Blurred Template Arrhythmia Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing implantable cardioverter defibrillators (ICDs) face challenges in accurately classifying cardiac arrhythmias due to the limitations of direct template comparison methods, particularly with clipped and double-peaked QRS signals, which result in erroneous classifications.
Innovation Solution
The implementation of a 'blurred template' approach, where adjusted templates are computed by shifting the original template in time and amplitude, defining upper and lower boundaries to create an expected morphology area, allowing for the classification of unknown cardiac EGM waveforms by comparing them to these boundaries rather than the original template.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct template comparison methods are used for waveform classification, then the classification process is simple and fast, but the classification accuracy deteriorates for clipped and double-peaked QRS signals
Solution Approach 1:
The patent transforms the single template comparison approach into a multiple template comparison approach by creating upper and lower boundary templates from the original template. This parameter change in the template structure allows the system to accommodate signal variations in clipped and double-peaked QRS signals, thereby improving classification accuracy without requiring complex additional hardware
Solution Approach 2:
The patent creates copies of the original template (upper boundary template and lower boundary template) to form a template family. These copied templates are used to define an expected morphology area, allowing the system to handle signal variations by comparing against multiple template variants rather than a single template, thus improving accuracy for challenging waveform types
2Reliability
If a single original template is used for comparison, then the method is computationally efficient, but it cannot account for signal quality variations and leads to erroneous classifications
Solution Approach 1:
The patent segments the single template into multiple templates (upper boundary template and lower boundary template) that together define an expected morphology area. This segmentation allows the system to account for signal quality variations by providing a range of acceptable morphologies rather than a single reference, improving reliability while keeping computational requirements manageable through structured template families
3Measurement precision
If traditional template comparison is used, then the processing speed is high, but the detection precision for arrhythmias deteriorates due to inability to handle waveform variations
Solution Approach 1:
The patent performs preliminary action by pre-computing the upper and lower boundary templates and defining the expected morphology area before actual waveform classification occurs. This preparation work is done once during template creation, and then during classification, the system only needs to compare the unknown waveform against these pre-established boundaries, maintaining processing speed while improving detection precision through more robust comparison criteria
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
An implantable medical device and associated method sense cardiac signals for deriving a template representing a known EGM waveform morphology and for classifying an unknown waveform morphology. A boundary of the template, offset from the template, is computed and compared to an unknown waveform morphology for classifying the unknown waveform morphology.