The Reproducibility of Ultrasonographic Findings of Rectosigmoid Endometriosis Among Examiners With Different Level of Expertise.

OBJECTIVE: To analyze the reproducibility of ultrasonographic (US) findings of rectosigmoid endometriosis among examiners with different level of expertise using stored three-dimensional (3D) volumes of the posterior compartment of the pelvis as a part of SANABA (Sardinia-Navarra-Barcelona) collaborative study. MATERIALS AND METHODS: Six examiners in 3 academic Department of Obstetrics and Gynecology, with different levels of experience and blinded to each other, evaluated 60 stored 3D volumes from the posterior compartment of the pelvis and looked for the presence or absence of features of rectosigmoid endometriotic lesions defined as an irregular hypoechoic nodule with or without hypoechoic foci at the level of the muscularis propria of the anterior wall rectum sigma. Multiplanar view and virtual navigation were used. All examiners had to assess the 3D volume of posterior compartment of the pelvis and classify it as present or absent disease. To analyze intra-observer and the inter-observer agreements, each examiner performed the assessment twice with a 2-week interval between the first and second assessments. Reproducibility was assessed by calculating the weighted Kappa index. RESULTS: Intra-observer reproducibility was moderate to very good for all observers (Kappa index ranging from 0.49 to 0.96) associated with a good diagnostic accuracy of each reader. Inter-observer reproducibility was fair to very good (Kappa index range: 0.21-0.87). CONCLUSIONS: The typical US sign of rectosigmoid endometriosis is reasonably recognizable to observers with different level of expertise when assessed in stored 3D volumes.

Artificial intelligence (AI) in the detection of rectosigmoid deep endometriosis.

OBJECTIVES: 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 endometriotic bowel involvement. MATERIALS AND METHODS: Input data to the models was retrieved from a database of a previously published study on bowel endometriosis performed on 333 patients. The following models have been tested: k-nearest neighbors 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 accuracy 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. Accuracy, Sensitivity, Specificity, positive (PPV) and negative (NPV) predictive value were calculated using a 50 % threshold. Presence of intestinal involvement was defined in all cases in the test dataset with an estimated probability greater than 0.5. RESULTS: In our previous study from where the inputs were retrieved, 106 women had a final expert US diagnosis of rectosigmoid endometriosis. In term of diagnostic accuracy the best model was the Neural Net (Accuracy, 0.73; sensitivity, 0.72; specificity 0.73; PPV 0.52; and NPV 0.86) but without significant difference with the others. CONCLUSIONS: The accuracy of ultrasound soft markers in raising suspicion of rectosigmoid endometriosis using Artificial Intelligence (AI) models showed similar results to the logistic model.