MEG Waveform Identification for Epilepsy Lesion Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for analyzing characteristic waveforms in magneto-encephalography, such as the equivalent current dipole method, face challenges in accurately determining the time and sensors for Interictal Epileptiform Discharge (IED) due to the large volume of data and low signal-to-noise ratio, making manual extraction difficult and inefficient.
Innovation Solution
A waveform generation identification method that compares individual waveform data from multiple sensors to determine the probability of characteristic waveform information appearance based on correlation, using machine learning to create an IED probability map that identifies the time and sensors of IED occurrence, enabling more accurate dipole estimation and localization of epilepsy lesions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual search methods are used to extract IED information, then measurement precision can be maintained, but loss of time increases significantly due to the enormous volume of data
Solution Approach 1:
The patent replaces manual mechanical search processes with automated computer-based algorithms. The system automatically processes MEG data by comparing waveform patterns against stored templates, calculating correlation coefficients, and identifying IED events without human intervention, thereby eliminating time loss while maintaining detection accuracy through computational analysis
Solution Approach 2:
The patent creates a database of template waveforms representing characteristic IED patterns. The system then searches through enormous amounts of MEG data by comparing and matching these template copies against actual waveform segments, enabling rapid identification of IED events without manually examining each data point
2Measurement precision
If wavelet transform methods are applied to detect characteristic waves, then measurement precision improves, but the technology cannot determine IED time and narrow down sensors appropriately for equivalent current dipole method
Solution Approach 1:
The patent introduces template waveforms as intermediary reference patterns that bridge the gap between raw MEG data and IED identification. By comparing actual waveforms against these standardized templates and calculating correlation coefficients, the system can precisely determine both the temporal occurrence and spatial sensor locations of IED events, enabling subsequent dipole estimation
Solution Approach 2:
The patent segments the MEG data into individual waveform segments from multiple sensors and time points. Each segment is independently compared against template waveforms, allowing the system to identify specific IED events and their associated sensors without requiring complex global analysis, thus enabling precise time and location determination
3Measurement precision
If sampling frequency and number of sensors are increased, then measurement precision improves, but device complexity and data volume increase enormously
Solution Approach 1:
The patent extracts only the relevant information needed for IED detection by comparing waveform patterns against templates. Rather than processing every data point from the high-resolution sensor array, the system identifies and extracts only those waveform segments that match characteristic IED patterns, significantly reducing processing complexity while maintaining the benefits of high sampling frequency and sensor density
Data Source
AI summary
A waveform generation identification method includes: comparing individual waveform data obtained by a plurality of sensors, with at least one piece of characteristic waveform information; determining appearance probability of characteristic waveform information in at least a certain section of the waveform data, based on a degree of correlation between a peak section of the waveform data and the characteristic waveform information; and identifying a time when a section matching with the characteristic waveform information appears and a concerned sensor, based on the appearance probability.