AUTHOR=Villalonga J. F. , Solari D. , Cuocolo R. , De Lucia V. , Ugga L. , Gragnaniello C. , Pailler J. I. , Cervio A. , Campero A. , Cavallo L. M. , Cappabianca P. TITLE=Clinical application of the “sellar barrier’s concept” for predicting intraoperative CSF leak in endoscopic endonasal surgery for pituitary adenomas with a machine learning analysis JOURNAL=Frontiers in Surgery VOLUME=Volume 9 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/surgery/articles/10.3389/fsurg.2022.934721 DOI=10.3389/fsurg.2022.934721 ISSN=2296-875X ABSTRACT=Background: Recently, it has been defined that the entity of sellar barrier could be identified as a predictor of CSF intraoperative leakage. The aim of this study is to validate the application of the sellar barrier´s concept for predicting intraoperative CSF leak in endoscopic endonasal surgery for pituitary adenomas with a machine learning approach. Methods: We conducted a prospective cohort study, from June 2019 to September 2020: data of 155 patients with pituitary subdiaphragmatic adenoma operated by mean of endoscopic approach at the Division of Neurosurgery, Università degli Studi di Napoli "Federico II", were included. Preoperative magnetic resonance images (MRI) and intraoperative findings were analyzed. After processing patient data, the experiment has been conducted as a novelty detection problem, splitting outliers (i.e., patients with intraoperative fistula, n=11/155) and inliers in separate datasets, the latter further separated into training (n=115/144) and inlier test (n=29/144) datasets. The machine learning analysis was performed using different novelty detection algorithms (isolation forest, local outlier factor, one-class SVM), whose performance was assessed separately and as an ensemble on the inlier and outlier test sets. Results: According to the type of sellar barrier, the patients were separated into 2 groups, i.e., strong, and weak barrier: a third category of mixed barrier was defined whether a case was neither non-weak non-strong criteria. Significant differences between the 3 datasets were found for: Knosp classification score (p=0.0015), MRI barrier: Strong (p=1.405e-06), MRI barrier: Weak (p=4.487e-08), Intraoperative barrier: Strong (p=2.788e-07), Intraoperative barrier: Weak (p=2.191e-10). We recorded 11 cases of intraoperative leakage and it was occurring in the vast majority of cases in those patients presenting a weak sellar barrier (p=4.487e-08) at preoperative MRI. Accuracy, sensitivity, and specificity for outlier detection were 0.70, 0.64 and 0.72 for IF, 0.85, 0.45 and 1.00 for LOF, 0.83, 0.64 and 0.90 for oSVM and 0.83, 0.55 and 0.93 for the ensemble. Conclusions: There is a true correlation between the type of sellar barrier at MRI and its in vivo features as observed during endoscopic endonasal surgery. The novelty detection models highlighted differences between patients who developed an intraoperative CSF leak and those who did not.