组方规律 128 图书情报与数字图书馆 0.82%
中医治疗 124 医药方针卫生政策 0.74%
关联规则 118 生物医学工程 0.58%
中医症候 98 中西医结合 0.54%
Table 1  NLP/机器学习/数据挖掘在中医领域的应用
Fig.1  融合多NLP任务的中医辅助诊疗流程框架
Fig.2  中医辅助诊疗的量化评诊和症状匹配流程框架
Fig.3  量化评诊和症状匹配查询过滤的维恩图表达
Fig.4  中医辅助诊疗的系统预诊和关联映射流程框架
糖尿病疾病类型 样本量
糖尿病肾病 236
糖尿病 456
1型糖尿病 83
2型糖尿病 69
糖尿病足 47
糖尿病性胃轻瘫 62
糖尿病性神经病变 67
Table 2  糖尿病疾病类型与样本量
Fig.5  量化评诊部分计算结果
症状匹配结果 症状匹配度
***主症:口渴引饮、易饥多食、心烦失眠、尿频便秘。
次症:急躁易怒、面红目赤、心悸怔忡、头晕目眩。
舌脉:舌红,苔黄,脉弦数或弦滑数。
兼血瘀证:心胸隐痛、肢体麻痛、舌质暗红。
兼湿热证:脘腹胀满、恶心呕吐、口干口臭,舌红,苔黄腻。
0.926 6
***尿频量多或尿浊,头晕眼花,口渴心烦,便干耳鸣,苔少脉细属气阴两虚型。伴尿浊浮肿,神倦乏力,纳少腹胀,腰膝疫软,或畏寒尿少,面色晦滞属脾肾阳虚型。若浮肿明显,腰以下尤甚,身重困倦,头重如裹,或呕恶,口干腥臭,尿少而浊,苔腻脉滑为夹湿浊型。若兼浮肿日久不消,腰痛如折,舌紫瘀斑,脉涩结代者为夹瘀型。 0.917 8
... ...
Table 3  症状匹配部分结果
症状概况 建议治法 置信度
***目眩|心悸怔忡|乏力|形寒肢冷|半身不遂|乏力|舌有裂纹|五心烦热|少尿|血栓性脉管炎|神疲乏力|面红目赤身热|尿频便塘|舌质淡胖|舌质黯红少津|舌质淡有瘀斑|脉细数|脉沉细无力|心情不畅难以入睡|肉芽色暗|舌淡|喉间有痰|脉象细数|多尿|消瘦|手足麻木|睡不解乏|舌体有瘀斑|苔黄粗燥|纳呆腹胀|形体消瘦|倦怠乏力|疼痛|舌质红|头昏且胀|伴乏力肢楚|脉濡数|多尿|舌淡苔白而干|偶头晕头痛|脉沉细等症状|近来日益明盈|痛觉|质较清稀|舌边齿痕|舌红苔黄|混浊|舌淡|胸闷恶心|形盛体胖|小便如膏| 滋阴清热,活血化瘀,豁痰通络,轻型温经散寒,活血通络。 0.993 4
***小便清长或淋漓不尽|舌瘦|尿多|大便粘滞腥臭|身重倦怠|下肢水肿|关节变形|痛连左臂|气短|偶有下肢转筋|脘腹不舒|腰酸|体重倦怠|舌淡有齿痕|半年来感乏力|心慌气短|口淡无味或黏腻|多处搔抓痕|心悸|口干喜饮|畏寒肢冷|多食易饥|苔少|面生褐斑|苔白或少苔|疲劳后尿沫增加|胖大|尿糖|苔薄白腻|眩晕:以头目眩晕|排便乏力|胃纳可|神呆为主症|甲状腺炎等病史|鼻孔干燥|苔色薄黄|日渐明显|颜色黧黑|苔黄 益气养阴,通阳利水,活血通络 0.465 7
... ... ...
