A hybrid recommendation method based on medical knowledge base and electronic medical records(PDF)
《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]
- Issue:
- 2019年第4期
- Page:
- 492-496
- Research Field:
- 其他（激光医学等）
- Publishing date:
Info
- Title:
- A hybrid recommendation method based on medical knowledge base and electronic medical records
- Author(s):
- HU Dehua; FANG Hao; ZHOU Yukui; LUO Aijing
- Department of Medical Information, School of Life Sciences, Central South University, Changsha 410083, China
- Keywords:
- Keywords: medical knowledge base; electronic medical record; collaborative filtering; content filtering
- PACS:
- R319；TP391
- DOI:
- DOI:10.3969/j.issn.1005-202X.2019.04.024
- Abstract:
- Abstract: Objective To explore an effective method for providing recommendation knowledge related to diagnosis and treatment for patients based on medical knowledge base and electronic medical records. Methods The content filtering method of medical knowledge base was combined with electronic medical records to establish an integrated scoring matrix. The fusion of the results obtained with patient-based collaborative filtering and diagnosis-based collaborative filtering was used to construct a hybrid recommendation method, so as to reduce the error of prediction results. With mean absolute percentage error as the evaluation index, the prediction results obtained with hybrid recommendation method, patient-based collaborative filtering and diagnosis-based diagnostic collaborative filtering were evaluated. Results The average absolute percent error of hybrid recommendation method was lower than that of patient-based collaborative filtering and diagnosis-based collaborative filtering. The limitation of this study is that the types of hypertensive diseases in medical knowledge base were less than those in electronic medical records, and that the effects of the disease coverage on the recommendation results remained to be further explored. Conclusion The hybrid recommendation method which is superior to patient-based collaborative filtering and diagnosis-based diagnostic collaborative filtering can be used to provide medical knowledge for patients, having a good application prospect.
Last Update: 2019-04-23