Force-directed layout visualization method for medical record knowledge graph presentation applications

doi:10.11772/j.issn.1001-9081.2025081037

Abstract

Abstract:

The force-directed layout method is a classical approach used for knowledge graph visualization， and it distributes nodes in the visualization space uniformly through attractive and repulsive forces between nodes. However， since visualization of domain-specific knowledge graphs is often closely related to domain applications， the classical force-directed layout method cannot meet the visualization requirements of domain-specific knowledge graphs. Therefore， an improved force-directed layout method was proposed by analyzing the graph structure， data characteristics， and visualization requirements of query applications of medical record knowledge graphs， with the aim of supporting visualization technology for medical record knowledge graphs in clinical assisted diagnosis and treatment applications. In the method， based on the classical force-directed layout model， attractive force between nodes of the same level and same type， repulsive force between nodes of the same level but different types， and repulsive force between non-leaf and non-adjacent nodes to maintain the spacing were introduced. As a result， nodes of the same level and the same type were grouped into the same region， nodes of the same level but different types were distributed into separate regions， and nodes at different levels were distributed with reasonable spacing， thereby reducing node crowding and edge crossings； so that the data’s type characteristics， quantity characteristics， and community characteristics were revealed by the final generated visualization layout clearly. Additionally， node type and the number of same-level， same-type nodes were taken as factors determining edge length between nodes， and the node importance was represented through edge length， thereby further improving the rationality of the visualization layout and enhancing semantic visualization. Experimental results show that in terms of semantic visualization capability in vertical domains， the proposed method achieves an average improvement of 165.5% in the mean Node-Type Spatial Discriminability （NTSD） compared with the FR （Fruchterman-Reingold） method. As the visualization results， this method has better performance for electronic medical record knowledge graph visualization.

Key words: medical record knowledge graph, visual expression, force-directed layout, node-link graph, force-directed method integrating type information

摘要：

力导向布局方法是知识图谱可视化采用的经典方法，它通过节点间的引力和排斥力将节点均匀地分布在可视化空间内。但垂直领域的知识图谱可视化常与领域应用密切相关，经典的力导向布局方法不能满足垂直领域知识图谱的可视化需求。因此，以支持临床辅助诊疗应用的病历知识图谱可视化技术为目的，通过分析病历知识图谱的图结构、数据特点以及查询应用的可视化需求，提出改进的力导向布局方法。该方法在经典力导向布局模型的基础上引入同级同类型节点间引力、同级异类型节点间排斥力和非叶且非邻接节点间维持间距排斥力，使在布局上同级同类型节点能集中在同一个区域，同级不同类型节点分布到不同的区域，不同级别的节点以合理的间距分布，从而减少节点挤压和边交叉现象，使最终生成的可视化布局能清晰地展示数据的类型特征、数量特征和社区特征。此外，将节点的类型和同级同类型节点数作为决定节点间边长的因子，并通过边的长度表达节点的重要性，从而进一步提升可视化布局的合理性，以强化语义的可视化表达。实验结果表明，在垂直领域语义可视化能力方面，本文方法的节点类型空间区分度（NTSD）均值相较于FR（Fruchterman-Reingold）方法平均提升了165.5%。从可视化结果来看，该方法在电子病历知识图谱可视化上获得了更好的效果。

关键词: 病历知识图谱, 可视化表达, 力导向布局, 节点链接图, 融合类型信息的力导向方法

CLC Number:

TP391.41

Dongyang ZHOU, Jinlian DU, Xueyun JIN. Force-directed layout visualization method for medical record knowledge graph presentation applications[J]. Journal of Computer Applications, 2026, 46(9): 2847-2856.

周东阳, 杜金莲, 金雪云. 面向病历知识图谱展示应用的力导向布局可视化方法[J]. 《计算机应用》唯一官方网站, 2026, 46(9): 2847-2856.

