--- title: "Data" author: Sebastian Zeki date: '`r format(Sys.Date(), "%B %d, %Y")`' output: rmarkdown::html_document: theme: lumen toc: true toc_depth: 5 css: style.css vignette: > %\VignetteIndexEntry{Data} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8} --- ```{r setup, include=FALSE} library(pander) library(EndoMineR) knitr::opts_chunk$set(echo = TRUE) ``` ```{r global_options, include=FALSE} knitr::opts_chunk$set( warning=FALSE, message=FALSE) ``` ##Overview It is envisaged that different users will start at different points of their data preparation. THis section is intended to explain the fake data I have created so the type of data used for the examples can be better understood. There are several data files used. These are detailed below ## Gastroscopy #### Raw datasets: #####TheOGDReportFinal A dataset containing fake upper GI endoscopy reports. The report field is provided as a whole report without any fields having been already extracted. There are 2000 rows #####PathDataFrameFinal A dataset containing fake upper GI pathology reports. The report field is provided as a whole report without any fields having been already extracted. There are 2000 rows ####Pre-extracted datasets: #####Myendo This has been extracted using the Extractor method as follows from the raw text within Mypath: ```{r MyendoExtract, eval = FALSE} mywords <- c("OGDReportWhole","HospitalNumber","PatientName", "GeneralPractitioner","Dateofprocedure","Endoscopist", "Secondendoscopist","Medications","Instrument","ExtentofExam", "Indications","ProcedurePerformed","Findings") Extractor(TheOGDReportFinal,"OGDReportWhole",mywords) ``` ####Mypath This has been extracted using the Extractor method as follows from the raw text within Mypath: ```{r MypathExtract, eval = FALSE} mywords<-c("HospitalNumber","PatientName","DOB","GeneralPractitioner", "Dateofprocedure","ClinicalDetails","Macroscopicdescription", "Histology","Diagnosis") Extractor(PathDataFrameFinal,"PathReportWhole",mywords) ``` The original dataset has also been added as "PathReportWhole", ##Colonoscopy ###Raw datasets ####ColonFinal A dataset containing fake lower GI endoscopy reports. The report field is provided as a whole report without any fields having been already extracted. There are 2000 rows ####PathDataFrameFinalColon A dataset containing fake lower GI pathology reports. The report field is provided as a whole report without any fields having been already extracted. There are 2000 rows.