Table 4  自动摘要部分结果
症状 症状判别
***瞌睡,以前上班都很精神饱满,会突然出现极度的无法抗拒的想昏睡的感觉,趴在桌上半个小时就马上改善 [('糖尿病肾病',0.3841),('糖尿病',0.9640),('糖尿病足',0.2663),('2型糖尿病',0.4414),('1型糖尿病',0.4521),('糖尿病性胃轻瘫',0.8900),('糖尿病性神经病变',0.3600)]
***每日腹泻次数达2~10次不等,常于餐后、夜间或清晨排便,大便不成形或是脂肪泻,表现为顽固或间歇性的腹泻 [('糖尿病肾病',0.2332),('糖尿病',0.7455),('糖尿病足',0.0),('2型糖尿病',0.8014),('1型糖尿病',0.8891),('糖尿病性胃轻瘫',0.3600),('糖尿病性神经病变',0.1295)]
... ...
Table 5  疾病判别部分结果
症状-用药-频繁项集 症状-用药-频繁规则
('丹参', '无力', '疲乏'): 20 {山药, 葛根} -> {乏力}
('无力', '疲乏', '黄连'): 16 {黄芪, 茯苓} -> {气短}
('舌', '黄芩', '黄连'): 18 {丹皮, 泽泻} -> {大便}
('乏力', '口渴', '心烦', '黄芪'): 16 {山茱萸, 山药} -> {酸软}
... ...
Table 6  糖尿病症状与药物关联挖掘部分结果
Fig.6  10轮评价矩阵及其平均值可视化
Round V1 V2 V3 V4
R1 0.1222 0.4165 0.1509 0.3104
R2 0.1164 0.2443 0.2136 0.4257
R3 0.3093 0.2602 0.1304 0.3001
R4 0.1356 0.4739 0.1344 0.2561
R5 0.3270 0.1108 0.3628 0.1993
R6 0.1449 0.3871 0.2058 0.2622
R7 0.2272 0.2253 0.1453 0.4022
R8 0.3240 0.2121 0.2912 0.1727
R9 0.0997 0.4154 0.2399 0.2450
R10 0.1427 0.3942 0.2987 0.1643
Average 0.1949 0.3140 0.2173 0.2738
Table 7  模糊综合评价矩阵的隶属度计算结果及其均值
[1] 蔡菲菲, 马立新, 胡元佳, 等. “AI+中医药”: 迈向智慧中医药时代的新途径[J]. 世界科学技术: 中医药现代化, 2019, 21(9):1968-1973.
[1] ( Cai Feifei, Ma Lixin, Hu Yuanjia, et al. “AI + Traditional Chinese Medicine”: A New Approach to the Era of Smart Traditional Chinese Medicine[J]. World Science and Technology: Modernization of Traditional Chinese Medicine, 2019, 21(9):1968-1973.)
[2] 夏淑洁, 杨朝阳, 周常恩, 等. 常见机器学习方法在中医诊断领域的应用述评[J]. 广州中医药大学学报, 2021, 38(4):826-831.
[2] ( Xia Shujie, Yang Chaoyang, Zhou Changen, et al. Review of Application of Machine Learning Methods in the Field of Diagnostics of Traditional Chinese Medicine[J]. Journal of Guangzhou University of Traditional Chinese Medicine, 2021, 38(4):826-831.)
[3] Chowdhury G G. Natural Language Processing[J]. Annual Review of Information Science and Technology, 2003, 37(1):51-89.
doi: 10.1002/aris.1440370103
[4] Gopalakrishnan K, Hedayatnia B, Chen Q, et al. Topical-Chat: Towards Knowledge-Grounded Open-Domain Conversations[C]// Proceedings of the INTERSPEECH 2019. 2019:1891-1895.
