Respiratory Rate Measurement via Gaussian Process Uncertainty

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for measuring respiratory rate, particularly in acutely ill or elderly patients, often provide point estimates with unclear uncertainty, making it difficult for clinicians to distinguish between significant readings and those dominated by noise, and are less effective due to poor signal quality and interference from movement artifacts.

Innovation Solution

A probabilistic method using Gaussian Process regression with a periodic covariance function to quantify uncertainty in respiratory rate estimates, allowing for a posterior distribution over hyperparameters that provides both an estimate and its uncertainty, enabling more informed clinical decisions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional measurement methods (spirometry, chest bands, impedance plethysmography) are used to measure respiratory rate, then respiratory rate can be measured, but the equipment interferes with natural breathing or causes discomfort to patients

Engineering Contradiction:
Improverespiratory rate measurementVSAvoidinterference with natural breathing and patient discomfort
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent replaces mechanical measurement devices (spirometry, chest bands, impedance plethysmography) with an optical sensing system using a camera to capture chest movement. This substitution eliminates physical contact and interference with natural breathing while maintaining measurement capability through video analysis of chest wall motion.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces an intermediary approach by using a camera to indirectly measure respiratory rate through chest wall motion rather than direct mechanical contact. The camera captures visual information that serves as a proxy for respiratory activity, eliminating the need for uncomfortable or interfering physical sensors on the patient.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If conventional signal processing methods are used to extract respiratory rate from ECG or PPG signals, then a point estimate of respiratory rate is obtained, but the uncertainty associated with the estimate cannot be directly quantified

Engineering Contradiction:
Improverespiratory rate estimateVSAvoiduncertainty information
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent changes the output parameter from a single point estimate to a probability distribution characterized by mean and variance. This parameter transformation enables simultaneous presentation of the respiratory rate estimate and its uncertainty, allowing clinicians to assess both the value and reliability of the measurement.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent implements feedback by continuously monitoring signal quality metrics and adjusting the uncertainty estimation accordingly. When signal quality deteriorates due to motion artifacts or poor contact, the system automatically increases the uncertainty parameter, providing real-time feedback about the reliability of the respiratory rate measurement.

Inventive Principle:
Principle #23Feedback

3Measurement precision

If existing respiratory rate estimation methods are applied to unwell or elderly patients, then measurements can be obtained, but the methods are less successful compared to healthy volunteers due to poor signal quality and movement artifacts

Engineering Contradiction:
Improverespiratory rate measurementVSAvoidmeasurement reliability in unwell and elderly patients
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent applies dynamic adaptation by adjusting measurement parameters and uncertainty estimates based on real-time signal quality assessment. The system responds to changing physiological conditions and motion artifacts by dynamically modifying processing parameters, enabling reliable respiratory rate measurement across diverse patient populations including unwell and elderly individuals.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent prepares for potential measurement failures by pre-establishing robust uncertainty models that account for motion artifacts and signal degradation. The uncertainty quantification framework is designed beforehand to compensate for poor signal quality, ensuring that measurements from unwell or elderly patients are interpreted with appropriate caution even when signal conditions are suboptimal.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

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

This approach directly quantifies uncertainty in respiratory rate measurements, improving accuracy and reliability, especially for elderly patients, by handling noise and artifacts robustly and generating predictive posterior distributions for better data estimation during missing data periods.

Implementation Method 1

A camera is used to capture chest wall motion

Methodology Applied
Scientific EffectVideo imaging:

Implementation Method 2

Respiratory activity may cause the ECG to be modulated in two fundamental ways: R-peak amplitude (RPA) modulation, which is caused by the movement of the chest due to the filling and emptying of the lungs (which in turn causes a rotation of the electrical axis of the heart and the consequent modulation of the amplitude of the ECG), and respiratory sinus arrhythmia (RSA), which is a frequency modulation, corresponding to a variation in heart rate that occurs throughout the respiratory cycle

Methodology Applied
Scientific EffectElectrical signal modulation:

Data Source

PatentEP3014501B1Measuring respiration
Publication Date: 2019.08.28 OXFORD UNIVERSITY INNOVATION LTD
  • EP3014501B1 patent drawingFigure 1
  • EP3014501B1 patent drawingFigure 2~3
  • EP3014501B1 patent drawingFigure 4(a)~4(e)

AI summary

A method of obtaining information about the rate of a periodic physiological process from a time series of measurements obtained from a patient, comprising: obtaining the time series of measurements; fitting a model defining a probability distribution over functions to the time series of measurements, wherein the model is defined by a mean function and a periodic covariance function; and outputting the result of the fitting as information about the rate of the periodic physiological process.