VP05.06: Differential diagnosis of deep rectal endometriosis by ultrasound

Abstract

The aim of this study was to compare the accuracy of seven classical machine learning (ML) models trained with ultrasound (US) soft markers to raise suspicion of forniceal endometriotic involvement. Methods: Input data to the models was retrieved from a database of 194 patients submitted to surgery for the suspicion of presence of deep endometriosis. The following models have been tested: k-nearest neighbours' algorithm (k-NN), Naive Bayes, Neural Networks (NNET-neuralnet), support vector machine (SVM), decision tree, random forest, and logistic regression. The data driven strategy has been to split randomly the complete dataset in two different datasets. The training dataset and the test dataset with a 67% and 33% of the original cases respectively. All models were trained on the training dataset and the predictions have been evaluated using the test dataset. The best model was chosen based on the best AUC demonstrated on the test dataset. The information used in all the models were: age; presence of US signs of uterine adenomyosis; presence of an endometrioma; adhesions of the ovary to the uterus; presence of ''kissing ovaries''; absence of sliding sign. All models have been trained using CARET package in R with ten repeated 10-fold cross-validation. Sensitivity, and specificity, were calculated using a Youden index threshold. Results: 28 women had a surgical diagnosis of forniceal endometriosis. In term of diagnostic accuracy, the best model was the Neural Net (AUC, 0.73; sensitivity, 0.78; specificity 0.70) but without significant difference with the others. Conclusions: The accuracy of ultrasound soft markers in raising suspicion of forniceal endometriosis using artificial intelligence (AI) models showed similar results to the logistic model.This study was partly supported by Fondazione di Sardegna grant F74I19001010007.