Deep learning to predict risk of lateral skull base cerebrospinal fluid leak or encephalocele

Steven D Curry; Kieran S Boochoon; Geoffrey C Casazza; Daniel L Surdell; Justin A Cramer

doi:10.1007/s11548-024-03259-z

Deep learning to predict risk of lateral skull base cerebrospinal fluid leak or encephalocele

Int J Comput Assist Radiol Surg. 2024 Aug 29. doi: 10.1007/s11548-024-03259-z. Online ahead of print.

Authors

Steven D Curry¹, Kieran S Boochoon², Geoffrey C Casazza², Daniel L Surdell³, Justin A Cramer⁴

Affiliations

¹ Department of Otolaryngology, Head and Neck Surgery, University of Nebraska Medical Center, 981225 Nebraska Medical Center, Omaha, NE, 68198-1225, USA. scurry@houseclinic.com.
² Department of Otolaryngology, Head and Neck Surgery, University of Nebraska Medical Center, 981225 Nebraska Medical Center, Omaha, NE, 68198-1225, USA.
³ Department of Neurosurgery, University of Nebraska Medical Center, 988437 Nebraska Medical Center, Omaha, NE, 68198-8437, USA.
⁴ Department of Radiology, Mayo Clinic, 5777 E Mayo Boulevard, Phoenix, AZ, 85054, USA.

PMID: 39207718
DOI: 10.1007/s11548-024-03259-z

Abstract

Purpose: Skull base features, including increased foramen ovale (FO) cross-sectional area, are associated with lateral skull base spontaneous cerebrospinal fluid (sCSF) leak and encephalocele. Manual measurement requires skill in interpreting imaging studies and is time consuming. The goal of this study was to develop a fully automated deep learning method for FO segmentation and to determine the predictive value in identifying patients with sCSF leak or encephalocele.

Methods: A retrospective cohort study at a tertiary care academic hospital of 34 adults with lateral skull base sCSF leak or encephalocele were compared with 815 control patients from 2013-2021. A convolutional neural network (CNN) was constructed for image segmentation of axial computed tomography (CT) studies. Predicted FO segmentations were compared to manual segmentations, and receiver operating characteristic (ROC) curves were constructed.

Results: 295 CTs were used for training and validation of the CNN. A separate dataset of 554 control CTs was matched 5:1 on age and sex with the sCSF leak/encephalocele group. The mean Dice score was 0.81. The sCSF leak/encephalocele group had greater mean (SD) FO cross-sectional area compared to the control group, 29.0 (7.7) mm² versus 24.3 (7.6) mm² (P = .002, 95% confidence interval 0.02-0.08). The area under the ROC curve was 0.69.

Conclusion: CNNs can be used to segment the cross-sectional area of the FO accurately and efficiently. Used together with other predictors, this method could be used as part of a clinical tool to predict the risk of sCSF leak or encephalocele.

Keywords: Automatic segmentation; Convolutional neural network; Deep learning; Encephalocele; Idiopathic intracranial hypertension; Spontaneous CSF leak.