Construction of domain knowledge graph of dementia care

doi:10.3761/j.issn.0254-1769.2024.04.007

Abstract

Abstract:

Objective To construct a domain knowledge graph of dementia care，so as to provide the foundation and guarantee for the next intelligent application based on the knowledge graph. Methods A top-down approach was adopted to construct a domain knowledge graph of dementia care. Firstly，the ontology concept is constructed from the top level，namely the schema layer of knowledge graph. Then，instances are filled，and knowledge extraction is carried out from the existing data sources，and the extracted entities and relationships are filled into the pattern layer ontology database to complete the data layer construction of the knowledge graph. Finally，the “entity relationship entity” triplet data was input into the Neo4j graph database for storage. Results In this study，the personalized care plan set of 1 012 dementia cases was used as the corpus to construct a domain knowledge graph of dementia care. The knowledge graph takes people with dementia as the core，and unfolds，one by one，around basic characteristics，care problems，and care plans in a standardized “entity-relationship-entity” triplet format，forming a large knowledge network，which contains a total of 1 522 specific dementia care knowledge entities and 8 kinds of inter-entity relationships. Conclusion The domain knowledge graph of dementia care constructed in this study clearly and intuitively shows the global pedigree and logical path of knowledge，which provides an efficient and intelligent basic guarantee for the browsing，retrieval and application of dementia care knowledge，so as to realize personalized and intelligent management of people with dementia，break through the bottleneck of lack of professionals，improve the health outcomes of people with dementia，promote the implementation of inclusive pension services，and promote healthy aging.

Key words: Dementia, Caregivers, Knowledge Graph, Artificial Intelligence, Geriatric Nursing

摘要：

目的构建失智照护领域知识图谱，为下一步基于知识图谱的智能应用提供基础和保障。方法采用自顶向下的方法构建失智照护领域知识图谱。首先从顶层开始构建本体概念，即知识图谱的模式层构建。之后进行实例填充，从已有数据源中进行知识抽取，将抽取的实体、关系等填充到模式层本体库中，完成知识图谱的数据层构建。最后将“实体-关系-实体”三元组数据输入Neo4j图数据库进行存储。结果以1 012例失智个案的个性化照护方案集为知识来源，构建失智照护领域知识图谱。该知识图谱以失智老年人为核心，围绕基本特征、照护问题、照护方案以规范的“实体-关系-实体”三元组格式的数据形式逐一展开，形成一张巨大的知识网络，共包含1 522种失智照护知识实体及8种实体间关系。结论该研究构建的失智照护领域知识图谱清晰、直观地展示了知识图谱的全局谱系及知识的逻辑路径，为失智照护知识浏览、检索、应用提供了高效智能的基础保障，从而实现失智老年人的个性化、智能化管理，突破专业人员不足的瓶颈，改善失智老年人的健康结局，推动普惠型养老服务落地实施，促进健康老龄化。

关键词: 痴呆, 照护者, 知识图谱, 人工智能, 老年护理学

LENG Minmin, SUN Yue, LU Weihua, LI Baihe, SHANG Zhiwei, WANG Zhiwen. Construction of domain knowledge graph of dementia care[J]. Chinese Journal of Nursing, 2024, 59(4): 432-437.

冷敏敏, 孙月, 鲁卫华, 李百合, 尚志嵬, 王志稳. 失智照护领域知识图谱的构建[J]. 中华护理杂志, 2024, 59(4): 432-437.

