Bayesian Signal Detection for Recurrent Feature Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for automatically detecting recurrent features in biological signals, such as electrocardiogram (ECG) signals, require human expertise and are inefficient for health monitoring applications, as they lack automated detection capabilities.
Innovation Solution
A method using probability density functions and Bayesian frameworks to parameterize ECG signals, determining posterior probabilities of feature hypotheses based on prior and conditional probabilities, enabling automated detection and updating of probability density functions for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated detection methods are implemented, then productivity and efficiency are improved, but measurement precision and reliability deteriorate due to lack of human expertise
Solution Approach 1:
The patent introduces probability density functions and Bayesian frameworks as intermediary mathematical tools that bridge automated detection and expert-level precision. These probabilistic models serve as mediators that process signal features and produce reliable detection results without requiring human expertise, thus resolving the contradiction between automation and accuracy
Solution Approach 2:
The patent transforms the detection problem by changing parameters from deterministic threshold-based detection to probabilistic parameter estimation. By using probability density functions to model signal features and applying Bayesian inference to update hypotheses, the system achieves both automated operation and expert-level precision through parameter transformation
2Measurement precision
If probability density functions and Bayesian frameworks are used, then measurement precision and reliability are improved, but device complexity increases
Solution Approach 1:
The patent segments the complex detection problem into manageable components: signal parameter extraction, probability density function construction for each parameter, hypothesis formulation, and Bayesian inference. This segmentation allows the complex probabilistic framework to be implemented as a structured sequence of simpler operations, reducing overall system complexity while maintaining high precision
Solution Approach 2:
The patent creates a universal detection framework that can handle multiple types of signal features and hypotheses using the same probabilistic machinery. The Bayesian framework serves multiple functions: updating hypotheses, combining evidence from multiple parameters, and providing unified decision criteria. This multi-functionality reduces complexity by avoiding the need for separate specialized algorithms for different detection tasks
Data Source
Figure 1
Figure 2A~2D
Figure 3~4
AI summary
A method of detection of a recurrent feature of interest within a signal comprising: obtaining evidence, based on a signal, the evidence including a probability density function for each of a plurality of parameters for parameterizing the signal, including at least one probability density function for a parameter, of the plurality of parameters, that positions a feature of interest within signal data of the signal; parameterizing a portion of the signal data from the signal based upon a hypothesis that a point of interest in the signal data is a position of the feature of interest; determining a posterior probability of the hypothesis being true given the portion of the signal data by combining a prior probability of the hypothesis and a conditional probability of observing the portion of the signal data given the hypothesis, wherein the conditional probability of observing the portion of the signal data given the hypothesis is based at least upon: the parameterization of the portion of the signal data and the probability density function for at least one of the plurality of parameters; using the posterior probability to determine whether or not the hypothesis is true; the method further comprising: updating at least one of the probability density functions for the plurality of parameters using the parameterization of the portion of the signal data.