Speech Analysis Lung Volume Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for monitoring lung volumes, particularly in patients with conditions like asthma, COPD, and CHF, are inconvenient and costly, requiring regular visits to hospitals or clinics for spirometer testing.
Innovation Solution
A system and method that uses a smartphone or tablet to capture and analyze a patient's speech, calculating lung volumes by mapping speech features to airflow rates through a learned function, allowing for remote and convenient monitoring.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If spirometer testing is used to measure lung volumes, then measurement precision is improved, but device complexity and ease of operation worsen due to requiring hospital visits and specialized equipment
Solution Approach 1:
The patent replaces the mechanical spirometer system with an acoustic analysis system using a microphone and signal processing algorithm. The speech signal is analyzed to extract features (energy, zero-crossing rate, spectral features) that are mapped to airflow rates, eliminating the need for complex mechanical lung volume measurement devices while maintaining monitoring capability
Solution Approach 2:
The patent creates a computational model that copies the relationship between speech acoustic features and airflow rates. By training a mapping function on calibration data, the system replicates the physiological relationship without requiring direct physical measurement, enabling remote monitoring through standard mobile device microphones
2Measurement precision
If spirometer testing is used to measure lung volumes, then measurement precision is improved, but loss of time increases due to regular hospital or clinic visits
Solution Approach 1:
The patent enables patients to perform self-monitoring at home using their own mobile devices. The system requires no professional operator or specialized facility - patients simply speak into their device and the algorithm automatically processes the speech signal to estimate lung volumes, eliminating travel time and scheduling constraints
Solution Approach 2:
The patent performs preliminary calibration during an initial hospital visit where speech samples are collected and a personalized mapping function is trained. This preliminary action stores the patient-specific relationship between speech features and airflow rates, enabling accurate subsequent measurements at home without requiring repeated professional calibration
3Ease of operation
If speech analysis is used to estimate lung volumes, then ease of operation is improved, but measurement precision may worsen compared to traditional spirometry
Solution Approach 1:
The patent transforms the measurement approach by changing from direct physical measurement of lung volumes to indirect estimation through speech acoustic parameters. Multiple speech features (energy, zero-crossing rate, spectral characteristics) are extracted and combined through a mapping function to estimate airflow rates, which are then integrated to obtain lung volume measurements
Solution Approach 2:
The patent implements a feedback mechanism where estimated lung volumes are compared against expected ranges and trends. The system can generate alerts when deviations from baseline values are detected, and the calibration process allows for iterative refinement of the mapping function based on periodic re-calibration or adjustment of model parameters
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A system (20) includes circuitry (26, 42) and one or more processors (28, 36), configured to cooperatively carry out a process that includes receiving, from the circuitry, a speech signal (62) that represents speech uttered by a subject (22), the speech including one or more speech segments, dividing the speech signal into multiple frames (64), such that one or more sequences (66) of the frames represent the speech segments, respectively, computing respective estimated total volumes of air exhaled by the subject while the speech segments were uttered, by, for each of the sequences, computing respective estimated flow rates of air exhaled by the subject during the frames belonging to the sequence and based on the estimated flow rates computing a respective one of the estimated total volumes of air, and in response to the estimated total volumes of air, generating an alert. Other embodiments are also described.