Figures/Tables 16

Fig. 1 Classical force-directed layout visualization for knowledge graph of medical records

Fig. 2 Schematic diagram of directions of forces between nodes

Fig. 3 Node distribution under classical force-directed layout

Fig. 4 Comparison of effects before and after adding attractive force for same level and same type

Fig. 5 Comparison of effects before and after adding repulsive force for heterogeneous same level and different types

Fig. 6 Comparison of effects before and after adding repulsive force to maintain spacing

Fig. 7 Schematic diagram of node spacing

Tab. 1 Layout generation time under different subgraph scales

节点数	边数	平均时间/s
69	173	2.830
372	880	15.578
694	1 720	30.392
1 108	3 441	99.229

Tab. 2 Knowledge graph entity data for electronic medical record

标签	描述	实体数	负载数据
Diagnosis	疾病	6 006	Id， label， name
Examination	检查	350	Id， label， name
Detection	检验	256	Id， label， name
MajorSymptom	症状	988	Id， label， name
MedicationTreatment	治疗	545	Id， label， name
Patient	患者	2 000	Id， age， gender， label， name， patientSN
M-E	就诊经历	2 120	Id， label， EXPTIME， EXP_ID， MR_ID，

Tab. 3 Knowledge graph relationship data for electronic medical record

标签	关系方向	关系数
FALL_IN	就诊经历→疾病	27 030
HAS_CHECK	就诊经历→检查	1 575
HAS_EXAMINE	就诊经历→检验	761
HAS_PATIENT	就诊经历→患者	2 120
SHOW_SYM	就诊经历→症状（主症状、伴随症状、无症状）	2 729
TAKE_MEDICATION	就诊经历→治疗（手术治疗、药物治疗）	1 878

Fig. 8 Explanation of node colors and shapes

Fig. 9 Visualization results of five methods （Query 1）

Tab. 4 Relevant statistical results of NTSD

方法	查询1		查询2
方法	NTSD均值	NTSD标准差	NTSD均值	NTSD标准差
FR方法	1.539 5	0.684 0	2.243 0	0.123 7
Neo4j方法	2.320 0	0.279 5	2.352 3	0.304 0
康方法	3.383 5	0.244 5	3.335 3	0.185 0
ForceAtlas2方法	1.338 5	0.067 3	2.215 8	0.165 5
本文方法	5.060 7	0.420 5	4.536 3	0.349 5