[5] Zhou L, Gao J, Li D, et al. The Design and Implementation of XiaoIce, an Empathetic Social Chatbot[J]. Computational Linguistics, 2020, 46(1):53-93.
doi: 10.1162/coli_a_00368
[6] Guzella T S, Caminhas W M. A Review of Machine Learning Approaches to Spam Filtering[J]. Expert Systems with Applications, 2009, 36(7):10206-10222.
doi: 10.1016/j.eswa.2009.02.037
[7] Wang H, Zhang F, Wang J, et al. Ripplenet: Propagating User Preferences on the Knowledge Graph for Recommender Systems[C]// Proceedings of the 27th ACM International Conference on Information and Knowledge Management. 2018: 417-426.
[8] 吴玺煜, 陈启买, 刘海, 等. 基于知识图谱表示学习的协同过滤推荐算法[J]. 计算机工程, 2018, 44(2):226-232, 263.
[8] ( Wu Xiyu, Chen Qimai, Liu Hai, et al. Collaborative Filtering Recommendation Algorithm Based on Representation Learning of Knowledge Graph[J]. Computer Engineering, 2018, 44(2):226-232, 263.)
[9] 胡佳慧, 方安, 赵琬清, 等. 面向知识发现的中文电子病历标注方法研究[J]. 数据分析与知识发现, 2019, 3(7):123-132.
[9] ( Hu Jiahui, Fang An, Zhao Wanqing, et al. Annotating Chinese E-Medical Record for Knowledge Discovery[J]. Data Analysis and Knowledge Discovery, 2019, 3(7):123-132.)
[10] Liang H, Tsui B Y, Ni H, et al. Evaluation and Accurate Diagnoses of Pediatric Diseases Using Artificial Intelligence[J]. Nature Medicine, 2019, 25(3):433-438.
doi: 10.1038/s41591-018-0335-9
[11] 侯梦薇, 卫荣, 陆亮, 等. 知识图谱研究综述及其在医疗领域的应用[J]. 计算机研究与发展, 2018, 55(12):2587-2599.
[11] ( Hou Mengwei, Wei Rong, Lu Liang, et al. Research Review of Knowledge Graph and Its Application in Medical Domain[J]. Journal of Computer Research and Development, 2018, 55(12):2587-2599.)
[12] 李健, 卢朋, 唐仕欢, 等. 基于中医传承辅助系统的治疗肺痈方剂组方规律分析[J]. 中国实验方剂学杂志, 2012, 18(2):254-257.
[12] ( Li Jian, Lu Peng, Tang Shihuan, et al. Analysis on Composition Principles for Lung Abscess by Using Traditional Chinese Medicine Inheritance System[J]. Chinese Journal of Experimental Traditional Medical Formulae, 2012, 18(2):254-257.)
[13] 姚美村, 艾路, 袁月梅, 等. 消渴病复方配伍规律的关联规则分析[J]. 北京中医药大学学报, 2002, 25(6):48-50.
[13] ( Yao Meicun, Ai Lu, Yuan Yuemei, et al. Analysis of the Association Rule in the Composition of the TCM Formulas for Diabetes[J]. Journal of Beijing University of TCM, 2002, 25(6):48-50.)
[14] Shi Q, Zhao H, Chen J, et al. Study on TCM Syndrome Identification Modes of Coronary Heart Disease Based on Data Mining[J]. Evidence-Based Complementary and Alternative Medicine, 2012:Article ID 697028.
[15] Yang H, Chen J, Tang S, et al. New Drug R&D of Traditional Chinese Medicine: Role of Data Mining Approaches[J]. Journal of Biological Systems, 2009, 17(3):329-347.
doi: 10.1142/S0218339009002971
[16] Fang Y C, Huang H C, Chen H H, et al. TCMGeneDIT: A Database for Associated Traditional Chinese Medicine, Gene and Disease Information Using Text Mining[J]. BMC Complementary and Alternative Medicine, 2008, 8: Article No.58
[17] 龚燕冰, 倪青, 王永炎. 中医证候研究的现代方法学述评(一)——中医证候数据挖掘技术[J]. 北京中医药大学学报, 2006, 29(12):797-801.
