Methods and systems for pulmonary artery pressure and cardiac synchronization monitoring
Inventors
Maidens, John • GUO, LING • Venkatraman, Subramaniam • Landgraf, Connor
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12303237-B2
Publication Date
2025-05-20
Expiration Date
Abstract
Various methods and systems are provided for monitoring a pulmonary artery pressure and cardiac synchronization of a subject. In one example, a method includes acquiring at least one of electrocardiogram (ECG) data, phonocardiogram (PCG) data, and seismocardiogram (SCG) data from a subject via a digital stethoscope, inputting one or more of the ECG data, the PCG data, and the SCG data into a machine learning algorithm, and estimating at least one of a pulmonary artery pressure and a cardiac synchronization of the subject using the machine learning algorithm. In this way, the pulmonary artery pressure and the cardiac synchronization may be estimated using artificial intelligence and inputs that are non-invasively measured by the digital stethoscope, allowing conditions like heart failure and pulmonary hypertension to be more simply and reliably detected and monitored.
Core Innovation
The invention provides a method executable via a processor that uses a handheld digital stethoscope to acquire time-synchronized electrocardiogram (ECG) data, phonocardiogram (PCG) data, and seismocardiogram (SCG) data. The ECG data, PCG data, and SCG data are captured from a subject via a microphone, an ECG sensor, and an accelerometer of the handheld digital stethoscope, and sensor location information for the microphone, ECG sensor, and accelerometer with respect to the heart is used together with the time-synchronized signals.
A machine learning model is trained with ECG data, PCG data, and SCG data acquired by the handheld digital stethoscope and corresponding invasively measured pulmonary artery pressure and echocardiographic imaging results. In application, the time-synchronized ECG, PCG, and SCG data together with locations of the accelerometer, microphone, and ECG sensor with respect to the heart are input into the trained machine learning algorithm to output at least one of an estimated pulmonary artery pressure and an estimated cardiac synchrony, and signals from at least one of a weight scale, a blood pressure monitor, a respiration monitor, and a pulse oximeter are additionally input into the machine learning algorithm.
Based on at least one of the estimated pulmonary artery pressure and the estimated cardiac synchrony, the method outputs a condition of the subject that includes a presence or absence of a cardiac pathology. The condition determined from the estimated outputs includes heart failure and pulmonary hypertension states, including presence or absence, risk score, severity, and subtype, and decision logic uses fuzzy logic or rules-based logic programmed according to clinically-relevant cardiac synchrony and pulmonary artery pressure values.
Claims Coverage
The document includes two independent claims, a method and a system, centered on a handheld digital stethoscope acquiring time-synchronized ECG, PCG, and SCG signals and using a machine learning model trained with invasively measured pulmonary artery pressure and echocardiographic imaging results, followed by decision logic to determine cardiac pathology states.
Handheld digital stethoscope acquisition of time-synchronized ECG, PCG, and SCG
Capturing ECG data, PCG data, and SCG data from a subject via a microphone, an ECG sensor, and an accelerometer of a handheld digital stethoscope to capture electrical signals, longitudinal chest oscillations, and transverse chest oscillations, with the signals being time-synchronized.
Training a machine learning model from invasively measured pulmonary artery pressure and echocardiographic imaging results
Training a machine learning model by inputting the ECG data, PCG data, and SCG data acquired by the handheld digital stethoscope along with corresponding invasively measured pulmonary artery pressure and echocardiographic imaging results, to output at least one of an estimated pulmonary artery pressure and an estimated cardiac synchrony.
Estimating pulmonary artery pressure and cardiac synchrony using signals and sensor locations
Inputting ECG data, PCG data, SCG data, and locations of the accelerometer, microphone, and ECG sensor with respect to the heart into the trained machine learning algorithm, with optional inputting of signals from at least one of a weight scale, a blood pressure monitor, a respiration monitor, and a pulse oximeter, to output at least one of the estimated pulmonary artery pressure and the estimated cardiac synchrony.
Determining cardiac pathology presence or absence from estimated pulmonary artery pressure and cardiac synchrony
Outputting a condition of the subject determined based on at least one of the estimated pulmonary artery pressure and the estimated cardiac synchrony, where the condition includes a presence or absence of a cardiac pathology and includes heart failure and pulmonary hypertension states.
Cardiac exam system architecture for machine learning estimation and decision logic output
A system including a handheld digital stethoscope having a microphone, an ECG sensor, and an accelerometer and a processor configured to capture time-synchronized PCG data, SCG data, and ECG data, to input these signals and locations into a machine learning model trained to output an estimated pulmonary artery pressure and an estimated cardiac synchrony, and to enter these estimated values into decision logic using fuzzy logic or rules-based logic to output a heart failure status and a pulmonary hypertension status and determine the presence or absence of cardiac pathology.
Across the independent claims, the inventive focus is the handheld digital stethoscope acquisition of time-synchronized ECG, PCG, and SCG signals and sensor locations, training and using a machine learning model tied to invasively measured pulmonary artery pressure and echocardiographic imaging results to estimate pulmonary artery pressure and cardiac synchrony, and applying decision logic to determine presence or absence of cardiac pathology including heart failure and pulmonary hypertension states.
Stated Advantages
Faster diagnosis or screening.
Less expensive diagnosis or screening.
More comfortable diagnosis or screening.
Reduced need for invasive catheterization and advanced imaging.
Documented Applications
Non-invasive monitoring and diagnosis or screening of pulmonary artery pressure and cardiac synchrony, used to determine heart failure and pulmonary hypertension statuses including presence or absence, risk score, severity, and subtype.
Interested in licensing this patent?