Machine-learning for the prediction of one-year seizure recurrence based on routine electroencephalography

Abstract

Predicting seizure recurrence risk is critical to the diagnosis and management of epilepsy. Routine electroencephalography (EEG) is a cornerstone of the estimation of seizure recurrence risk. However, EEG interpretation relies on the visual identification of interictal epileptiform discharges (IEDs) by neurologists, with limited sensitivity. Automated processing of EEG could increase its diagnostic yield and accessibility. The main objective was to develop a prediction model based on automated EEG processing to predict one-year seizure recurrence in patients undergoing routine EEG. We retrospectively selected a consecutive cohort of 517 patients undergoing routine EEG at our institution (training set) and a separate, temporally shifted cohort of 261 patients (testing set). We developed an automated processing pipeline to extract linear and non-linear features from the EEGs. We trained machine learning algorithms on multichannel EEG segments to predict one-year seizure recurrence. We evaluated the impact of IEDs and clinical confounders on performances and validated the performances on the testing set. The receiver operating characteristic area-under-the-curve for seizure recurrence after EEG in the testing set was 0.63 (95% CI 0.55–0.71). Predictions were still significantly above chance in EEGs with no IEDs. Our findings suggest that there are changes other than IEDs in the EEG signal embodying seizure propensity.

The value of linear and non-linear quantitative EEG analysis in paediatric epilepsy surgery: a machine learning approach.

“…Out of the three SETs used, results coming from SET3 revealed the lowest mean square error (MSE = 9.2), indicating that our sample size was adequate to achieve a stable model. Our ANN results demonstrated an accuracy of 64.8% in predicting seizure outcome using SET 3, this was also consistent with previous published data [30,42,43].…”

“…Lemoine at al evaluated the combination of linear and non-linear EEG features in predicting one-year seizure recurrence after a routine EEG using four BML algorithms (general linear model, support vector machine, Random Forest and LightGBM), achieving the 62-67% of accuracy [30]. No studies showed that a combination of linear and non-linear interictal scalp EEG features had a high predictive accuracy for post-surgical seizure outcome in paediatric patients.…”

The value of linear and non-linear quantitative EEG analysis in paediatric epilepsy surgery: a machine learning approach

“…Lemoine et al investigated how combining linear and non-linear EEG features could predict seizure recurrence within 1 year after EEG, using four BML algorithms (general linear model, support vector machine, Random Forest and LightGBM). They achieved an accuracy rate between 62 and 67% 30 . Previously, no studies had showed that such a combination of linear and non-linear interictal scalp EEG features could accurately predict surgical outcome in children with epilepsy.…”

“…We applied two different approaches. To compute SET 1 we performed a single channel extraction: all predicting values are computed at each channel of standard 10–20 montage and epoch as reported in Lemoine et al 30 . To compute SET 2 and 3 we extracted all values as the average across all channels according to Lin et al 31 , 32 .…”

Polygenic risk scores as a marker for epilepsy risk across lifetime and after unspecified seizure events

“…In this study we could demonstrate that common genetic variants, in the form of PRSGGE have a significant quantitative effect on GGE lifetime cumulative incidence that we could reproduce in several biobanks with hazard ratios of 3-4 for the upper tails of the PRSGGE distribution in line with previous studies 23 . Prediction is modest, but comparable to performance of models using standard variables such as clinical factors or EEG 34,35 , so we expect PRSs to have most utility as a supportive but not standalone tool.…”