Retrieving Attitudes: Sentiment Analysis from Clinical Narratives

Physicians and nurses express their judgments and observations towards a patient’s health status in clinical narratives. Thus, their judgments are explicitly or implicitly included in patient records. To get impressions on the current health situation of a patient or on changes in the status, analysis and retrieval of this subjective content is crucial. In this paper, we approach this question as sentiment analysis problem and analyze the feasibility of assessing these judgments in clinical text by means of general sentiment analysis methods. Specifically, the word usage in clinical narratives and in a general text corpus is compared. The linguistic characteristics of judgments in clinical narratives are collected. Besides, the requirements for sentiment analysis and retrieval from clinical narratives are derived.