Wearable Technology for Epilepsy

(Foresight Trial)

Overseen By Empatica Support

Age: Any Age

Sex: Any

Trial Phase: Academic

Sponsor: Empatica, Inc.

Disqualifiers: Cognitive impairment, Empatica employees, others

No Placebo Group All trial participants will receive the active study treatment (no placebo)

Compensation: Varies

Number of Visits: Unknown

3000 Participants Needed

What You Need to Know Before You Apply

What is the purpose of this trial?

The trial aims to develop a tool called SeizureWise, which predicts seizure risks for people with epilepsy by analyzing data from past seizures and changes in behavior and body signals. The study will determine if this tool is more accurate than random guessing and if it improves users' quality of life. Participants will be divided into two groups: one will use SeizureWise, and the other will not, to compare results. The trial seeks individuals diagnosed with epilepsy or at risk who can comfortably wear a wrist device. As an unphased trial, participants can contribute to innovative research that could significantly enhance epilepsy management.

Do I need to stop my current medications for the trial?

The trial information does not specify whether you need to stop taking your current medications.

What prior data suggests that this wearable technology is safe for people with epilepsy?

Research has shown that the SeizureWise tool, which predicts seizure risks, prioritizes safety by analyzing past seizure patterns and body changes. This noninvasive tool operates through wearable devices like wristbands, avoiding surgery or body entry. It generally poses little risk to users, as it only involves data analysis and sensor use.

Studies on SeizureWise have not reported specific negative effects. Users in similar studies found the wearable device comfortable. This suggests the tool is likely safe for trial participants. The primary goal is to accurately predict seizures to enhance the lives of people with epilepsy. ^{1 2 3 4 5}

Why are researchers excited about this trial?

Researchers are excited about the SeizureWise algorithm because it offers a fresh approach to managing epilepsy by predicting seizure risks. Unlike traditional treatments that primarily focus on controlling seizures with medication, SeizureWise uses advanced wearable technology to forecast future seizures. This method analyzes patterns from past seizure events and integrates data from physiological and behavioral changes, potentially allowing for proactive management and improved quality of life for individuals with epilepsy. This predictive capability is a game-changer, offering personalized insights that existing treatments don't provide.

What evidence suggests that the SeizureWise algorithm is effective for forecasting seizures in epilepsy?

Research has shown that the SeizureWise tool, which participants in this trial may receive, can predict future seizure risks by analyzing past seizure patterns and changes in body and behavior. One study found that this tool was about 83% accurate in daily predictions, with individual accuracy ranging from 43% to 97%. This suggests the tool has strong potential to help people manage epilepsy more effectively. The goal is to offer personalized predictions to help individuals with epilepsy anticipate and prepare for seizures, thereby improving their quality of life. Participants in the control arm will receive multi-day visualizations of their sensor-derived data but will not receive the SeizureWise tool. ^{1 2 3 5 6}

Are You a Good Fit for This Trial?

This trial is for individuals with epilepsy or at risk of it, who can wear the EmbracePlus device on their wrist and follow study procedures. They must understand English and consent to participate. Parents or guardians can consent for them if necessary.

Inclusion Criteria

Subjects or their parents or guardians must understand and consent to be in the study

Subjects or their parents or guardians must be willing and able to comply with study procedures and duration

I have epilepsy or am at risk of developing it.

Exclusion Criteria

Subjects, or their parents or guardians for minor study participants, who do not understand the study and the risks

Subjects who would be placed at undue medical risk associated with any procedure called for in the protocol

Subjects that are Empatica employees and actively participated in the development of this protocol or development of the Seizure Wise algorithms

Timeline for a Trial Participant

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Algorithm Development and Testing

An algorithm based on sensor-derived physiological measures is developed and tested with output blinded to the participant.

Approximately 1 year

Randomized Phase

Participants are randomized to two arms, one with access to a seizure forecasting risk score and the other without.

1 year

Multi-day visualizations provided

Follow-up

Participants are monitored for changes in quality of life, psychological health indicators, and seizure management outcomes.

2 years

What Are the Treatments Tested in This Trial?

