DOI QR코드

DOI QR Code

Similarity Analysis of Sibling Nodes in SNOMED CT Terminology System

SNOMED CT 용어체계에서 형제 노드의 유사도 분석 기법

  • Woo-Seok Ryu (Dept. of Health Care Management, Catholic University of Pusan)
  • 류우석 (부산가톨릭대학교 병원경영학과)
  • Received : 2023.11.18
  • Accepted : 2024.02.17
  • Published : 2024.02.29

Abstract

This paper discusses the incompleteness of the SNOMED CT and proposes a noble metric which evaluates similarity among sibling nodes as a method to address this incompleteness. SNOMED CT encompasses an extensive range of medical terms, but it faces issues of ontology incompleteness, such as missing concepts in the hierarchy. We propose a noble metric for evaluating similarity among nodes within a node group, composed of multiple sibling nodes, to identify missing concepts, and identify groups with low similarity. Analyzing the similarity of sibling node groups in the March 2023 international release of SNOMED CT, the average similarity of 29,199 sibling node groups, which are sub-concepts of the clinical finding concept and are consist of two or more sibling nodes, was found to be 0.81. The group with the lowest similarity was associated with child concepts of poisoning, with a similarity of 0.0036.

본 논문에서는 SNOMED CT 용어체계가 가지는 불완전성을 논의하고 이를 유지하는 방법으로 형제 노드 간 유사성을 평가하는 지표를 제안한다. SNOMED CT는 방대한 양의 의학용어를 포함하고 있으나 계층구조 내 컨셉의 누락 등 온톨로지의 불완전성 문제가 존재한다. 누락된 컨셉 발견을 위해 다수의 노드로 구성된 형제 노드 그룹 내에서의 노드 간 유사도 평가를 위한 지표를 제안하고 유사도가 낮은 그룹을 도출하였다. 2023년 3월 SNOMED CT 국제 배포판에 적용하여 형제 노드 그룹들의 유사도를 분석한 결과 임상적 발견 컨셉의 하위 컨셉들 중 2개 이상의 형제 노드를 가진 29,199개의 형제 노드 그룹의 평균 유사도는 0.81로 나타났다. 반면, 유사도가 가장 낮은 그룹은 중독 컨셉의 자식 컨셉으로 그 유사도는 0.0036으로 확인되었다.

Keywords

Acknowledgement

이 논문은 2021년도 부산가톨릭대학교 교내연구비에 의하여 연구되었음

References

  1. H. Park, S. Yu, and H. Jung, "Strategies for Adopting and Implementing SNOMED CT in Korea," Healthcare Informatics Research, vol. 27, no. 1, Jan. 2021, pp. 3-10. https://doi.org/10.4258/hir.2021.27.1.3
  2. W. Del-Pinto, R. A. Schmidt, and Y. Gao, "Extracting Subontologies from SNOMED CT," In European Semantic Web Conf., Crete, Greece, May 2022, pp. 291-294.
  3. X. Hao, R. Abeysinghe, K. Roberts, and L. Cui, "Logical definition-based identification of potential missing concepts in SNOMED CT," BMC Medical Informatics and Decision Making, vol. 23, no. 1, May 2023, pp. 1-18.
  4. L. Cui, W. Zhu, S. Tao, J. T. Case, O. Bodenreider, and G. Q. Zhang, "Mining non-lattice subgraphs for detecting missing hierarchical relations and concepts in SNOMED CT," J. of the American Medical Informatics Association, vol. 24, no. 4, July 2017, pp. 788-798. https://doi.org/10.1093/jamia/ocw175
  5. X. Hao, R. Abeysinghe, F. Zheng, and L. Cui, "Leveraging non-lattice subgraphs for suggestion of new concepts for SNOMED CT," In 2021 IEEE Int. Conf. on Bioinformatics and Biomedicine, San Diego, USA, Dec. 2021, pp. 1805-1812.
  6. F. Zheng, J. Shi, and L. Cui, "A lexical-based approach for exhaustive detection of missing hierarchical IS-A relations in SNOMED CT," In American Medical Informatics Association Annual Symp. Proc., Virtual Event, Nov. 2020, pp. 1392-1401.
  7. X. Hao, R. Abeysinghe, J. Shi, and L. Cui, "A substring replacement approach for identifying missing IS-A relations in SNOMED CT," In 2022 IEEE Int. Conf. on Bioinformatics and Biomedicine, Las Vegas, USA, Dec. 2022, pp. 2611-2618.
  8. E. Lim, "IoB Based Scenario Application of Health and Medical AI Platform," J. of the Korea Institute of Electronic Communication Sciences, vol. 17, no. 6, Dec. 2022, pp. 1283-1292.
  9. H. Sim and H. Kim, "Development of Type 2 Prediction Prediction Based on Big Data," J. of the Korea Institute of Electronic Communication Sciences, vol. 18, no. 5, Oct. 2023, pp. 999-1008.
  10. R. Abeysinghe, F. Zheng, E. V. Bernstam, J. Shi, O. Bodenreider, and L. Cui, "A deep learning approach to identify missing is-a relations in SNOMED CT," J. of the American Medical Informatics Association, vol. 30, no. 3, Mar. 2023, pp. 475-484. https://doi.org/10.1093/jamia/ocac248