ECG Signal Pattern Detection Using Digital Signatures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual detection of patterns in electrocardiogram (ECG) signals is tedious, time-consuming, and prone to errors, while specifying patterns quantitatively for automatic detection is difficult.
Innovation Solution
A computer-implemented method that segments ECG signals into time segments, represents changes as characters, forms a signature, and automatically identifies pattern occurrences by matching these signatures in other segments, using scaling parameters to align dominant frequencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual detection of patterns in ECG signals is performed, then diagnostic accuracy can be maintained, but time consumption and labor effort increase significantly
Solution Approach 1:
The patent creates a digital signature (copy) of the ECG pattern that captures its essential characteristics. This signature is then used to automatically identify and detect occurrences of the pattern in ECG signals, eliminating the need for manual visual inspection while maintaining detection accuracy. The signature serves as a reusable template that can be quickly compared against new signal data.
Solution Approach 2:
The patent replaces the mechanical/manual process of visual pattern recognition with an automated computational system. The system uses algorithmic comparison of signal characteristics against stored signatures to detect patterns, substituting human expertise with machine-based automated detection that operates faster and without fatigue.
2Productivity
If automatic detection of patterns is implemented, then time efficiency improves, but the complexity of specifying patterns quantitatively increases
Solution Approach 1:
Instead of requiring users to define complex quantitative parameters, the system captures the pattern characteristics by copying/extraction of key features from example signals. The signature generation process automatically identifies and stores the essential characteristics, simplifying the user task to providing example patterns rather than specifying complex detection criteria.
Solution Approach 2:
The system performs self-configuration by automatically generating detection signatures from example patterns provided by the user. The algorithm autonomously identifies relevant features and creates the detection template, eliminating the need for users to manually configure complex detection parameters or understand the underlying technical specifications.
3Extent of automation
If detailed quantitative pattern specification is required, then automation capability is achieved, but ease of operation decreases
Solution Approach 1:
The system uses example-based copying where users provide sample patterns and the system automatically creates detection signatures from these examples. This approach maintains full automation capability while dramatically simplifying operation, as users only need to provide representative examples rather than specify complex quantitative criteria.
Solution Approach 2:
The patent introduces an intermediary signature representation that bridges between simple user input (example patterns) and complex automated detection requirements. The signature acts as a mediator that translates intuitive example-based input into precise automated detection rules, eliminating the need for users to directly handle complex quantitative specifications.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Automatically detects patterns of interest in ECG signals, reducing manual effort and improving diagnostic efficiency by marking and displaying occurrences with statistical information.
Implementation Method 1
segmenting each first time interval of the multiple first time intervals into a plurality of time segments; determining whether the physiological signal increases or decreases in each time segment of the plurality of time segments within the first time intervals, wherein determining whether the physiological signal increases or decreases comprises combining physiological signal characteristics in parallel time segments
Implementation Method 2
calculating a scaling parameter of the signal in the first time interval, wherein calculating the scaling parameter comprises identifying a first dominant frequency in a first spectrum of the signal in the first time interval
Implementation Method 3
scaling the signal in the second time interval responsively to the scaling parameter to match the signal in the first time interval, wherein scaling the signal in the second time interval comprises identifying a second dominant frequency in a second spectrum of the signal in the second time interval and scaling the second spectrum so that the second dominant frequency matches the first dominant frequency
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method for analyzing a physiological signal includes selecting a first time interval containing a pattern of interest in a recording of the physiological signal. Respective values of a characteristic of the physiological signal are computed in a plurality of time segments within the first time interval. The computed values are concatenated to form a signature of the pattern of interest. A further occurrence of the pattern of interest is identified in the physiological signal during a second time interval by matching the signal in the second time interval to the signature.