Untitled Item

R script for the analysis of the histopathological features of thoracic endometriosis (TE) using ggplot2, DT, pROC and vcd packages in R statistical software. The aim is to review the histopathological findings in TE, summarise the diagnostic features, identify any major histo-morphological indicator(s) hitherto unrecognised as such, suggest diagnostic algorithms; all with the aim of improving the diagnostic capacity and reducing observer error even where the clinical suspicion is low. Statistical methods applied include chi-squared test of independence, Cramér’s V test of strength of association and the area under the curve of the receiver operating characteristic (AUC-ROC) curve. The final conclusion is summarised with a Venn plot identifying endometrioid glands, ceroid macrophages, siderophages and lymphoid clusters as the major markers of TE.