Search

OBJECTIVE: In recent years, natural language processing (NLP) techniques have progressed, and their application in the medical field has been tested. However, the use of NLP to detect symptoms from medical progress notes written in Japanese, remains limited. We aimed to detect 2 gastrointestinal symptoms that interfere with the continuation of chemotherapy--nausea/vomiting and diarrhea--from progress notes using NLP, and then to analyze factors affecting NLP. MATERIALS AND METHODS: In this study, 200 patients were randomly selected from 5277 patients who received intravenous injections of cytotoxic anticancer drugs at Kagawa University Hospital, Japan, between January 2011 and December 2018. We aimed to detect the first occurrence of nausea/vomiting (Group A) and diarrhea (Group B) using NLP. The NLP performance was evaluated by the concordance with a review of the physicians' progress notes used as the gold standard. RESULTS: Both groups showed high concordance: 83.5% (95% confidence interval [CI] 74.1-90.1) in Group A and 97.7% (95% CI 91.3-99.9) in Group B. However, the concordance was significantly better in Group B (P = .0027). There were significantly more misdetection cases in Group A than in Group B (15.3% in Group A; 1.2% in Group B, P = .0012) due to negative findings or past history. CONCLUSION: We detected occurrences of nausea/vomiting and diarrhea accurately using NLP. However, there were more misdetection cases in Group A due to negative findings or past history, which may have been influenced by the physicians' more frequent documentation of nausea/vomiting. [ABSTRACT FROM AUTHOR]

We performed the multi-year project to collect discharge summary from multiple hospitals and made the big text database to build a common document vector space, and developed various applications. We extracted 243,907 discharge summaries from seven hospitals. There was a difference in term structure and number of terms between the hospitals, however the differences by disease were similar. We built the vector space using TF-IDF method. We performed a cross-match analysis of DPC selection among seven hospitals. About 80% cases were correctly matched. The use of model data of other hospitals reduced selection rate to around 10%; however, integrated model data from all hospitals restored the selection rate.

We started a multi-year project to collect discharge summaries from multiple hospitals and create a big text database to build a common document vector space, and develop various applications such as the autoselection of the disease. As the first step, we extracted discharge summary from two hospitals. Using a text mining method, we carried out a DPC selection. There was a difference in term structure and number of terms between the discharge summaries from both hospitals. Nevertheless, the selection rate of the disease is resembled closely.