Abstract:Accurate segmentation of surgical resection sites is critical for clinical assessments and neuroimaging research applications, including resection extent determination, predictive modeling of surgery outcome, and masking image processing near resection sites. In this study, an automated resection cavity segmentation algorithm is developed for analyzing postoperative MRI of epilepsy patients and deployed in an easy-to-use graphical user interface (GUI) that estimates remnant brain volumes, including postsurgical hippocampal remnant tissue. This retrospective study included postoperative T1-weighted MRI from 62 temporal lobe epilepsy (TLE) patients who underwent resective surgery. The resection site was manually segmented and reviewed by a neuroradiologist (JMS). A majority vote ensemble algorithm was used to segment surgical resections, using 3 U-Net convolutional neural networks trained on axial, coronal, and sagittal slices, respectively. The algorithm was trained using 5-fold cross validation, with data partitioned into training (N = 27) testing (N = 9), and validation (N = 9) sets, and evaluated on a separate held-out test set (N = 17). Algorithm performance was assessed using Dice-Sørensen coefficient (DSC), Hausdorff distance, and volume estimates. Additionally, we deploy a fully-automated, GUI-based pipeline that compares resection segmentations with preoperative imaging and reports estimates of resected brain structures. The cross-validation and held-out test median DSCs were 0.84 ± 0.08 and 0.74 ± 0.22 (median ± interquartile range) respectively, which approach inter-rater reliability between radiologists (0.84-0.86) as reported in the literature. Median 95 % Hausdorff distances were 3.6 mm and 4.0 mm respectively, indicating high segmentation boundary confidence. Automated and manual resection volume estimates were highly correlated for both cross-validation (r = 0.94, p < 0.0001) and held-out test subjects (r = 0.87, p < 0.0001). Automated and manual segmentations overlapped in all 62 subjects, indicating a low false negative rate. In control subjects (N = 40), the classifier segmented no voxels (N = 33), <50 voxels (N = 5), or a small volumes<0.5 cm3 (N = 2), indicating a low false positive rate that can be controlled via thresholding. There was strong agreement between postoperative hippocampal remnant volumes determined using automated and manual resection segmentations (r = 0.90, p < 0.0001, mean absolute error = 6.3 %), indicating that automated resection segmentations can permit quantification of postoperative brain volumes after epilepsy surgery. Applications include quantification of postoperative remnant brain volumes, correction of deformable registration, and localization of removed brain regions for network modeling.

Automated segmentation of epilepsy surgical resection cavities: comparison of four methods to manual segmentation

Automated segmentation of epilepsy surgical resection cavities: comparison of four methods to manual segmentation

Deep learning-based automated segmentation of resection cavities on postsurgical epilepsy MRI

Simulation of Brain Resection for Cavity Segmentation Using Self-Supervised and Semi-Supervised Learning

Fully automatic segmentation of brain lacunas resulting from resective surgery using a 3D deep learning model

A self-supervised learning strategy for postoperative brain cavity segmentation simulating resections

Comparison of different automatic solutions for resection cavity segmentation in postoperative MRI volumes including longitudinal acquisitions

Application of MRI Post-processing in Presurgical Evaluation of Non-lesional Cingulate Epilepsy.

Evaluating Traditional, Deep Learning, and Subfield Methods for Automatically Segmenting the Hippocampus from MRI

Hippocampus segmentation on epilepsy and Alzheimer's disease studies with multiple convolutional neural networks

Automated volumetric assessment of pituitary adenoma

Machine learning algorithm for predicting seizure control after temporal lobe resection using peri-ictal electroencephalography

Deep learning for automatic segmentation of vestibular schwannoma: a retrospective study from multi-center routine MRI

Incorporation of quantitative MRI in a model to predict temporal lobe epilepsy surgery outcome

Deep Learning for Automatic Segmentation of Vestibular Schwannoma: A Retrospective Study from Multi-Centre Routine MRI

Open access segmentations of intraoperative brain tumor ultrasound images

Automated deep learning segmentation of high-resolution 7 Tesla postmortem MRI for quantitative analysis of structure-pathology correlations in neurodegenerative diseases

Machine Learning-Derived Multimodal Neuroimaging of Presurgical Target Area to Predict Individual's Seizure Outcomes After Epilepsy Surgery

Postoperative glioblastoma segmentation: Development of a fully automated pipeline using deep convolutional neural networks and comparison with currently available models

The Imaging Database for Epilepsy And Surgery (IDEAS)

Machine Learning for Localizing Epileptogenic-Zone in the Temporal Lobe: Quantifying the Value of Multimodal Clinical-Semiology and Imaging Concordance