Adaptive Filtering for Lung Sound Signal Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional electronic stethoscopes often filter out both heart and lung sounds, making it difficult to separate and retain lung sound signals effectively.
Innovation Solution
A filtering system and method that uses an adaptive filtering device to decompose sound signals into primary lung and reference heart sound signals, adjusts the reference heart sound signal with a weighted value, and subtracts it from the lung sound signal to generate a filtered lung sound signal, thereby separating heart sound noise from lung sound signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If traditional filtering methods are used in electronic stethoscopes, then heart sounds can be filtered out, but lung sounds are also filtered out simultaneously
Solution Approach 1:
The sound signal is segmented into multiple frequency bands using bandpass filters. The filtering system divides the overall frequency spectrum into different ranges, with each bandpass filter targeting specific frequency characteristics of heart sounds versus lung sounds. This segmentation allows selective processing of different sound components without affecting the entire signal spectrum uniformly.
Solution Approach 2:
Different filtering operations are applied to different frequency bands. The system applies stronger filtering to frequency ranges where heart sounds dominate, while applying minimal or no filtering to frequency ranges where lung sounds are prominent. This local quality approach ensures that filtering intensity is adapted to the specific characteristics of each frequency band, preserving lung sounds while removing heart sound noise.
2Object-affected harmful factors
If aggressive filtering is applied to remove heart sound noise, then heart sound interference is reduced, but lung sound signals are also attenuated
Solution Approach 1:
The filtering system dynamically adjusts filter coefficients and parameters based on the detected sound signal characteristics. The bandpass filters and adaptive filtering components modify their behavior in real-time according to the incoming signal's frequency content and amplitude distribution. This dynamic adaptation allows the system to maintain high lung sound signal quality while effectively suppressing heart sound noise, as the filtering strength is continuously optimized rather than fixed.
Solution Approach 2:
The system changes filtering parameters such as cutoff frequencies, bandwidth, and filter order based on the analyzed sound signal properties. By adjusting these parameters dynamically, the system can optimize the balance between noise removal and signal preservation for each specific listening condition, ensuring reliable lung sound detection without excessive attenuation.
3Device complexity
If simple filtering is used, then device complexity is low, but the ability to separate heart and lung sounds is insufficient
Solution Approach 1:
The filtering system is segmented into multiple functional modules: bandpass filter stage, adaptive filtering stage, and signal combination stage. Each module performs a specific function in the signal processing chain, making the overall complex task of sound separation manageable through modular design. This segmentation allows the system to achieve high measurement precision while keeping each individual module relatively simple.
Solution Approach 2:
The system merges multiple filtering techniques (bandpass filtering and adaptive filtering) into a unified processing pipeline. By combining these different filtering approaches, the system achieves superior sound separation capability that neither method could achieve alone, while the integrated design prevents excessive complexity through coordinated operation of the merged components.
Data Source
AI summary
A filtering method includes the following steps: receiving an sound signal; decomposing the sound signal into a primary lung sound signal and a reference heart sound signal; adjusting the reference heart sound signal according to a weighted value to generate an adjusted heart sound signal; and subtracting the adjusted heart sound signal from the primary lung sound signal to generate a filtered lung sound signal.