2023 Fiscal Year Final Research Report
Sentiment analysis of nursing notes and association with clinical outcomes in palliative care
| Project/Area Number |
22K21167
|
|---|
| Research Category |
Grant-in-Aid for Research Activity Start-up
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
0908:Society medicine, nursing, and related fields
|
|---|
| Research Institution | Tohoku University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2022-08-31 – 2024-03-31
|
| Keywords | 緩和ケア / 電子カルテ / 自然言語処理 / 感情分析 |
|---|
| Outline of Final Research Achievements |
In this study, we decided to utilize sentiment analysis, one of the natural language processing techniques, for medical records, mainly nursing records in electronic medical records, and to examine its usefulness.
This study was a secondary analysis of electronic medical record data collected in another study, and data collection had already been completed. As a preliminary study, we decided to evaluate the feasibility of applying sentiment analysis to a questionnaire survey with a small number of data.As a result, through discussions with other nursing researchers, the difficulties in applying emotion analysis became apparent. In particular, it was difficult to determine the criteria for assigning an sentiment score, which is the correct label for the data, and this was not realized.
|
| Free Research Field |
緩和ケア
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究の意義の1つとして、既存の感情分析を緩和ケア領域での実臨床や研究での応用することは難しいことが明らかになった。既存の感情分析はソーシャルメディアなど臨床とは異なる状況に適しており臨床ではそのままの応用は難しい。特に緩和ケア領域では、終末期であり「死」や「看取り」など一般的にはネガティブな表現が多く使用される。緩和ケア以外の医学領域でも、状況にはよるがそれら表現はネガティブなものと考えられるかもしれない。しかし、緩和ケアに携わる医療者ではそれらはネガティブではない場合が多い。そのため、今後感情分析を応用するためには緩和ケアに特化した新たな手法を検討する必要があると考える。
|
