Clinical Medicine Retrieval System Oriented to Treatment Plan

In order to overcome the problem of low retrieval efficiency in the field of clinical medicine. According to the TREC evaluation task, the paper proposes a method for constructing a clinical medical retrieval system for treatment plan to improve retrieval efficiency, aiming at being able to obtain effective treatment options by entering patient-related information to facilitate the solution given by physicians or biomedical scientists. For the specificity of the dataset, this method considers genetic and demographic characteristics to preprocess documents and topics, and improve the query using the query expansion technique based on Medical Subject Headings (MeSH); Set up a Lucene-based retrieval system to read in the converted files, and get the primary feedback results by inputting the processed query into the retrieval, and then filter the result document to get the final result document. Experimental results show that this method can obtain documents with high relevance, which is of great significance to the field of clinical medicine information retrieval.