No abstract

Background: The combination of high-frequency oscillations (HFOs) with single-mode imaging methods has been proved useful in identifying epileptogenic zones, whereas few studies have examined HFOs combined with multimodal imaging methods. The aim of this study was to evaluate the prognostic value of ripples, an HFO subtype with a frequency of 80 to 200 Hz is combined with multimodal imaging methods in predicting epilepsy surgery outcome. Methods: HFOs were analyzed in 21 consecutive medically refractory epilepsy patients who underwent epilepsy surgery. All patients underwent positron emission tomography (PET) and deep electrode implantation for stereo-electroencephalography (SEEG); 11 patients underwent magnetoencephalography (MEG). Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy in predicting surgical outcome were calculated for ripples combined with PET, MEG, both PET and MEG, and PET combined with MEG. Kaplan-Meier survival analyses were conducted in each group to estimate prognostic value. Results: The study included 13 men and 8 women. Accuracy for ripples, PET, and MEG alone in predicting surgical outcome was 42.9%, 42.9%, and 81.8%, respectively. Accuracy for ripples combined with PET and MEG was the highest. Resection of regions identified by ripples, MEG dipoles, and combined PET findings was significantly associated with better surgical outcome (P < 0.05). Conclusions: Intracranial electrodes are essential to detect regions which generate ripples and to remove these areas which indicate good surgical outcome for medically intractable epilepsy. With the assistance of presurgical noninvasive imaging examinations, PET and MEG, for example, the SEEG electrodes would identify epileptogenic regions more effectively.

ObjectiveThis study aimed to differentiate temporal‐plus epilepsy (TPE) from temporal lobe epilepsy (TLE) using extraction of radiomics features from three‐dimensional magnetization‐prepared rapid acquisition gradient echo (3D‐MPRAGE) imaging data.MethodsData from patients with TLE or TPE who underwent epilepsy surgery between January 2019 and January 2021 were retrospectively analyzed. Thirty‐three regions of interest in the affected hemisphere of each patient were defined on 3D‐MPRAGE images. A total of 3531 image features were extracted from each patient. Four feature selection methods and 10 machine learning algorithms were used to build 40 differentiation models. Model performance was evaluated using receiver operating characteristic analysis.ResultsEighty‐two patients were included for analysis, 47 with TLE and 35 with TPE. The model combining logistic regression and the relief selection method had the best performance (area under the receiver operating characteristic curve, .779; accuracy, .875; sensitivity, .800; specificity, .929; positive predictive value, .889; negative predictive value, .867).SignificanceRadiomics analysis can differentiate TPE from TLE. The logistic regression classifier trained with radiomics features extracted from 3D‐MPRAGE images had the highest accuracy and best performance.