Pharyngeal Phenotyping in OSA Using Manometry and Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current diagnostic methods for obstructive sleep apnea, such as drug-induced sleep endoscopy and imaging modalities, are expensive, require sedation, lack standardization, and have variable inter- and intra-rater assessments, while esophageal manometry is labor-intensive and costly, limiting their effectiveness in predicting pharyngeal collapse patterns.

Innovation Solution

A method utilizing high-resolution manometry data combined with machine learning algorithms to predict pharyngeal collapse locations and degrees by extracting features like high-level breath features, frequency features, and largest negative connected component features, enabling accurate prediction of pharyngeal collapse without the need for sedation or high labor costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If drug-induced sleep endoscopy (DISE) is used to assess pharyngeal collapse patterns, then diagnostic accuracy is improved, but cost and procedural complexity increase due to sedation requirements and specialized equipment

Engineering Contradiction:

Improvepharyngeal collapse assessment accuracyVSAvoidprocedural complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical/optical DISE system (endoscope, sedation apparatus) with a manometry-based system that uses pressure sensors to detect pharyngeal collapse. The manometry system substitutes the complex visual assessment mechanism with a simpler pressure detection mechanism that achieves comparable diagnostic accuracy without requiring sedation or specialized endoscopic equipment

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent employs disposable or reusable manometry catheters with pressure sensors that can be easily inserted and removed, replacing the expensive, complex DISE equipment. The manometry system uses affordable pressure sensing technology rather than costly endoscopic systems, making the diagnostic process more accessible and less complex

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

2Device complexity

If conventional manometry is used for pharyngeal assessment, then cost and procedural complexity are reduced, but measurement precision and objectivity deteriorate due to labor-intensive analysis and unintuitive waveforms

Engineering Contradiction:

Improveprocedural simplicityVSAvoidpharyngeal collapse detection accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent introduces machine learning algorithms as an intermediary between the raw manometry data and the final diagnostic interpretation. The ML model processes the complex pressure waveform data and translates it into objective, standardized pharyngeal collapse assessments, eliminating the need for labor-intensive manual analysis while improving measurement precision and objectivity

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent transforms the complex, multi-dimensional manometry waveform data into simplified, standardized parameters that represent pharyngeal collapse characteristics. By changing the representation parameters from raw pressure values to ML-derived collapse metrics, the system achieves both procedural simplicity and high measurement precision

Inventive Principle:

Principle #35Parameter changes

3Adaptability or versatility

If multiple anesthetic agents and classification systems are used in DISE, then diagnostic comprehensiveness is improved, but inter- and intra-rater variability increases due to lack of standardization

Engineering Contradiction:

Improvediagnostic comprehensivenessVSAvoidassessment reproducibility

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent standardizes the diagnostic parameters by using a unified machine learning model that processes manometry data according to fixed algorithms. This replaces the variable human interpretation and multiple classification systems with consistent, reproducible computational parameters, eliminating inter- and intra-rater variability while maintaining diagnostic comprehensiveness

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent creates a universal machine learning model that can assess all types of pharyngeal collapse patterns through a single standardized system. This universal approach replaces the need for multiple specialized classification systems and anesthetic protocols, achieving both diagnostic comprehensiveness and assessment reproducibility through one unified method

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250226094A1Systems and methods for pharyngeal phenotyping in obstructive sleep apnea

Publication Date: 2025.07.10 VANDERBILT UNIV

AI summary

Systems and methods for pharyngeal phenotyping in obstructive sleep apnea are described herein. An example method includes receiving manometry data for a subject; extracting a plurality of features from the manometry data, where the extracted features include one or more of a high-level breath feature, a frequency feature, or a largest negative connected component (LNCC) feature; inputting the extracted features into a trained machine learning model; and predicting, using the trained machine learning model, at least one of a location of pharyngeal collapse for the subject or a degree of pharyngeal collapse for the subject.