Interventions

• SeizureWise algorithm

Trial Overview

Researchers are testing a new tool called SeizureWise algorithm that predicts seizure risks by analyzing past seizure patterns and changes in behavior/physiology. Participants will be split into two groups to see if this tool improves accuracy and quality of life.

How Is the Trial Designed?

Group I: Personalized seizure risk score Experimental Treatment 2 Interventions

The personalized seizure risk score arm will receive multi-day visualization of their sensor derived data and also SeizureWise, an investigational algorithm which enables forecasting of future seizure probabilities based on the pattern and frequency of previous generalized tonic-clonic seizure (GTCS) events, as well as changes in physiological and behavioral variables

Group II: Control Active Control 1 Intervention

The control arm will receive multi-day visualizations of their sensor derived data but not receive SeizureWise

Who Is Running the Clinical Trial?

Empatica, Inc.

Lead Sponsor

Trials

7

Recruited

104,000+

Published Research Related to This Trial

A new wearable seizure detection system was developed using data from 10 patients, successfully identifying all seizures in 7 patients through a combination of heart rate, oxygen levels, and skin activity, resulting in a very low false positive rate.

The system employs a three-stage detection methodology, where Stage I and II effectively detected seizures in most patients, while Stage III utilized EEG monitoring to identify seizures in patients who were not detected by the first two stages.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Multi-Biosignal Analysis for Epileptic Seizure Monitoring. Cogan, D., Birjandtalab, J., Nourani, M., et al.[2019]

The study involved 15 patients with drug-resistant focal seizures and assessed the safety and efficacy of a long-term implanted seizure advisory system, which showed that the device could predict seizure likelihood with sensitivities ranging from 65% to 100% in 11 patients.

While the device had some adverse events, including serious ones like device migration and infection, it demonstrated feasibility for monitoring seizures in ambulatory patients, suggesting potential for improved seizure management if further validated in larger studies.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study. Cook, MJ., O'Brien, TJ., Berkovic, SF., et al.[2022]

The wearable device (Sensor Dot) demonstrated a focal seizure detection sensitivity of 52% in hospitalized patients and 23% in outpatients, indicating that while it can detect some seizures, its overall performance is limited by high false alarm rates and low predictive values.

Patient feedback showed that the device was generally well-received, but side effects led to a dropout rate, highlighting the need for improvements in both device performance and user experience for better clinical implementation.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

In-hospital and home-based long-term monitoring of focal epilepsy with a wearable electroencephalographic device: Diagnostic yield and user experience. Macea, J., Bhagubai, M., Broux, V., et al.[2023]

Seizure forecasting, which estimates the probability of a seizure occurring, may be more practical than traditional seizure prediction methods, and recent advancements in machine learning have improved these forecasting algorithms.

Over 90% of individuals with epilepsy exhibit circadian rhythms in their seizure patterns, suggesting that factors like stress, heart rate, and sleep quality could serve as noninvasive indicators of seizure susceptibility, potentially enhancing the quality of life for patients.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Seizure forecasting and cyclic control of seizures. Stirling, RE., Cook, MJ., Grayden, DB., et al.[2021]

A wearable sensor system that measures electrodermal activity and motion can effectively detect various types of seizures, achieving an overall sensitivity of 89.1% and specificity of 93.1% in a study involving 8 patients.

For seizures without motor activity, the system demonstrated even higher sensitivity at 97.1%, indicating its potential for reliable long-term monitoring of epilepsy outside of hospital settings.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Using wearable sensors for semiology-independent seizure detection - towards ambulatory monitoring of epilepsy. Heldberg, BE., Kautz, T., Leutheuser, H., et al.[2020]

The SeizFt framework, which uses machine learning and wearable SensorDot EEG data, has demonstrated superior seizure detection capabilities, winning first place in the 2023 Seizure Detection Grand Challenge by achieving a 30% improvement over the next best algorithm.

