Abstract found on PubMed

Objective: The accurate prediction of seizure freedom after epilepsy surgery remains challenging. We investigated if 1) training more complex models, 2) recruiting larger sample sizes, or 3) using data-driven selection of clinical predictors would improve our ability to predict post-operative seizure outcome using clinical features. We also conducted the first substantial external validation of a machine learning model trained to predict post-operative seizure outcome.

Methods: We performed a retrospective cohort study of 797 children who had undergone resective or disconnective epilepsy surgery at a tertiary center. We extracted patient information from medical records and trained three models – a logistic regression, a multilayer perceptron, and an XGBoost model – to predict one-year post-operative seizure outcome on our dataset. We evaluated the performance of a recently published XGBoost model on the same patients. We further investigated the impact of sample size on model performance, using learning curve analysis to estimate performance at samples up to N=2,000. Finally, we examined the impact of predictor selection on model performance.

Results: Our logistic regression achieved an accuracy of 72% (95% CI=68-75%,AUC=0.72), while our multilayer perceptron and XGBoost both achieved accuracies of 71% (95% CI_MLP =67-74%,AUC_MLP =0.70; 95% CI_{XGBoost own} =68-75%,AUC_{XGBoost own} =0.70). There was no significant difference in performance between our three models (all P>0.4) and they all performed better than the external XGBoost, which achieved an accuracy of 63% (95% CI=59-67%,AUC=0.62; P_LR=0.005,P_MLP =0.01,P_{XGBoost own} =0.01) on our data. All models showed improved performance with increasing sample size, but limited improvements beyond our current sample. The best model performance was achieved with data-driven feature selection.

Significance: We show that neither the deployment of complex machine learning models nor the assembly of thousands of patients alone is likely to generate significant improvements in our ability to predict post-operative seizure freedom. We instead propose that improved feature selection alongside collaboration, data standardization, and model sharing is required to advance the field.

Abstract found on Wiley Online Library

Genetic variants in KCNQ2 are associated with a range of epilepsies, from self-limited (familial) neonatal-infantile epilepsy to developmental and epileptic encephalopathy (DEE). We retrospectively reviewed clinical data from eight patients with KCNQ2-related DEE who were treated with ezogabine. Treatment was initiated at a median age of 8 months (range 7 weeks to 2.5 years) and continued for a median of 2.6 years (range 7 months to 4.5 years). Five individuals had daily seizures at baseline and experienced at least 50% seizure reduction with treatment, sustained in four. One individual with two to four yearly seizures improved to rare events. Two individuals were seizure-free; treatment targeted cognition and development. Developmental improvements were reported in all eight patients. Weaning of ezogabine was associated with increased seizure frequency (N=4), agitation and irritability (N=2), poor sleep (N=1), and developmental regression (N=2). These data suggest that treatment with ezogabine is effective at reducing seizure burden and is associated with improved development. Minimal side effects were observed. Weaning was associated with increased seizures and behavioral disturbances in a subset. An approach targeting potassium channel dysfunction with ezogabine is warranted in patients with KCNQ2-related DEE.

Abstract found on PubMed

Objective: Automated detection of spikes and seizures has been a subject of research for several decades now. There have been important advances, yet automated detection in EMU (Epilepsy Monitoring Unit) settings has not been accepted as standard practice. We intend to implement this software at our EMU and so carried out a qualitative study to identify factors that hinder (‘barriers’) and facilitate (‘enablers’) implementation.

Method: Twenty-two semi-structured interviews were conducted with 14 technicians and neurologists involved in recording and reporting EEGs and eight neurologists who receive EEG reports in the outpatient department. The study was reported according to the Consolidated Criteria for Reporting Qualitative Studies (COREQ).

Results: We identified 14 barriers and 14 enablers for future implementation. Most barriers were reported by technicians. The most prominent barrier was lack of trust in the software, especially regarding seizure detection and false positive results. Additionally, technicians feared losing their EEG review skills or their jobs. Most commonly reported enablers included potential efficiency in the EEG workflow, the opportunity for quantification of EEG findings and the willingness to try the software.

Conclusions: This study provides insight into the perspectives of users and offers recommendations for implementing automated spike and seizure detection in EMUs.