Fig. 10 Visualization results of five methods （Query 2）

Fig. 11 Visualization results of four models （Query 1）

Fig. 12 Visualization results of four models （Query 2）

References 24

[1]	Chandak P， Huang K， Zitnik M. Building a knowledge graph to enable precision medicine ［J］. Scientific Data， 2023， 10： No.67.
[2]	Hu X， Wang L， Wei Q， et al. Research on time-series visualization methods for scientific and technical consulting graph data［C］// FRSE 2023. Piscataway： IEEE， 2023： 393-400.
[3]	Fruchterman T M J， Reingold E M. Graph drawing by force-directed placement ［J］. Software： Practice and Experience， 1991， 21（11）： 1129-1164.
[4]	杨卓，谢雅淇，陈谊，等. 图可视化布局方法最新研究进展综述［J］. 计算机工程与应用， 2023， 59（16）： 1-15.
	Yang Zhuo， Xie Yaqi， Chen Yi， et al. Review of latest research for layout methods of graph visualization ［J］. Computer Engineering and Applications， 2023， 59（16）： 1-15.
[5]	Zhou H， Huang S， Zheng Y， et al. DKGV： a dynamic knowledge graph visualization method based on force-directed layout ［C］// VRCAI 2024. New York： ACM， 2024： No.3.
[6]	Hu Y. Efficient， high-quality force-directed graph drawing ［J］. The Mathematica Journal， 2005， 10（1）： 37-71.
[7]	Walshaw C. A multilevel algorithm for force-directed graph drawing［C］// GD 2000， LNCS 1984. Berlin： Springer， 2001： 171-182.
[8]	Noack A. An energy model for visual graph clustering［C］// International Symposium on Graph Drawing. Berlin： Springer， 2003： 425-436.
[9]	Jacomy M， Venturini T， Heymann S， et al. ForceAtlas2， a continuous graph layout algorithm for handy network visualization designed for the Gephi software ［J］. PLoS ONE， 2014， 9（6）： No.e98679.
[10]	Grama A Y， Kumar V， Sameh A. Scalable parallel formulations of the Barnes-Hut method for n-body simulations ［C］// Supercomputing 1994. Piscataway： IEEE， 1994： 439-448.
[11]	康健梓，周虹. 基于力导向布局的知识图谱可视化方法［J］. 计算机辅助设计与图形学学报， 2024， 36（8）： 1214-1223.
	Kang Jianzi， Zhou Hong. Force-directed layout based knowledge graph visualization method ［J］. Journal of Computer-Aided Design and Computer Graphics， 2024， 36（8）： 1214-1223.
[12]	Purchase H C. Metrics for graph drawing aesthetics ［J］. Journal of Visual Languages and Computing， 2002， 13（5）： 501-516.
[13]	Du F， Cao N， Lin Y R， et al. iSphere： Focus+context sphere visualization for interactive large graph exploration ［C］// CHI 2017. New York： ACM， 2017： 2916-2927.
[14]	Ahmed R， de Luca F， Devkota S， et al. Graph drawing via gradient descent，（GD）² ［C］// GD 2020， LNCS 12590. Cham： Springer， 2020： 3-17.
[15]	王鑫，傅强，王林，等. 知识图谱可视化查询技术综述［J］. 计算机工程， 2020， 46（6）： 1-11.
	Wang Xin， Fu Qiang， Wang Lin， et al. Survey on visualization query technology of knowledge graph［J］. Computer Engineering， 2020， 46（6）： 1-11.
[16]	王勇超，罗胜文，杨英宝，等. 知识图谱可视化综述［J］. 计算机辅助设计与图形学学报， 2019， 31（10）： 1666-1676.
	Wang Yongchao， Luo Shengwen， Yang Yingbao， et al. A survey on knowledge graph visualization［J］. Journal of Computer-Aided Design and Computer Graphics， 2019， 31（10）： 1666-1676.
[17]	陈谊，张梦录，万玉钗. 图的表示与可视化方法综述［J］. 系统仿真学报， 2020， 32（7）： 1232-1243.
	Chen Yi， Zhang Menglu， Wan Yuchai. A survey on graph representation and visualization techniques ［J］. Journal of System Simulation， 2020， 32（7）： 1232-1243.
[18]	王春杰，何进荣，王文发. 基于图的多视角聚类算法综述［J］. 计算机与数字工程， 2022， 50（2）： 229-237， 255.
	Wang Chunjie， He Jinrong， Wang Wenfa. A survey of graph-based multi-view clustering algorithms ［J］. Computer and Digital Engineering， 2022， 50（2）： 229-237， 255.
[19]	Santos A， Colaço A R， Nielsen A B， et al. Clinical knowledge graph integrates proteomics data into clinical decision-making ［PP/OL］. bioRxiv （2020-05-10）［2025-07-09］..
[20]	Gong F， Wang M， Wang H， et al. SMR： medical knowledge graph embedding for safe medicine recommendation ［J］. Big Data Research， 2021， 23： 100174.
[21]	Zhou C， Wang Z， Jiang Z. Construction of medical knowledge atlas and information visualization ［C］// ICCIR 2022. New York： ACM， 2022： 247-253.
[22]	Peng J， Xu D， Lee R， et al. Expediting knowledge acquisition by a web framework for Knowledge Graph Exploration and Visualization （KGEV）： case studies on COVID-19 and Human Phenotype Ontology ［J］. BMC Medical Informatics and Decision Making， 2022， 22（）： 147.
[23]	Du J， Duan S， Jin X， et al. The methodology of mapping relational data to graph data for the construction of knowledge graph in Chinese medical records ［C］// SPIE 13208， IC-BIS 2024. Bellingham， WA： SPIE， 2024： No.1320826.
[24]	Kamada T， Kawai S. An algorithm for drawing general undirected graphs ［J］. Information Processing Letters， 1989， 31（1）： 7-15.