|
|
|
| A Knowledge Representation Method Based on Two-Layer Modeling for Constructing Medical Knowledge Base |
| Zhang Yinsheng1, Wang Rui2, Qiao Qingzhi2, Li Haomin3,4*, Lv Xudong5, Duan Huilong5 |
1School of Management and E-Busineess, Zhejiang Gongshang University, Hangzhou 310018, China
2Department of Information, Shanxi Dayi Hospital, Taiyuan 030032, China
3Children’s Hospital, Zhejiang University, Hangzhou 310003, China
4Institute of Translational Medicine, Zhejiang University, Hangzhou 310027, China
5College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou 310027, China |
|
|
|
|
Abstract Medical knowledge base is an important asset for clinical decision support system (CDSS). Current CDSSs deployed in hospitals are mostly targeted at specific requirements (e.g. rational drug use, computer-aided diagnosis), and their knowledge bases usually contain one specific type of knowledge. A comprehensive knowledge base that can coordinate different types of medical knowledge is essential for building integral and effective decision support applications. This paper proposes a knowledge representation method based on two-layer modeling. First, an epistemological model is constructed to cover various diagnostic and therapeutic knowledge types used in clinical settings, as well as their coordinative relationships. Then, based on specific reasoning and computation requirements, concrete representation formulisms are assigned to each knowledge type to form the final computational model. A clinical decision support system based on this two-layer modeling method has been running for more than 4 years in a Class III hospital. Practices show that the system effectively solved the need of integrating and reasoning of different knowledge types, and has established a foundation for building integral and comprehensive knowledge applications.
|
|
Received: 25 November 2016
|
|
|
|
|
|
[1] Nebel B. Reasoning and Revision in Hybrid Representation Systems [M]. Berlin: Springer-Verlag Heidelberg, 1990.
[2] Sloman A, Why we need many knowledge representation formalisms [C]//Research and Development in Expert Systems IX-Proceedings BCS Expert Systems Conference 1984. Bcston: Cambridge University Press, 1985: 235.
[3] Weng C, Tu SW, Sim I,et al. Formal representations of eligibility criteria: A literature review [J]. Journal of Biomedical Informatics, 2010, 43(3): 451-467.
[4] Kraus S, Enders M, Prokosch HU,et al. Accessing complex patient data from Arden Syntax Medical Logic Modules [J]. Artificial Intelligence in Medicine, 2015, In Press.
[5] Fobel A, Subramanian N, Comparison of the performance of Drools and Jena rule-based systems for event processing on the semantic web [C]//IEEE International Conference on Software Engineering Research, Management and Applications. Hammamet: IEEE, 2016: 24-30.
[6] McClay JC, Campbell JR, Parker C, et al., Structuring Order Sets for Interoperable Distribution [C]//AMIA 2006 Symposium: Biomedical and Health Informatics. Washington, DC: American Medical Informatics Association, 2006: 549.
[7] Peleg M. Computer-interpretable clinical guidelines: a methodological review [J]. Journal of Biomedical Informatics, 2013,46(4): 744-763.
[8] Weng JT, Ameedeen MA. A survey of petri net tools [J]. Lecture Notes in Electrical Engineering, 2015, 315(8): 537-551.
[9] Hirsch JA, Lesliemazwi TM, Nicola GN,et al. Current procedural terminology: A primer [J]. Journal of NeuroInterventional Surgery, 2015, 7(4): 309-312.
[10] Zhu Q, Jiang G, Chute CG. Profiling structured product labeling with NDF-RT and RxNorm [J]. Journal of Biomedical Semantics, 2012,3(1): 16-20.
[11] Torres A, Galante R, Pimenta MS,et al. Twenty years of object-relational mapping: A survey on patterns, solutions, and their implications on application design [J]. Information and Software Technology, 2017, 82: 1-18.
[12] 张寅升, 南山, 段会龙, 等. 基于工作流管理技术的医学知识库集成与协同应用 [J]. 中华医院管理杂志, 2016, 32(11): 833-836. |
|
|
|
