SectionSegmentation: Section Segmentation of Radiology Text Reports

In wlktan/LireNLPSystem: LireNLPSystem

Description

This function segments the imagereporttext column of a data frame into columns containing text history, exam, comparison, technique, finding, impression

Usage

SectionSegmentation(df, site = NULL, hst_rgx = NULL, exam_rgx = NULL,
  comparison_rgx = NULL, technique_rgx = NULL, body_rgx = NULL,
  imp_rgx = NULL, datetime_rgx = NULL,
  liretext_rgx = "(Among\\s*people\\s*(over|under|between)\\s*the\\s*age\\s*of.*$)|(Some\\s*findings\\s*are\\s*so\\s*common.*$)|(The\\s*following\\s*findings\\s*are\\s*so\\s*common\\s*in\\s*normal.*$)")

Arguments

df	Input data frame
site	1= , 2= , 3=, 4=; defaults to 2
hst_rgx	Defaults to regex for Kaiser
exam_rgx	Defaults to regex for Kaiser
comparison_rgx	Defaults to regex for Kaiser
technique_rgx	Defaults to regex for Kaiser
body_rgx	Defaults to regex for Kaiser
imp_rgx	Defaults to regex for Kaiser
liretext_rgx	Defaults to regular expressions for LIRE macro; don't change unless there's a good reason to.

Value

The input data frame with additional columns exam, comparison, technique, body, impression. The body column is always non-empty: the entire imagereporttext is returned in body if separation is not possible.

Examples

SectionSegmentation(input.df, site = 2)

wlktan/LireNLPSystem documentation built on May 27, 2019, 12:13 p.m.