Figures/Tables 5

Figure 1 Relationship construction of domain knowledge graph of dementia care

Table 1 Socio-demographic characteristics of people with dementia（n=1 012）

Table 2 Entity framework of domain knowledge graph of dementia care

Table 3 Inter-entity relationship of domain knowledge graph of dementia care

Figure 2 Domain knowledge graph of dementia care（part）

References 29

[1]	陈晓春, 张杰文, 贾建平, 等. 2018中国痴呆与认知障碍诊治指南(一):痴呆及其分类诊断标准[J]. 中华医学杂志, 2018, 98(13):965-970.
	Chen XC, Zhang JW, Jia JP, et al. 2018 Chinese guidelines for the diagnosis and treatment of dementia and cognitive impairment(1):dementia and its classification diagnostic criteria[J]. Natl Med J China,2018, 98(13):965-970.
[2]	Jia LF, Du YF, Chu L, et al. Prevalence,risk factors,and management of dementia and mild cognitive impairment in adults aged 60 years or older in China:a cross-sectional study[J]. Lancet Public Health, 2020, 5(12):e661-e671. DOI URL
[3]	GBD Dementia Forecasting Collaborators. Estimation of the glo-bal prevalence of dementia in 2019 and forecasted prevalence in 2050:an analysis for the global burden of disease study 2019[J]. Lancet Public Health, 2022, 7(2):e105-e125. DOI URL
[4]	van den Kieboom R, Snaphaan L, Mark R, et al. The trajectory of caregiver burden and risk factors in dementia progression:a systematic review[J]. J Alzheimers Dis, 2020, 77(3):1107-1115. DOI URL
[5]	康祎陈, 刘文琳, 丁玎, 等. 年轻型痴呆症患者及其照护者疾病体验的质性研究[J]. 中华护理杂志, 2022, 57(13):1591-1598. DOI
	Kang YC, Liu WL, Ding D, et al. Qualitative study on the dyadic experience of patients with young-onset dementia and their caregivers[J]. Chin J Nurs, 2022, 57(13):1591-1598. DOI
[6]	Rigby T, Johnson DK, Taylor A, et al. Comparison of the caregiving experience of grief,burden,and quality of life in dementia with lewy bodies,Alzheimer’s disease,and Parkinson’s disease dementia[J]. J Alzheimers Dis, 2021, 80(1):421-432. DOI PMID
[7]	冷敏敏, 赵雅洁, 李立玉, 等. 网络干预在痴呆老年人居家照护中应用的范围综述[J]. 中华护理杂志, 2021, 56(5):694-699. DOI
	Leng MM, Zhao YJ, Li LY, et al. Internet-based interventions for family care of older people with dementia:a scoping review[J]. Chin J Nurs, 2021, 56(5):694-699.
[8]	Paulheim H. Knowledge graph refinement:a survey of approaches and evaluation methods[J]. Semantic Web, 2016, 8(3):489-508. DOI URL
[9]	康风建. 基于知识图谱的跨境商品海关编码预测系统研究与实现[D]. 济南: 山东大学, 2022.
	Kang FJ. Research and implementation of cross-border commodity customs code prediction system based on knowledge map[D]. Jinan: Shandong University, 2022.
[10]	杨敬慧. 基于知识图谱的新闻推荐系统研究[J]. 科学与信息化, 2021(18):187-189.
	Yang JH. Research on news recommendation system based on knowledge graph[J]. Scie Inform, 2021(18):187-189.
[11]	Long J, Chen Z, He W, et al. An integrated framework of deep learning and knowledge graph for prediction of stock price trend:an application in Chinese stock exchange market[J]. Appl Soft Comput, 2020, 91:106205. DOI URL
[12]	Yin YT, Zhang L, Wang YG, et al. Question answering system based on knowledge graph in traditional Chinese medicine diagnosis and treatment of viral hepatitis B[J]. Biomed Res Int, 2022, 2022:7139904.
[13]	Chai XQ. Diagnosis method of thyroid disease combining knowledge graph and deep learning[J]. IEEE Access, 2020, 8:149787-149795. DOI URL
