CPAP Algorithm Distinguishing OSA and CSA via Breath Morphology

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current respiratory treatment devices cannot accurately distinguish between obstructive sleep apnea (OSA) and central sleep apnea (CSA), leading to inappropriate pressure adjustments that may not effectively treat CSA.

Innovation Solution

A CPAP device equipped with an auto-adjusting algorithm that classifies sleep apnea as either OSA or CSA using a passive machine learning algorithm, analyzing the morphology of pre-apnea breaths to determine the appropriate pressure response.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If current respiratory treatment devices classify all sleep apnea as obstructive sleep apnea and increase positive airway pressure, then treatment for OSA is improved, but treatment for CSA becomes inappropriate and may trigger additional induced CSA events

Engineering Contradiction:

Improvetreatment effectivenessVSAvoidinduced CSA events

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The invention segments sleep apnea into distinct types (OSA and CSA) based on breath morphology analysis. By dividing the homogeneous classification approach into differentiated categories, the system applies appropriate pressure adjustments for each type, preventing inappropriate treatment of CSA that would otherwise trigger induced events.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Instead of assuming all apnea events are OSA and applying pressure increases (conventional approach), the invention inverts the logic by first analyzing breath morphology to determine apnea type, then applying pressure increases only for OSA events. This reversal prevents induced CSA events caused by inappropriate pressure application to CSA patients.

Inventive Principle:

Principle #13The other way round (Inversion)

2Extent of automation

If a passive machine learning algorithm is used to classify sleep apnea type, then automated differentiation between OSA and CSA is achieved, but device complexity increases

Engineering Contradiction:

Improveautomated apnea classificationVSAvoidalgorithm processing complexity

Core Design Contradiction:

Extent of automationVSDevice complexity

Solution Approach 1:

The system employs a passive machine learning algorithm that automatically analyzes breath morphology and classifies apnea type without requiring active user input or complex manual processing. The algorithm serves itself by learning from training data and autonomously making classification decisions, achieving high automation while managing complexity through self-contained processing.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The invention performs preliminary classification of apnea type using breath morphology analysis before applying pressure adjustments. By pre-analyzing the breath pattern and determining whether the event is OSA or CSA, the system prepares the appropriate treatment response in advance, automating the decision-making process while structuring the complexity in a manageable sequential workflow.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20250144335A1Distinguishing between central and obstructive sleep apnea

Publication Date: 2025.05.08 FISHER & PAYKEL HEALTHCARE LTD

AI summary

Apparatuses and methods for detecting sleep apnea and classifying the events as obstructive sleep apnea (OSA) and/or central sleep apnea (CSA) are disclosed herein. The apparatuses can include respiratory treatment devices that have an auto adjusting algorithm that is able to classify a sleep apnea as CSA or OSA so that an appropriate pressure can be applied to the patient depending on the type of sleep apnea detected. The apparatuses and methods can use characteristics of at least one breath preceding the apnea event in classifying the event.