Physiological Rhythm Signal Estimation via Prior Probability Refinement

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Unobtrusive and unconstrained measurement systems for physiological rhythms face challenges in providing reliable and accurate estimations of instantaneous frequencies due to varying signal quality and changes in posture or movement.

Innovation Solution

A computer-implemented method that processes a signal representing a physiological rhythm by estimating a first group of period characteristic estimations, generating a prior probability based on a subset of these estimations, and then estimating a second group of period characteristic estimations using the prior probability, thereby improving accuracy especially in cases of movement or posture change.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If unobtrusive measurement systems are used for continuous monitoring, then user compliance and comfort are improved, but signal quality and measurement reliability deteriorate due to uncontrolled environments and posture changes

Engineering Contradiction:
Improveuser complianceVSAvoidsignal quality
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The system performs preliminary actions by continuously estimating instantaneous frequencies and building a probability distribution model before actual measurement needs arise. This allows the system to pre-characterize the relationship between sensor readings and physiological parameters under various conditions, enabling more reliable measurements when actually needed without requiring controlled environments or user compliance during the measurement process itself

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by using estimated instantaneous frequencies to update and refine the probability distribution model over time. This continuous feedback loop allows the system to adapt to individual user characteristics and environmental variations, improving measurement reliability while maintaining the unobtrusive nature of the monitoring system

Inventive Principle:
Principle #23Feedback

2Measurement precision

If conventional algorithms based on feature detection are used, then measurement precision is improved for controlled signals, but adaptability deteriorates when signal morphology changes due to movement or posture changes

Engineering Contradiction:
Improveinstantaneous frequency estimationVSAvoidsignal morphology robustness
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The system applies parameter changes by transforming the measurement approach from direct feature detection to probabilistic parameter estimation. Instead of relying on fixed morphological features that change with posture, the system estimates instantaneous frequencies and uses these to build a probability distribution that adapts to different signal conditions, maintaining measurement precision across varying morphologies

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system achieves universality by creating a measurement approach that works across multiple signal types and conditions. The probability distribution model can handle different physiological rhythms and signal morphologies uniformly, making the system adaptable to various postures and movement states while maintaining consistent measurement precision through the unified probabilistic framework

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4493049B1Method, computer program product, computer-readable storage medium and system
Publication Date: 2025.04.23 KONINKLIJKE PHILIPS NV
  • EP4493049B1 patent drawingFigure 1~2
  • EP4493049B1 patent drawingFigure 3
  • EP4493049B1 patent drawingFigure 4

AI summary

There is provided a method to estimate a period characteristic, such as period length or a period location, of a physiological rhythm, such as a cardiac rhythm or a spontaneous breathing rhythm. The method comprises receiving the signal representing the physiological rhythm, estimating a first group of period characteristic estimations based on the signal, generating a prior probability for the period characteristic based on at least a subset of the first group of period characteristic estimations, and estimating a second group of period characteristic estimations based on the first group and the prior probability. By using this method, the accuracy of the signal is improved, especially when the signal is obtained from an accelerometer arranged on the chest of the subject.