Adaptive Filter Leak Estimation for Ventilation Circuits
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing leak estimation methods in gas delivery systems, particularly for non-invasive and invasive ventilation, face challenges such as circular dependence between breath detection and leak estimation, leading to errors in pediatric patients and patients with irregular breathing, and are prone to false triggers due to assumptions about constant discharge coefficients and orifice areas.
Innovation Solution
A method employing a transfer function φ(x) to estimate leak flow Qleak, where x is a set of independent measured or fixed variables, using an adaptive filter to constrain patient flow Qp to zero, ensuring accurate leak estimation over time by decoupling breath detection from leak estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If Bernoulli's orifice flow model is used to estimate leak flow, then the leak estimation is simple and computationally efficient, but it produces errors in pediatric patients and patients with irregular breathing due to circular dependence between breath detection and leak estimation
Solution Approach 1:
The patent segments the leak estimation process into distinct phases: an initial phase using the Bernoulli model for computational efficiency, and a subsequent phase using an adaptive filter for accuracy. This segmentation allows the system to use the simpler model when appropriate while transitioning to the more robust model when reliability is compromised, particularly in pediatric or irregular breathing cases.
Solution Approach 2:
The patent implements a dynamic approach by using an adaptive filter that continuously adjusts the leak estimation based on incoming breath data. The system dynamically switches between estimation methods and updates parameters in real-time, allowing the leak estimation to adapt to changing patient conditions, particularly for irregular breathing patterns where static models fail.
2Device complexity
If constant discharge coefficients and orifice areas are assumed in leak estimation, then the calculation is simplified, but false triggers occur due to the inability to account for varying leak characteristics
Solution Approach 1:
The patent implements feedback mechanisms where the adaptive filter continuously monitors breath detection events and adjusts leak estimation parameters accordingly. When false triggers are detected or breath patterns deviate from expectations, the system uses feedback to recalculate leak characteristics, preventing false positives while maintaining computational simplicity through the adaptive nature of the filter.
Solution Approach 2:
The patent changes the parameters used in leak estimation from fixed constant values to dynamically adjusted parameters through the adaptive filter. The discharge coefficient and orifice area are no longer assumed constant but are continuously updated based on observed breath patterns, allowing the system to maintain low computational complexity while achieving high measurement precision through parameter adaptation.
3Reliability
If breath detection is used to drive leak estimation, then the leak model can be updated in real-time, but circular dependence causes errors in patients with irregular breathing patterns
Solution Approach 1:
The patent introduces an adaptive filter as an intermediary layer between breath detection and leak estimation. This intermediary processes the raw breath detection data, filters out artifacts and false triggers, and provides cleaned input to the leak estimation algorithm. This breaks the circular dependence by decoupling the breath detection feedback loop from the leak model updates, preventing errors in irregular breathing cases while maintaining real-time updating capability.
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 provides a more accurate and reliable estimation of leak flow, reducing errors in breath detection and false triggers, and effectively models leak flow in both non-invasive and invasive ventilation systems, especially in pediatric patients with irregular breathing patterns.
Implementation Method 1
determining a transfer function φ(x) that estimates the leak flow Qleak, where x is a set of independent measured or fixed variables, based on an adaptive filter constraining patient flow Qp to zero
Data Source
AI summary
The disclosed concept maintains that Qp=Qc−Qleak, where, Qp is the estimated patient flow, Qleak is the estimated leak and Qc is the measured total circuit flow. Qleak is given by a transfer function φ(χ) where x is a set of independent measured or fixed variables. The transfer function is thus Qleak=φ(χ). The transfer function (φ(χ) is adjusted given the constraint that, Qp shall be zero. The transfer function converges over time to accurately estimate the leak because over an extended time the mean patient flow will always be zero. In one example, φ(χ)=gorfPγ and the coefficient gorf is adapted until Qp is zero.