EMG Seizure Detection Using Wavelet Frequency Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current seizure detection methods, particularly EEG-based systems, are cumbersome, require technical expertise, and are not suited for long-term home use or daily wearability, often failing to differentiate between epileptic seizures and psychogenic non-epileptic seizures (PNES) accurately, leading to delayed or incorrect diagnoses.
Innovation Solution
A method involving EMG signal analysis using wavelet transforms to differentiate between PNES and generalized tonic-clonic (GTC) seizures by organizing data into high and low frequency groups, determining magnitude, scaling, and comparing to thresholds to detect tonic and clonic phases, enabling classification of seizure events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If EEG-based seizure detection systems are used, then seizure activity can be detected, but the systems become cumbersome and require technical expertise to apply and monitor
Solution Approach 1:
The patent replaces the complex EEG mechanical system (multiple electrodes, wires, amplifiers) with an EMG-based system that uses simpler electrodes placed on muscles. This substitution maintains seizure detection capability while reducing device complexity and making the system more suitable for home use.
Solution Approach 2:
The patent introduces EMG signals as an intermediary to detect seizure activity. Instead of directly measuring brain electrical activity through complex EEG systems, the system uses muscle electrical activity as a mediator to infer seizure events, thereby simplifying the detection mechanism.
2Measurement precision
If multiple EEG electrodes and video recording equipment are used, then seizure confirmation can be achieved, but the equipment becomes cumbersome and requires staffed clinical environments
Solution Approach 1:
The EMG-based system is designed to be self-sufficient and does not require staffed clinical environments for operation. The system automatically detects and records seizure events using muscle electrical activity, eliminating the need for continuous caregiver observation or specialized clinical staff, thereby improving ease of operation for home use.
3Productivity
If accelerometer-based seizure alerting systems are used, then motion detection can be achieved, but the systems fail to detect seizures where muscles work to make the person rigid rather than cause violent movement
Solution Approach 1:
The patent replaces accelerometer-based mechanical motion detection with EMG-based electrical signal detection. Since EMG measures the electrical activity of muscles directly, it can detect the subtle electrical signals of muscle rigidity even when no gross body movement occurs, thereby maintaining high detection accuracy for all seizure types.
Data Source
AI summary
Systems and methods are described for detecting and characterizing seizures or seizure-related events. The methods herein may include determining magnitude and/or scaled magnitude data for each of at least one high and low frequency group of signals. Based on the determined magnitudes and/or scaled magnitude data, seizures or seizure-related events may be characterized.