Non-Invasive Seizure Forecasting Wearable System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current solutions for drug-resistant epilepsy (DRE) are either invasive, costly, and unreliable, or they only detect seizures during the ictal period, failing to predict upcoming seizures and address social and safety issues.
Innovation Solution
A non-invasive end-to-end seizure prediction system using physiological data (heart rate, blood volume pulse, electrodermal activity, etc.) and physical data (voice/audio) with a deep learning-enabled ecosystem that personalizes and optimizes predictions for individual DRE patients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If subcutaneous EEG is used for seizure prediction, then prediction accuracy is improved, but invasiveness and cost increase
Solution Approach 1:
The patent replaces the mechanical/invasive EEG electrode implantation system with a non-invasive wearable device system that uses physiological sensors (heart rate, blood volume pulse, electrodermal activity, temperature, accelerometry) to detect seizure precursors, thereby maintaining prediction capability while eliminating surgical invasion
Solution Approach 2:
The patent introduces physiological parameters (heart rate, BVP, EDA, temperature) as intermediary indicators that reflect brain activity changes during pre-ictal periods, allowing indirect measurement of seizure risk without direct brain contact
2Reliability
If detection systems are used for seizure monitoring, then seizure detection capability is improved, but ability to predict upcoming seizures deteriorates
Solution Approach 1:
The patent implements preliminary detection of pre-ictal physiological changes (heart rate variability, electrodermal activity, blood volume pulse patterns) that occur before seizures, enabling early warning and prediction rather than only detecting seizures after they begin
Solution Approach 2:
The system continuously monitors physiological parameters and provides real-time feedback about seizure risk levels, allowing dynamic adjustment of alerts based on detected patterns and improving both detection reliability and prediction capability
3Adaptability or versatility
If deep learning models are trained on multiple patients' data, then model generalization is improved, but patient data privacy deteriorates
Solution Approach 1:
The patent creates synthetic copies of patient data through data augmentation techniques and privacy-preserving methods that allow training on multiple patients' patterns without exposing actual sensitive health information, maintaining generalization while protecting privacy
Solution Approach 2:
The system allows each patient's data to be processed and modeled locally with personalized parameters, enabling generalization across patients through shared architecture while maintaining local data privacy and individualized treatment patterns
Data Source
AI summary
A system for seizure forecasting is provided. The system comprises a wearable device, and a mobile device. The wearable device includes one or more sensors that are configured to obtain physiological data of a user that describes pre-ictal, ictal and post-ictal and normal phases of the user's seizure related activity. The mobile device includes at least a processor, one or more machine learning models and a memory. The memory is encoded with instructions that, when executed by the at least processor of the system, cause the system to perform operations comprises periodically receiving the physiological data of the user from the wearable device; and analyzing the received physiological data using the machine learning models and forecast upcoming seizures.