EIT Lung Data Visualization Using Sigmoid Probability Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional electrical impedance tomography (EIT) devices are insufficient in providing quick and accurate detection of ventilation disorders, particularly in pneumothorax cases, with existing methods failing to deliver reliable and rapid results.
Innovation Solution
A method and device that transform EIT data using a non-linear sigmoid function, with parameters adjusted based on a histogram distinguishing healthy and ventilator-disordered lung data, to assign color values indicating the probability of ventilation disturbances, allowing for rapid and accurate visualization of lung function.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional EIT data visualization methods are used, then the device complexity is low and ease of operation is maintained, but the measurement precision and reliability of ventilation disorder detection are insufficient
Solution Approach 1:
The patent applies preliminary action by pre-defining transformation functions and parameters based on histogram analysis of healthy versus diseased lung data. The sigmoid function parameters (amplitude, position, width) are predetermined through training data, allowing rapid real-time transformation without complex runtime calculations, thus improving detection accuracy while maintaining processing efficiency
Solution Approach 2:
The patent transforms EIT pixel values through parameter changes by applying a sigmoid function with specific parameters (amplitude, position, width) that are optimized based on histogram differences between healthy and diseased states. This parameter-based transformation converts raw impedance data into probability maps, significantly improving ventilation disorder detection precision without requiring complex device architecture
2Productivity
If conventional EIT visualization is used, then the device is simple to operate, but the speed and reliability of pneumothorax detection are insufficient
Solution Approach 1:
The system performs preliminary analysis by pre-determining transformation parameters through histogram comparison of training data. This preliminary action enables rapid real-time detection without sacrificing reliability, as the optimized parameters are already established before clinical use, allowing quick pneumothorax detection with high reliability
Solution Approach 2:
The patent implements feedback by using histogram analysis of training data to optimize sigmoid function parameters. The system continuously refines detection accuracy by comparing transformed output with known diagnostic outcomes, creating a feedback loop that improves both detection speed and reliability through iterative parameter optimization
3Measurement precision
If a previously defined transformation function is applied to tidal images, then false positives are reduced and accuracy improves, but the processing requires additional computational steps
Solution Approach 1:
The patent applies parameter changes by transforming pixel values through a sigmoid function with optimized parameters (amplitude, position, width). This single-parameter transformation approach improves detection accuracy by converting raw data into probability-based visualizations, while the parameter optimization from training data ensures the transformation is computationally efficient rather than overly complex
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method for processing and visualizing data from an electrical impedance tomography of a lung comprises the steps of: acquiring (8) data using an electrical impedance tomography device (3) or providing data acquired using the electrical impedance tomography device (3), generating (9) a visual representation of the data, forming (10) at least one tidal image (12) of a respiratory cycle, and transforming (11) each pixel of the at least one tidal image (12) using a predefined function, whereby the transformation assigns to each of the pixels of the at least one tidal image (12) an absolute value corresponding to a probability of a ventilation disturbance occurring.