DOI QR코드

DOI QR Code

Analysis on Correlation between Prescriptions and Test Results of Diabetes Patients using Graph Models and Node Centrality

그래프 모델과 중심성 분석을 이용한 당뇨환자의 처방 및 검사결과의 상관관계 분석

  • 유강민 (서울대학교 컴퓨터공학부) ;
  • 박성찬 (서울대학교 컴퓨터공학부) ;
  • 이수진 (서울대학교 임상약리학교실) ;
  • 유경상 (서울대학교 임상약리학교실) ;
  • 이상구 (서울대학교 컴퓨터공학부)
  • Received : 2015.03.18
  • Accepted : 2015.05.08
  • Published : 2015.07.15

Abstract

This paper presents the results and the process of extracting correlations between events of prescriptions and examinations using graph-modeling and node centrality measures on a medical dataset of 11,938 patients with diabetes mellitus. As the data is stored in relational form, RDB2Graph framework was used to construct effective graph models from the data. Personalized PageRank was applied to analyze correlation between prescriptions and examinations of the patients. Two graph models were constructed: one that models medical events by each patient and another that considers the time gap between medical events. The results of the correlation analysis confirm current medical knowledge. The paper demonstrates some of the note-worthy findings to show the effectiveness of the method used in the current analysis.

본 논문은 11,938명의 당뇨환자 의료데이터를 그래프 모델로 변환하고 중심성 분석 기법으로 처방과 검사결과 간 상관관계를 추출해내는 과정에 대해 다루고 있다. 관계형 데이터베이스로 저장되어있는 데이터를 RDB2Graph 프레임워크를 사용하여 유의미한 그래프로 변환하였다, 변환된 그래프에 Personalized PageRank를 적용하여 처방과 검사 간 상관관계를 분석했다. 사용된 그래프 모델에는 환자 별 의료 기록 모델과 의료 기록의 시간적 간격을 고려한 모델이 있다. 분석 결과 기존의 의학적 지식에 부합하는 상관관계를 다수 발견할 수 있었으며, 본 논문에서는 발견한 상관관계 중 주요 사례를 소개하여 본 분석 방법의 유효함을 보인다.

Keywords

Acknowledgement

Supported by : 한국연구재단(NRF)

References

  1. S. Lee, R. Park, "Data Mining Concepts and Principles in Pharmacology," Proc. of the Korean Society of Medical Informatics, pp. 175-189, 2009. (in Korean)
  2. C. C. Aggarwal, H. Wang et al., Managing and Mining Graph Data, pp. 1-13, Springer, New York, 2010.
  3. S. Auer, S. Dietzold, J. Lehmann, S. Hellmann, and D. Aumueller, "Triplify: light-weight linked data publication from relational databases," Proc. of the 18th international conference on World Wide Web, pp. 621-630, 2009.
  4. J. Park, A Graph-based Framework for Processing Keyword Queries over Relational Databases, PhD thesis, Seoul National University, 2012. (in Korean)
  5. S. Lee, S. Park, M. Kahng, and S. Lee, "Pathrank: Ranking nodes on a heterogeneous graph for flexible hybrid recommender systems," Expert Systems with Applications, Vol. 40, No. 2, pp. 684-697, Feb. 2013. https://doi.org/10.1016/j.eswa.2012.08.004
  6. H. Lee, J. Myung, S. Lee, "A personalized ranking algorithm using user defined paths over a heterogeneous graph," Proc. of the KIISE Korea Computer Congress 2013, pp. 268-270, 2013. (in Korean)
  7. S. Das, S. Sundara, and R. Cyganiak. (2012, September 01). "R2RML: RDB to RDF Mapping Language" [Online], Available: http://www.w3.org/TR/r2rml, (downloaded 2015, Mar. 3).
  8. K. Yoo, S. Park, and S. Lee, "RDB2Graph: A Generic Framework for Modeling Relational Databases as Graphs," Poster presented at The 4th Joint International Semantic Technology Conference, 2014.
  9. L. Page, S. Brin, R. Motwani, and T. Winograd, "The PageRank Citation Ranking: Bringing Order to the Web," 1999.
  10. D. Fogaras, B. Racz, K. Csalogany, and T. Sarlos, "Towards scaling fully personalized pagerank: Algorithms, lower bounds, and experiments," Internet Mathematics, Vol. 2, No. 3, pp. 333-358, Sep. 2005. https://doi.org/10.1080/15427951.2005.10129104
  11. T. C. Alba-Loureiro, S. M. Hirabara, J. R. Mendoca, R. Curi, and T. C. Pithon-Curi, "Diabetes causes marked changes in function and metabolism of rat neutrophils," Journal of Endocrinology, Vol. 188, No. 2, pp. 295-303, Jul. 2006. https://doi.org/10.1677/joe.1.06438
  12. R. Otton, J. R. Mendonca, and R. Curi, "Diabetes causes marked changes in lymphocyte metabolism," Journal of endocrinology, Vol. 174, No. 1, pp. 55-61, Jul. 2002. https://doi.org/10.1677/joe.0.1740055