Aligning Epilepsy Monitoring with Real Life
Mentis Care is creating an AI-powered platform that detects and predicts seizures in daily life, expanding neurological understanding beyond the clinic and across time. We move monitoring out of the clinical setting and into where it matters most: everyday life.
Reframing epilepsy Monitoring
From Isolated Observations to Longitudinal Insight
In-clinic epilepsy monitoring offers limited visibility into a condition that evolves over days and nights and across real-world contexts.
Epilepsy care increasingly demands objective insights that go beyond isolated recordings. Monitoring neurological activity during everyday life helps clinicians understand both individual events and how they accumulate over time.
The clinical goal remains the same. The method of generating insight needs an upgrade.
Continuous, at-home EEG monitoring enables several distinct capabilities:
Seizure Detection
Recording events in real time provides objective, time-stamped seizure records without relying solely on retrospective self-reports.
Seizure Prediction
Identifying and alerting to imminent risk empowers patients and caregivers to prepare and respond before a seizure occurs. As seizure data accumulate over time, objective records can be analyzed to reveal patterns, cycles, triggers, and treatment response.
Our Platform
A System Designed for Real-World Neurological Monitoring
Mentis Care is building an integrated platform focused on wearable EEG and AI-driven signal analysis, designed to work outside controlled clinical settings.
The platform allows extended EEG collection in daily life, combined with algorithms that analyze signals continuously in real time. Secure data infrastructure enables aggregation and review over weeks and months through interfaces for clinicians and patients.
Foundation Model: Contextualized Learning
Seizure activity is infrequent, heterogeneous, and highly variable across patients, contexts, and time. AI models trained on short EEG segments often struggle to generalize beyond narrow conditions or controlled datasets.
Mentis Care is using a foundation model approach that learns the underlying neurological structure from a large longitudinal EEG dataset before being adapted for seizure detection and prediction.
By learning shared representations across individuals and contexts, this approach aims to support more reliable signal interpretation as data scale, while remaining adaptable to various clinically relevant goals.
Impact
Supporting Clinical Decisions Beyond the Clinic
For patients:
Continuous monitoring increases visibility into their condition and provides early warning of impending seizures.
For families:
Sustained monitoring offers peace of mind through ongoing observation and timely alerts.
For clinicians:
Objective seizure records collected over time support clearer assessment of seizure burden, treatment response, and medication titration.