[14]	Gong F, Wang M, Wang HF, et al. SMR:medical knowledge graph embedding for safe medicine recommendation[J]. Big Data Res, 2021, 23:100174. DOI URL
[15]	Ye Q, Hsieh CY, Yang ZY, et al. A unified drug-target interaction prediction framework based on knowledge graph and recommendation system[J]. Natl Commun, 2021, 12(1):6775. DOI
[16]	Chicaiza J, Valdiviezo-Diaz P. A comprehensive survey of knowledge graph-based recommender systems:technologies,development,and contributions[J]. Information, 2021, 12(6):2078-2489.
[17]	朱冬亮, 文奕, 万子琛. 基于知识图谱的推荐系统研究综述[J]. 数据分析与知识发现, 2021(12):1-13.
	Zhu DL, Wen Y, Wan ZC. Review of recommendation systems based on knowledge graph[J]. Data Anal Knowl Discov, 2021(12):1-13.
[18]	李丞. 动态环境下的知识图谱表示学习方法的研究[D]. 南京: 东南大学, 2020.
	Li C. Research on knowledge map representation learning method in dynamic environment[D]. Nanjing: Southeast University, 2020.
[19]	刘凡. 基于知识图谱技术的名老中医慢性胃炎辨证论治方案研究[D]. 北京: 中国中医科学院, 2020.
	Liu F. Study on the scheme of syndrome differentiation and treatment of chronic gastritis of famous and old Chinese me-dicine based on knowledge mapping technology[D]. Beijing: China Academy of Chinese Medical Sciences, 2020.
[20]	Guo Q, Cao S, Yi Z. A medical question answering system using large language models and knowledge graphs[J]. Int J Intelligent Sys, 2022, 37(11):8548-8564. DOI URL
[21]	程默. 基于知识图谱的乳腺肿瘤辅助诊断模型的研究[D]. 武汉: 湖北工业大学, 2021.
	Cheng M. Study on auxiliary diagnosis model of breast tumor based on knowledge map[D]. Wuhan: Hubei University of Technology, 2021.
[22]	付子轩, 周鹏, 任海燕, 等. 基于知识图谱的中西医结合急腹症诊疗推理分析[J]. 中国实验方剂学杂志, 2023, 29(11):190-199.
	Fu ZX, Zhou P, Ren HY, et al. Diagnosis and treatment reasoning of integrated traditional Chinese and western medicine against acute abdomen based on knowledge graph[J]. Chin J Exp Tradit Med Formulae, 2023, 29(11):190-199.
[23]	Taneja SB, Callahan TJ, Paine MF, et al. Developing a knowledge graph for pharmacokinetic natural product-drug interactions[J]. J Biomed Inform, 2023, 140:104341. DOI URL
[24]	李伟, 王竣生, 秦鹏. 基于知识图谱的医疗问答系统研究[J]. 长江信息通信, 2023, 36(6):107-109.
	Li W, Wang JS, Qin P. Research on medical Q & A system based on knowledge graph[J]. Chang Inf Commun, 2023, 36(6):107-109.
[25]	汪纯丽. 基于静脉曲张知识图谱的辅助诊断系统的设计与实现[D]. 武汉: 华中科技大学, 2021.
	Wang CL. Design and implementation of an auxiliary diagnosis system based on varicose vein knowledge map[D]. Wuhan: Huazhong University of Science and Technology, 2021.
[26]	张文卓. 基于知识图谱的药品推荐系统的研究与实现[D]. 邯郸: 河北工程大学, 2022.
	Zhang WZ. Research and implementation of drug recommendation system based on knowledge map[D]. Handan: Hebei University of Engineering, 2022.
[27]	Ren ZH, You ZH, Yu CQ, et al. A biomedical knowledge graph-based method for drug-drug interactions prediction through combining local and global features with deep neural networks[J]. Brief Bioinform, 2022, 23(5):bbac363. DOI URL
[28]	Xu GW, Jia GY, Shi L, et al. Personalized course recommendation system fusing with knowledge graph and collaborative filtering[J]. Comput Intell Neurosci, 2021, 2021:9590502.
[29]	王丽敏, 陈泓伯, 王琦, 等. 以公众健康教育与非药物干预为主的膝关节骨性关节炎疾病知识图谱的构建[J]. 中华护理杂志, 2022, 57(10):1172-1177. DOI
	Wang LM, Chen HB, Wang Q, et al. Construction of the disease knowledge graph of knee osteoarthritis focused on public health education and non-drug interventions[J]. Chin J Nurs, 2022, 57(10):1172-1177. DOI