Research on the Extraction of Medical Terms from Electronic Medical Records
Download as PDF
Yao Yin, Fangfang Li, Xingliang Mao, Hao Wang
Electronic medical records contain a large number of patient-related medical information. It is very important for doctors to diagnose diseases by mining useful information in electronic medical records. The extraction of medical terms is an important step in information mining from electronic medical records. Eyes are very important organs of human beings, therefore, this paper taking electronic medical records in the ophthalmology department in a hospital as the object, the main works of this paper are as follows. Firstly, the Conditional Random Field (CRF) model and its main steps of terms extraction are introduced. Secondly, the data set used in this paper and its annotation under the guidance of professional doctors are introduced. Finally, the extraction of medical terms by the CRF model is described in detail, and its result was compared with Hidden Markov Model (HMM) model. The experiment results show that CRF model based medical terms extraction in electronic medical records has made good performance.
Medical Terms, CRF Model, Hmm Model, Electronic Medical Record