Vieluf, Solveig

Abstract Seizure forecasting has progressed from theoretical aspiration to a rapidly advancing research domain, yet clinical translation remains limited. Over the past decades, advances in algorithm development, chronic electroencephalography (EEG), wearable sensors, and the characterization of seizure cycles have demonstrated that seizure risk is not random but fluctuates according to identifiable biological rhythms and patient‐specific patterns. Forecasting algorithms have shown promising performance across diverse retrospective datasets, including intracranial EEG, subscalp recordings, wearable physiological signals, and even self‐reported diaries. However, the clinical value of forecasting is contentious. Prospective real‐world validation and regulatory approval of patient‐facing forecasting systems remain rare. This review incorporates perspectives presented at the 5th International Congress on Mobile Health and Digital Technology in Epilepsy (2025). We examine barriers impeding clinical translation and current attempts to address them. Crucially, forecasting performance cannot be evaluated in isolation from intended use. Applications range from low‐risk uses, like scheduling diagnostic monitoring or visualization of historical trends, to higher risk interventions, including medication titration and adaptive neuromodulation. Each application entails distinct performance thresholds, ethical considerations, and regulatory requirements. Translational challenges include reliable seizure annotation, nonstationarity dynamics of biological cycles, and practical constraints for real‐time deployment. Ethical concerns center on miscalibrated reliance on low‐risk states, potential anxiety associated with high‐risk advisories, and the heterogeneity of patient preferences and risk tolerance. Regulatory pathways are likely to depend on clearly defined use cases and clinically meaningful endpoints, which may extend beyond seizure counts to include quality of life, anxiety, locus of control, and other patient‐reported outcomes. Ultimately, translation will require rigorous prospective evaluation against transparent benchmarks, sustainable scientific–commercial partnerships, and integration of probabilistic risk information into clinical workflows. With careful implementation, seizure forecasting may evolve from proof‐of‐concept research into a clinically meaningful component of epilepsy management, and we remain cautiously optimistic.

Abstract Objective Wearable nonelectroencephalographic biosignal recordings captured from the wrist offer enormous potential for seizure monitoring. However, signal quality remains a challenging factor affecting data reliability. Models trained for seizure detection depend on the quality of recordings in peri‐ictal periods in performing a feature‐based separation of ictal periods from interictal periods. Thus, this study aims to investigate the effect of epileptic seizures on signal quality, ensuring accurate and reliable monitoring. Methods This study assesses the signal quality of wearable data during peri‐ictal phases of generalized tonic–clonic and focal to bilateral tonic–clonic seizures (TCS), focal motor seizures (FMS), and focal nonmotor seizures (FNMS). We evaluated accelerometer (ACC) activity and the signal quality of electrodermal activity (EDA) and blood volume pulse (BVP) data. Additionally, we analyzed the influence of peri‐ictal movements as assessed by ACC (ACC activity) on signal quality and examined intraictal subphases of focal to bilateral TCS. Results We analyzed 386 seizures from 111 individuals in three international epilepsy monitoring units. BVP signal quality and ACC activity levels differed between all seizure types. We found the largest decrease in BVP signal quality and increase in ACC activity when comparing the ictal phase to the pre‐ and postictal phases for TCS. Additionally, ACC activity was strongly negatively correlated with BVP signal quality for TCS and FMS, and weakly for FNMS. Intraictal analysis revealed that tonic and clonic subphases have the lowest BVP signal quality and the highest ACC activity. Significance Motor elements of seizures significantly impair BVP signal quality, but do not have significant effect on EDA signal quality, as assessed by wrist‐worn wearables. The results underscore the importance of signal quality assessment methods and careful selection of robust modalities to ensure reliable seizure detection. Future research is needed to explain whether seizure detection models' decisions are based on signal responses induced by physiological processes as opposed to artifacts.

Purpose: Hypsarrhythmia is one of the major diagnostic and treatment response criteria in infantile spasms (IS). The clinical and electrophysiological effect of repository corticotropin injection treatment on IS was evaluated using electrophysiological biomarkers. Methods: Consecutive infants (<24 months) treated with repository corticotropin injection for IS were included in this retrospective descriptive study. Inclusion criteria were (1) clinical IS diagnosis, (2) repository corticotropin injection treatment, and (3) consecutive EEG recordings before and after repository corticotropin injection treatment. Patients with tuberous sclerosis complex were excluded. Response to treatment was defined as freedom from IS for at least 7 consecutive days during the treatment and resolution of hypsarrhythmia. The authors defined “relapse” as the recurrence of seizures after an initial response. Electrophysiological biomarker assessment included evaluation of semiautomatic spike counting algorithm, delta power, and delta coherence calculation during non-REM sleep EEG. Results: One hundred fifty patients (83 males; 55%; median age of IS onset: 5.9 months) with complete data were included, including 101 responders (67%, 71 with sustained response, and 30 relapses). Fifty patients (33%) with complete EEG data also underwent advanced EEG analysis. Baseline delta coherence was higher in sustained responders than in nonresponders or patients who relapsed. Greater decreases in semiautomatic spike counting algorithm, delta power, and delta coherence were found in sustained responders compared with nonresponders or patients who relapsed. Conclusions: Repository corticotropin injection treatment was associated with a 67% response rate in patients with IS. Computational biomarkers beyond hypsarrhythmia may provide additional information during IS treatment, such as early determination of treatment response and outcome assessment.