Clinical decision support system for patient-ventilator asynchrony detection and management
Inventors
Gholami, Behnood • Phan, Timothy S.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12194237-B2
Publication Date
2025-01-14
Expiration Date
Abstract
The present disclosure describes a system that automatically detects patient-ventilator asynchrony and trends in patient-ventilator asynchrony. The present disclosure describes a framework that uses pressure, flow, and volume waveforms to detect patient-ventilator asynchrony and the presence of secretions in the ventilator circuit.
Core Innovation
The invention relates to indicating patient-ventilator interaction by receiving ventilator pressure, flow, and/or volume waveform data relating to a ventilator. It extracts a first feature set from the received waveform data and determines whether at least one feature is absent from the first feature set. This absence, by way of a first classifier and/or labeling algorithm, indicates an irregular breath.
The method generates a patient-ventilator interaction indicator waveform comprising a Delta waveform from the ventilator pressure and flow waveform data. The Delta waveform represents a difference between normalized pressure, after correcting for positive end expiratory pressure (PEEP), and normalized flow waveforms. It then extracts a second feature set from the Delta waveform and/or the ventilator pressure, flow, and/or volume waveform data.
A second classifier and/or second labeling algorithm determines at least one class and/or label based at least on the second feature set. Based on the determined class and/or label, the method determines the presence or absence of one or more types of asynchronous patient-ventilator interactions. The approach uses the Delta-derived representation to support classification into asynchronous interaction types.
Claims Coverage
The provided material includes one independent claim. It contains a two-stage classification pipeline: a first classifier/labeling algorithm to indicate irregular breath based on absence of features, and a second classifier/labeling algorithm to assign classes/labels used to determine presence or absence of asynchronous patient-ventilator interaction types. The inventive features appear centered on the Delta waveform (PEEP-corrected, normalized pressure minus normalized flow) and feature-set-based classification for asynchronous interaction detection.
Two-stage feature extraction and classification for interaction indication
extracting a first feature set from the ventilator pressure, flow, and/or volume waveform data; determining whether at least one feature is absent from the first feature set, which absence by way of a first classifier and/or labeling algorithm indicates an irregular breath; extracting a second feature set from the Delta waveform and/or the ventilator pressure, flow, and/or volume waveform data; determining at least one class and/or label by a second classifier and/or second labeling algorithm based at least on the second feature set; and determining a presence or absence of one or more types of asynchronous patient-ventilator interactions based at least on the at least one class and/or label.
Delta waveform generation using PEEP-corrected normalized pressure and normalized flow
generating a patient-ventilator interaction indicator waveform comprising a Delta waveform from said ventilator pressure and flow waveform data, wherein the Delta waveform represents a difference between normalized pressure, after correcting for positive end expiratory pressure (PEEP), and normalized flow waveforms.
Overall, the claim coverage is anchored in generating a PEEP-corrected Delta waveform from normalized pressure and normalized flow, extracting first and second feature sets, using absence of features to indicate irregular breath, and using a second classifier/labeling algorithm to assign classes/labels that drive a presence/absence determination of asynchronous patient-ventilator interaction types.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?