Negation Detection for Clinical Text Mining in Russian

TL;DR

This paper presents a machine learning-based negation detection module for Russian clinical texts, improving medical record analysis by accurately identifying negated or absent disease mentions, which enhances predictive modeling in healthcare.

Contribution

It introduces a corpus-free gradient boosting approach for negation detection in Russian clinical texts, addressing the lack of NLP tools in this language and domain.

Findings

Achieved average F-score of 0.81 to 0.93 in negation detection

Demonstrated improved prediction of surgery presence in acute coronary syndrome patients

Validated the effectiveness of the method across five diseases

Abstract

Developing predictive modeling in medicine requires additional features from unstructured clinical texts. In Russia, there are no instruments for natural language processing to cope with problems of medical records. This paper is devoted to a module of negation detection. The corpus-free machine learning method is based on gradient boosting classifier is used to detect whether a disease is denied, not mentioned or presented in the text. The detector classifies negations for five diseases and shows average F-score from 0.81 to 0.93. The benefits of negation detection have been demonstrated by predicting the presence of surgery for patients with the acute coronary syndrome.