[17] ( Gong Yanbing, Ni Qing, Wang Yongyan. Modern Methodology of TCM Syndrome Study(I): Data Mining Technology of TCM Syndrome[J]. Journal of Beijing University of TCM, 2006, 29(12):797-801.)
[18] 戴霞, 郭伟星. 中医证候诊断标准规范化研究概况[J]. 中医杂志, 2011, 52(2):168-171.
[18] ( Dai Xia, Guo Weixing. Survey of Research on Standardization of Diagnostic Criteria of TCM Syndromes[J]. Journal of Traditional Chinese Medicine, 2011, 52(2):168-171.)
[19] Zhou X, Peng Y, Liu B. Text Mining for Traditional Chinese Medical Knowledge Discovery: A Survey[J]. Journal of Biomedical Informatics, 2010, 43(4):650-660.
doi: 10.1016/j.jbi.2010.01.002
[20] 沈毅, 傅萍, 孔丽娅. 数据挖掘方法在名老中医用药规律研究中的应用[J]. 中医杂志, 2016, 57(10):890-893.
[20] ( Shen Yi, Fu Ping, Kong Liya. Application of Data Mining Method in the Research of Famous and Old Traditional Chinese Medicine Drugs[J]. Journal of Traditional Chinese Medicine, 2016, 57(10):890-893.)
[21] 王华珍, 彭淑娟, 缑锦, 等. 基于随机森林的中医数据可视化研究[J]. 系统仿真学报, 2014, 26(11):2751-2756.
[21] ( Wang Huazhen, Peng Shujuan, Tiao Jin, et al. Data Visualization of Traditional Chinese Medicine Based on Random Forest[J]. Journal of System Simulation, 2014, 26(11):2751-2756.)
[22] Lukman S, He Y, Hui S C. Computational Methods for Traditional Chinese Medicine: A Survey[J]. Computer Methods and Programs in Biomedicine, 2007, 88(3):283-294.
doi: 10.1016/j.cmpb.2007.09.008
[23] Feng Y, Wu Z, Zhou X, et al. Knowledge Discovery in Traditional Chinese Medicine: State of the Art and Perspectives[J]. Artificial Intelligence in Medicine, 2006, 38(3):219-236.
pmid: 16930966
[24] Zhou X, Chen S, Liu B, et al. Development of Traditional Chinese Medicine Clinical Data Warehouse for Medical Knowledge Discovery and Decision Support[J]. Artificial Intelligence in Medicine, 2010, 48(2-3):139-152.
doi: 10.1016/j.artmed.2009.07.012
[25] Yang J, Tian S, Zhao J, et al. Exploring the Mechanism of TCM Formulae in the Treatment of Different Types of Coronary Heart Disease by Network Pharmacology and Machining Learning[J]. Pharmacological Research, 2020, 159:105034.
doi: 10.1016/j.phrs.2020.105034
[26] Li Y H, Aslam M S, Yang K L, et al. Classification of Body Constitution Based on TCM Philosophy and Deep Learning[J]. Symmetry, 2020, 12(5):803.
doi: 10.3390/sym12050803
[27] 刘耀, 穗志方, 周扬, 等. 中医药本体概念描述体系的自动构建研究[J]. 现代图书情报技术, 2008(5):21-26.
[27] ( Liu Yao, Sui Zhifang, Zhou Yang, et al. Research on Automatic Construction of Chinese Traditional Medicine Ontology Concept's Description Architecture[J]. New Technology of Library and Information Service, 2008(5):21-26.)
[28] 张晓航, 石清磊, 王斌, 等. 机器学习算法在中医诊疗中的研究综述[J]. 计算机科学, 2018, 45(S2):32-36.
[28] ( Zhang Xiaohang, Shi Qinglei, Wang Bin, et al. Review of Machine Learning Algorithms in Traditional Chinese Medicine[J]. Computer Science, 2018, 45(S2):32-36.)
