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

Artificial intelligence applied to electroencephalography in epilepsy

Artificial intelligence (AI) is progressively transforming all fields of medicine, promising substantial changes in clinical practice. In the context of epilepsy, electroencephalography (EEG), a technique used for over a century, has historically been resistant to automated analysis due to the complexity of the signals and the challenges posed by artifact management. While the human eye excels at recognizing patterns, algorithms have demonstrated superior capabilities in detecting and characterizing specific features, such as long-term dynamics and synchrony. Furthermore, the advent of wearable EEG devices has led to an exponential increase in data volume, surpassing the limits of visual interpretation. AI algorithms are now being developed to address these limitations, offering enhanced efficiency in both identifying subtle signal features and managing massive datasets. This review explores the fundamental principles of AI and its transformative potential in the field of EEG. It discusses the implications and the current limitations, including improvements limited to aggregation of already known knowledge, for epilepsy diagnosis, medical and surgical treatment, and innovative approaches to patient monitoring, including seizure forecasting, highlighting how AI is poised to redefine the management of epilepsy.

How accurate are machine learning models in predicting anti-seizure medication responses: A systematic review

IMPORTANCE

Current epilepsy management protocols often depend on anti-seizure medication (ASM) trials and assessment of clinical response. This may delay the initiation of the ASM regimen that might optimally balance efficacy and tolerability for individual patients. Machine learning (ML) can offer a promising tool for efficiently predicting ASM response.

OBJECTIVE

The objective of this review is to synthesize the available information about the effectiveness and limitations of ML models in predicting and classifying the response of patients with epilepsy to ASMs, and to assess the impact of various data inputs on prediction performance.

EVIDENCE REVIEW

We conducted a comprehensive search of studies utilizing ML models for ASM response prediction using PubMed and Scopus up until November 2024.

FINDINGS

The review included 37 studies. Various data types, including clinical information, brain MRI, EEG, and genetic data, are useful in predicting responses to ASMs. Tree-based ML algorithms and Support Vector Machines are the most used models. Reported results vary widely, with certain models achieving near-perfect accuracy and others performing similar to random classifiers. The review also highlights the limitations of this research field, especially concerning the quality and quantity of data.

CONCLUSIONS AND RELEVANCE

The findings indicate that while ML models show great promise in predicting ASM responses in epilepsy, further research is required to refine these models for practical clinical application. The review underscores both the potential of ML in advancing precision medicine in epilepsy management and the need for continued research to improve prediction accuracy.

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

A diagnosis of epilepsy has significant consequences for an individual but is often challenging in clinical practice. Novel biomarkers are thus greatly needed. Here, we investigated how common genetic factors (epilepsy polygenic risk scores, [PRSs]) influence epilepsy risk in detailed longitudinal electronic health records (EHRs) of > 700k Finns and Estonians. We found that a high genetic generalized epilepsy PRS (PRSGGE) increased risk for genetic generalized epilepsy (GGE) (hazard ratio [HR] 1.73 per PRSGGE standard deviation [SD]) across lifetime and within 10 years after an unspecified seizure event. The effect of PRSGGE was significantly larger on idiopathic generalized epilepsies, in females and for earlier epilepsy onset. Analogously, we found significant but more modest focal epilepsy PRS burden associated with non-acquired focal epilepsy (NAFE). Here, we outline the potential of epilepsy specific PRSs to serve as biomarkers after a first seizure event.

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

Epilepsy surgery is effective for patients with medication-resistant seizures, however 20-40% of them are not seizure free after surgery. Aim of this study is to evaluate the role of linear and non-linear EEG features to predict post-surgical outcome. We included 123 paediatric patients who underwent epilepsy surgery at Bambino Gesù Children Hospital (January 2009-April 2020). All patients had long term video-EEG monitoring. We analysed 1-min scalp interictal EEG (wakefulness and sleep) and extracted 13 linear and non-linear EEG features (power spectral density (PSD), Hjorth, approximate entropy, permutation entropy, Lyapunov and Hurst value). We used a logistic regression (LR) as feature selection process. To quantify the correlation between EEG features and surgical outcome we used an artificial neural network (ANN) model with 18 architectures. LR revealed a significant correlation between PSD of alpha band (sleep), Mobility index (sleep) and the Hurst value (sleep and awake) with outcome. The fifty-four ANN models gave a range of accuracy (46-65%) in predicting outcome. Within the fifty-four ANN models, we found a higher accuracy (64.8% ± 7.6%) in seizure outcome prediction, using features selected by LR. The combination of PSD of alpha band, mobility and the Hurst value positively correlate with good surgical outcome.

