Zygomatic EEG Detection of Medial Temporal Lobe Epileptic Events
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Detecting pathological electrophysiological events such as Interictal Epileptiform Discharges (IEDs) in the medial temporal lobe (MTL) regions is challenging with scalp EEG, as these events often occur spontaneously and are difficult to capture non-invasively.
Innovation Solution
A machine learning (ML) based platform using zygomatic EEG data to detect MTL epileptic activity by training ML models with intracranial depth electrode data, followed by non-invasive facial EEG channels, enabling reliable detection of IEDs and other events with high precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If scalp EEG is used to detect EP events, then non-invasive detection is achieved, but detection reliability deteriorates for MTL regions
Solution Approach 1:
The patent uses zygomatic bone EEG electrodes as an intermediary structure that bridges the gap between non-invasive scalp EEG and invasive MTL detection. The zygomatic bone serves as a physical mediator that transmits electrical signals from deep MTL structures to surface electrodes, enabling non-invasive detection with improved reliability compared to conventional scalp EEG.
Solution Approach 2:
The patent applies local quality by specifically targeting the zygomatic bone region for electrode placement, rather than using general scalp EEG positions. This localized approach concentrates the detection capability on the specific anatomical pathway that provides optimal access to MTL structures, improving detection reliability for this critical region while maintaining non-invasive operation.
2Device complexity
If conventional scalp EEG is used, then simple electrode placement is maintained, but detection precision deteriorates for deep brain structures
Solution Approach 1:
The invention maintains relative simplicity by using surface electrodes but achieves improved precision through local quality enhancement - specifically placing electrodes on the zygomatic bone, which has unique anatomical properties that provide better signal transmission from MTL structures compared to conventional scalp positions.
Solution Approach 2:
The patent creates a functional copy of the invasive depth electrode detection capability using non-invasive zygomatic EEG. By training ML models on paired datasets (intracranial and zygomatic EEG), the system replicates the detection precision of invasive methods through a non-invasive proxy measurement approach.
3Measurement precision
If ML models are trained on intracranial data, then detection precision improves, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-training ML models using intracranial EEG data before deploying them for non-invasive detection. This offline training phase prepares the models in advance with ground-truth labeled data, so that during actual operation, the system can achieve high detection precision without requiring real-time complex processing or additional invasive procedures.
Solution Approach 2:
The system creates a computational copy of the intracranial detection capability by training ML models to learn the mapping between zygomatic EEG signals and MTL EP events. This computational model serves as a virtual replica of the invasive detection system, achieving similar precision without the physical complexity of intracranial electrode implantation.
Data Source
AI summary
A system and method of detecting Electrophysiological events such as Interictal Epileptiform Discharge (IED) events, or other pathological and physiological electrophysiological events, in a human subject by at least one processor may include, for example: placing at least one first electroencephalogram (EEG) electrode over a zygomatic bone or a maxilla of the subject, directly below the subject's orbit in the subject's inferior direction; receiving a first EEG signal from the at least one first EEG electrode; processing the first EEG signal, to obtain one or more first EEG data elements; and inferring at least one machine-learning (ML) based model on the one or more first EEG data elements, to predict occurrence of at least one IED event in the subject.