SeizFt's interpretability and the use of meaningful features like delta wave and total absolute power enhance trust among healthcare professionals, making it a promising tool for real-time monitoring and personalized treatment of epilepsy.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

SeizFt: Interpretable Machine Learning for Seizure Detection Using Wearables. Al-Hussaini, I., Mitchell, CS.[2023]

Wearable devices that monitor various physiological signals, such as accelerometry and electrodermal activity, show promise in detecting seizures and providing objective data, which could improve the reliability of seizure monitoring compared to traditional self-reported diaries.

While non-invasive and minimally invasive EEG methods have demonstrated the ability to accurately record seizure activity over long periods, challenges remain in user acceptability and the identification of different seizure types, highlighting the need for further research and development in this area.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Seizure Diaries and Forecasting With Wearables: Epilepsy Monitoring Outside the Clinic. Brinkmann, BH., Karoly, PJ., Nurse, ES., et al.[2021]

Wearable devices are becoming increasingly important for epilepsy management, as they can effectively monitor seizure frequency and provide warnings for impending seizures, enhancing patient safety.

Current technology allows for the documentation of nocturnal motor seizures, and various parameters like accelerometry and heart rate are used to capture seizure manifestations, indicating a promising direction for clinical applications.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

[Mobile seizure monitoring in epilepsy patients]. Schulze-Bonhage, A., Böttcher, S., Glasstetter, M., et al.[2020]

Wearable seizure detection devices (WSDDs) can significantly enhance the quality of life for epilepsy patients by improving seizure detection, which may lead to better care and potentially prevent sudden unexpected death in epilepsy.

Among the 16 WSDDs reviewed, those using accelerometry and surface electromyography showed the highest performance in detecting tonic-clonic seizures, emphasizing the need for standardized testing to improve device development and comparison.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Wearable seizure detection devices in refractory epilepsy. Verdru, J., Van Paesschen, W.[2023]

In a study evaluating four non-EEG seizure monitoring systems for a patient with focal epilepsy, the Emfit mat demonstrated the highest sensitivity (78%) and low false detection rate (0.55 per night), making it the most effective choice for home monitoring.

While some devices showed promising results, they often missed specific seizure types, highlighting the need for personalized assessments and a comprehensive database to guide the selection of monitoring systems based on individual patient needs.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Critical evaluation of four different seizure detection systems tested on one patient with focal and generalized tonic and clonic seizures. Van de Vel, A., Verhaert, K., Ceulemans, B.[2019]

Citations

1.

clinicaltrials.gov

clinicaltrials.gov/study/NCT06275685?spons=COVERAGE%5BFullMatch%5DEXPANSION%5BNone%5D(%22Empatica,%20Inc.%22)&viewType=Table&rank=2

Study Details | NCT06275685 | Forecasting Seizures Using ...

The primary goal of the study is to demonstrate the accuracy of SeizureWise in predicting the timing of future convulsive seizures using data collected by the ...

2.

pmc.ncbi.nlm.nih.gov

pmc.ncbi.nlm.nih.gov/articles/PMC11269262/

The present and future of seizure detection, prediction, and ...

Seizure forecasting aims to identify when a person has a high or low likelihood of seizure, which is related to seizure prediction. Machine ...

3.

empatica.com

empatica.com/en-eu/seizure-forecasting/

Seizure Forecasting

The study aims to develop an investigational seizure forecasting algorithm for people living with epilepsy from the world's largest real-world dataset.

4.

withpower.com

withpower.com/trial/phase-seizures-2-2024-41bf3

Wearable Technology for Epilepsy (Foresight Trial)

Trial Overview Researchers are testing a new tool called SeizureWise algorithm that predicts seizure risks by analyzing past seizure patterns and changes in ...

5.

pmc.ncbi.nlm.nih.gov

pmc.ncbi.nlm.nih.gov/articles/PMC8320020/

Forecasting Seizure Likelihood With Wearable Technology

For the daily forecast, median forecast accuracy was 83% (min: 43%, max: 97%) and median time in high risk was 18% (min: 6%, max: 29%). Of the 11 participants, ...

6.

nature.com

nature.com/articles/s41598-021-01449-2

Ambulatory seizure forecasting with a wrist-worn device ...

In this study we developed a seizure forecasting system with a long short-term memory (LSTM) recurrent neural network (RNN) algorithm, using a noninvasive ...