Wearable ECG and Phonocardiogram Screening for Structural Heart Disease
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for screening structural heart disease are limited by the need for skilled human interpretation of heart sounds and lack of continuous, accurate monitoring capabilities, particularly in environments with background noise and motion artifacts.
Innovation Solution
A wearable sensor device measuring electrocardiogram, phonocardiogram, and accelerometer signals at the chest skin surface, using auditory filter models and machine learning algorithms to segment and classify heart sounds for structural heart disease screening.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional stethoscope is used for heart sound auscultation, then the physician can listen to heart sounds, but the diagnosis is limited by human threshold of audibility and physician skill
Solution Approach 1:
The patent replaces the mechanical stethoscope system with an electronic recording and digital signal processing system. Electronic sensors capture heart sounds, which are then processed through digital filters and analyzed by computers, eliminating the limitations of human auditory thresholds and physician skill variability.
Solution Approach 2:
The patent introduces digital signal processing algorithms and computer-based analysis as intermediaries between heart sound capture and diagnosis. These intermediaries enhance the detection capability by filtering background noise, amplifying relevant signals, and providing objective measurements that supplement physician evaluation.
2Loss of information
If digital stethoscope with PCG recording is used, then heart sounds can be recorded as digital waveforms, but the PCG can be obscured by background noise, breathing sounds, and sensor motion
Solution Approach 1:
The patent extracts the harmful background noise and motion artifacts from the PCG signal through digital filtering techniques. By separating and removing these interfering components, the system isolates the genuine heart sound signals for accurate analysis.
Solution Approach 2:
The patent converts motion artifacts and background noise from harmful factors into useful information. By analyzing the characteristics of these干扰 signals, the system can identify and compensate for their effects, or use them to validate the authenticity of detected heart sounds.
3Extent of automation
If automated classification of heart sounds is implemented, then objective diagnosis can be achieved, but accurate heart sound segmentation and classification remains a challenge
Solution Approach 1:
The patent applies segmentation techniques to divide the continuous PCG signal into discrete heart sound events (S1, S2, and abnormal sounds). By segmenting the signal based on temporal patterns, amplitude thresholds, and frequency characteristics, the system enables automated identification and classification of individual heart sounds.
Solution Approach 2:
The patent employs dynamic thresholding and adaptive filtering methods that adjust to varying signal conditions in real-time. This allows the automated classification system to maintain high accuracy across different recording environments, patient conditions, and heart rate variations.
Data Source
AI summary
A wireless wearable sensor device, system, method, and non-transitory computer readable medium for screening for structural heart disease based on electrocardiogram, phonocardiogram, and/or accelerometer signals on a patient's skin surface.