Prediction of the Need for Anticonvulsants in the Management of Orofacial Neuropathic Pain Using Machine Learning

Background and aim Orofacial neuropathic pain is a medical condition caused by a lesion or dysfunction of the nervous system and is one of the most challenging for dental clinicians to diagnose. Anticonvulsants, antidepressants, analgesics, nonsteroidal anti-inflammatory drugs, and other classes of medications are frequently used to treat this condition. Our study aimed to build a machine learning-based classifier to predict the need for anticonvulsant drugs in patients with orofacial neuropathic pain. Materials and methods A machine learning tool that was trained and tested on patients for predicting and detecting algorithms, which would in turn predict the need for anticonvulsants in the treatment of orofacial neuropathic pain, was employed in this study. Results Three machine learning algorithms successfully detected and predicted the need for anticonvulsants to treat patients with orofacial neuropathic pain. All three models showed a high accuracy, that is, 97%, 94%, and 89%, in predicting the need for anticonvulsants. Conclusion Machine learning algorithms can accurately predict the need for anticonvulsant drugs for treating orofacial neuropathic pain. Further research is needed to validate these findings using larger sample sizes and imaging modalities.

Machine Learning and Artificial Intelligence Applications to Epilepsy: a Review for the Practicing Epileptologist

PURPOSE OF REVIEW

Machine Learning (ML) and Artificial Intelligence (AI) are data-driven techniques to translate raw data into applicable and interpretable insights that can assist in clinical decision making. Some of these tools have extremely promising initial results, earning both great excitement and creating hype. This non-technical article reviews recent developments in ML/AI in epilepsy to assist the current practicing epileptologist in understanding both the benefits and limitations of integrating ML/AI tools into their clinical practice.

RECENT FINDINGS

ML/AI tools have been developed to assist clinicians in almost every clinical decision including (1) predicting future epilepsy in people at risk, (2) detecting and monitoring for seizures, (3) differentiating epilepsy from mimics, (4) using data to improve neuroanatomic localization and lateralization, and (5) tracking and predicting response to medical and surgical treatments. We also discuss practical, ethical, and equity considerations in the development and application of ML/AI tools including chatbots based on Large Language Models (e.g., ChatGPT). ML/AI tools will change how clinical medicine is practiced, but, with rare exceptions, the transferability to other centers, effectiveness, and safety of these approaches have not yet been established rigorously. In the future, ML/AI will not replace epileptologists, but epileptologists with ML/AI will replace epileptologists without ML/AI.

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

A diagnosis of epilepsy has significant consequences for an individual but is often challenging in clinical practice. Novel biomarkers are thus greatly needed. Here, we investigated how common genetic factors (epilepsy polygenic risk scores, [PRSs]) influence epilepsy risk in detailed longitudinal electronic health records (EHRs) of > 360k Finns spanning up to 50 years of individuals' lifetimes. Individuals with a high genetic generalized epilepsy PRS (PRSGGE) in FinnGen had an increased risk for genetic generalized epilepsy (GGE) (hazard ratio [HR] 1.55 per PRSGGE standard deviation [SD]) across their lifetime and after unspecified seizure events. Effect sizes of epilepsy PRSs were comparable to effect sizes in clinically curated data supporting our EHR-derived epilepsy diagnoses. Within 10 years after an unspecified seizure, the GGE rate was 37% when PRSGGE > 2 SD compared to 5.6% when PRSGGE < -2 SD. The effect of PRSGGE was even larger on GGE subtypes of idiopathic generalized epilepsy (IGE) (HR 2.1 per SD PRSGGE). We further report significantly larger effects of PRSGGE on epilepsy in females and in younger age groups. Analogously, we found significant but more modest focal epilepsy PRS burden associated with non-acquired focal epilepsy (NAFE). We found PRSGGE specifically associated with GGE in comparison with >2000 independent diseases while PRSNAFE was also associated with other diseases than NAFE such as back pain. Here, we show that epilepsy specific PRSs have good discriminative ability after a first seizure event i.e. in circumstances where the prior probability of epilepsy is high outlining a potential to serve as biomarkers for an epilepsy diagnosis.