Audio-Transducer EIT Emulation for Portable Lung Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing EIT systems are expensive, bulky, and require trained personnel, limiting their accessibility to remote or underserved populations, while stethoscopes are accessible and easy to use but not utilized for generating clinically useful emulations of EIT images.
Innovation Solution
A deep neural network is trained using EIT images and audio signals from the thoracic cavity to encode and decode audio signals into emulations of EIT images, utilizing systems like digital or analog stethoscopes and wearable garments with integrated sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional EIT systems are used, then lung function assessment capability is provided, but device cost and complexity increase significantly
Solution Approach 1:
The patent replaces the complex electrical measurement system with an acoustic measurement system. Instead of using electrodes and electrical current to assess lung function, the system uses audio transducers (microphones) to capture breath sounds and converts these acoustic signals into EIT image emulations through AI processing. This substitution dramatically reduces system complexity while maintaining diagnostic capability.
Solution Approach 2:
The patent creates a computational copy or emulation of EIT images from audio data. Rather than directly capturing electrical impedance data, the system generates synthetic EIT images that replicate the diagnostic information through machine learning models trained on the relationship between audio signals and EIT images. This copying approach simplifies the hardware requirements while preserving the essential diagnostic output.
2Measurement precision
If trained personnel are required for EIT measurements, then measurement accuracy is maintained, but accessibility to remote populations is reduced
Solution Approach 1:
The system enables self-service operation by automatically processing audio recordings through AI algorithms without requiring trained personnel. The machine learning models perform the diagnostic analysis autonomously, converting audio signals into EIT image emulations automatically. This eliminates the need for skilled operators while maintaining measurement accuracy, thereby improving accessibility to remote and underserved populations.
3Productivity
If EIT imaging is performed, then continuous monitoring capability is provided, but equipment portability is reduced
Solution Approach 1:
The patent replaces bulky electrical measurement equipment with lightweight audio transducers. The EIT system requires extensive hardware including electrode arrays and signal processing equipment, whereas the alternative system uses simple microphones that can be integrated into wearable garments or handheld devices. This substitution enables continuous monitoring with high portability.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables clinicians to assess lung and organ function more quickly and easily, especially for remote patients, facilitating earlier disease detection and improving patient care without the need for expensive equipment or trained personnel.
Implementation Method 1
audio signals captured at one or more positions on the thoracic cavity
Data Source
AI summary
Systems and techniques for creating clinically useful emulations of EIT scans by using audio sensors to capture sounds from the thoracic cavity, particularly but not solely lung sounds, where a neural network-based Audio Encoder is, in an embodiment, trained by simultaneously performing both an EIT scan and an audio recording of a cohort of test patients. The EIT scans are passed through an image encoder/decoder pair, each of which can also be neural network-based, and the image encoder creates an embedding representative of the EIT scan. The frequency characteristics of the audio signals are captured as an intermediate representation and supplied to the audio encoder which is trained to map an embedding of the converted audio to the embedding of the EIT scan from the image encoder. At run time audio of a patient is processed through the frequency conversion and supplied to the now-trained Audio Encoder to generate an embedding. The embedding is supplied to the image decoder which produces the EIT emulation for review by a clinician.