Abstract found on Wiley Online Library

Many state-of-the-art methods for seizure prediction, using the electroencephalogram, are based on machine learning models that are black boxes, weakening the trust of clinicians in them for high-risk decisions. Seizure prediction concerns a multidimensional time-series problem that performs continuous sliding window analysis and classification. In this work, we make a critical review of which explanations increase trust in models’ decisions for predicting seizures. We developed three machine learning methodologies to explore their explainability potential. These contain different levels of model transparency: a logistic regression, an ensemble of fifteen Support Vector Machines, and an ensemble of three Convolutional Neural Networks. For each methodology, we evaluated quasi-prospectively the performance in 40 patients (testing data comprised 2055 hours and 104 seizures). We selected patients with good and poor performance to explain the models’ decisions. Then, with Grounded Theory, we evaluated how these explanations helped specialists (data scientists and clinicians working in epilepsy) to understand the obtained model dynamics. We obtained four lessons for better communication between data scientists and clinicians. We found that the goal of explainability is not to explain the system’s decisions but to improve the system itself. Model transparency is not the most significant factor in explaining a model decision for seizure prediction. Even when using intuitive and state-of-the-art features, it is hard to understand brain dynamics and their relationship with the developed models. We achieve an increase in understanding by developing, in parallel, several systems that explicitly deal with signal dynamics changes that help develop a complete problem formulation.

Abstract found on PubMed

Purpose: Wearable automated detection devices of focal epileptic seizures are needed to alert patients and caregivers and to optimize the medical treatment. Heart rate variability (HRV)-based seizure detection devices have presented good detection sensitivity. However, false alarm rates (FAR) are too high.

Methods: In this phase-2 study we pursued to decrease the FAR, by using patient-adaptive logistic regression machine learning (LRML) to improve the performance of a previously published HRV-based seizure detection algorithm. ECG-data were prospectively collected using a dedicated wearable electrocardiogram-device during long-term video-EEG monitoring. Sixty-two patients had 174 seizures during 4,614 h recording. The dataset was divided into training-, cross-validation-, and test-sets (chronological) in order to avoid overfitting. Patients with >50 beats/min change in heart rate during first recorded seizure were selected as responders. We compared 18 LRML-settings to find the optimal algorithm.

Results: The patient-adaptive LRML-classifier in combination with using only responders to train the initial decision boundary was superior to both the generic approach and including non-responders to train the LRML-classifier. Using the optimal setting of the LRML in responders in the test dataset yielded a sensitivity of 78.2% and FAR of 0.62/24 h. The FAR was reduced by 31% compared to the previous method, upholding similar sensitivity.

Conclusion: The novel, patient-adaptive LRML seizure detection algorithm outperformed both the generic approach and the previously published patient-tailored method. The proposed method can be implemented in a wearable online HRV-based seizure detection system alerting patients and caregivers of seizures and improve seizure-count which may help optimizing the patient treatment.

Abstract found on PubMed

Objective: This study aimed to explore the association between epilepsy and cognitive impairment, and to determine factors associated with cognitive impairment in older people with epilepsy.

Methods: People with epilepsy and controls aged ?50 years were recruited and their global and domain-specific cognitive functions were evaluated by a comprehensive neuropsychological battery. Clinical characteristics were obtained from medical records. Analysis of covariance was used to examine the difference of cognition between two groups, after adjusting for age, gender, education years, hypertension, diabetes, and heart disease. A multiple linear regression model was used to explore the potential impact factors of cognitive functions among people with epilepsy.

Results: This study recruited 90 people with epilepsy and 110 controls. The proportion of cognitive impairment among older adults with epilepsy was 62.2%, which was significantly higher than controls (25.5%, p<0.001). People with epilepsy performed worse on global cognition (p<0.001), especially in domains of memory (p<0.001), executive function (p<0.001), language (p<0.001), and attention (p=0.031). Among older adults with epilepsy, age was negatively correlated with the scores of memory (?=-0.303, p=0.029), executive function (?=-0.354, p=0.008), and attention (?=-0.558, p<0.001). Females performed better on executive function (?=-0.350, p=0.002) than males. Education years had a positive correlation with global cognition (?=0.314, p=0.004). Number of anti-seizure medications was also negatively correlated with scores of spatial construction function (?=-0.272, p=0.019).

Significance: Our results indicated that cognitive impairment was a major comorbidity of epilepsy. Number of anti-seizure medications is suggested as a potential risk factor of impaired cognition in older people with epilepsy.