[29] Zhang Z, Han X, Liu Z, et al. ERNIE: Enhanced Language Representation with Informative Entities[OL]. arXiv Preprint, arXiv: 1905.07129.
[30] Agarwal R, Srikant R. Fast Algorithms for Mining Association Rules [C]//Proceedings of the 20th VLDB Conference. 1994, 487:499.
[31] 胡珊, 刘晶. 模糊综合评价法在产品设计方案决策中的应用[J]. 机械设计, 2020, 37(1):135-139.
[31] ( Hu Shan, Liu Jing. Application of Fuzzy Comprehensive Evaluation Method in Product Design Scheme Decision[J]. Machinery Design, 2020, 37(1):135-139.)
[32] Reimers N, Gurevych I. Sentence-Bert: Sentence Embeddings Using Siamese Bert-Networks[OL]. arXiv Preprint, arXiv: 1908.10084.
[1] 王一钒,李博,史话,苗威,姜斌. 古汉语实体关系联合抽取的标注方法*[J]. 数据分析与知识发现, 2021, 5(9): 63-74.
[2] 周志超. 基于机器学习技术的自动引文分类研究综述*[J]. 数据分析与知识发现, 2021, 5(12): 14-24.
[3] 黄名选,卢守东,徐辉. 基于加权关联模式挖掘与规则后件扩展的跨语言信息检索 *[J]. 数据分析与知识发现, 2019, 3(9): 77-87.
[4] 胡佳慧,方安,赵琬清,杨晨柳,任慧玲. 面向知识发现的中文电子病历标注方法研究 *[J]. 数据分析与知识发现, 2019, 3(7): 123-132.
[5] 杨春雷. 面向语用消歧的量化约束条件系统: 从语言学设计到计算实现*[J]. 数据分析与知识发现, 2017, 1(11): 1-11.
[6] 杨春雷. 基于HPSG的汉语词库和语法规则系统构建*[J]. 现代图书情报技术, 2016, 32(7-8): 129-136.
[7] 刘天祎,步一,赵丹群,黄文彬. 自动引文摘要研究述评[J]. 现代图书情报技术, 2016, 32(5): 1-8.
[8] 彭浩, 徐健, 肖卓. 基于比较句的网络用户评论情感分析[J]. 现代图书情报技术, 2015, 31(12): 48-56.
[9] 杨春雷, Dan Flickinger. 汉构:面向深层语言处理的语法工程[J]. 现代图书情报技术, 2014, 30(3): 57-64.
[10] 邱均平, 方国平. 基于知识图谱的中外自然语言处理研究的对比分析[J]. 现代图书情报技术, 2014, 30(12): 51-61.
[11] 佘贵清, 张永安. 审判案例自动抽取与标注模型研究[J]. 现代图书情报技术, 2013, (6): 23-29.
[12] 王秀艳, 崔雷. 采用混合方法抽取生物医学实体间语义关系[J]. 现代图书情报技术, 2013, 29(3): 77-82.
[13] 张运良 梁健 朱礼军 乔晓东. 基于术语定义的科技知识组织系统自动丰富关键技术研究*[J]. 现代图书情报技术, 2010, 26(7/8): 66-71.
[14] 仲夏 张志平 王惠临. 词汇化树邻接语法研究述评及中文应用初探*[J]. 现代图书情报技术, 2010, 26(5): 35-42.
[15] 傅继彬,刘杰,贾可亮,毛金涛. 基于知网和术语相关度的本体关系抽取研究*[J]. 现代图书情报技术, 2008, 24(9): 36-40.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
版权所有 © 2015 《数据分析与知识发现》编辑部
地址:北京市海淀区中关村北四环西路33号 邮编:100190
电话/传真:(010)82626611-6626,82624938
E-mail:jishu@mail.las.ac.cn