Pulse Arrival Time Estimation from EIT Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Electrical Impedance Tomography (EIT) methods provide only qualitative information and fail to detect smaller perfused regions like the aorta or pulmonary arteries, as they rely on relative impedance changes and require additional calibration for absolute measurements, and are limited by respiratory and cardiac cycle interference.
Innovation Solution
A method to estimate Pulse Arrival Time (PAT) values from EIT images by generating time-based images where pixel values correspond to PAT values, using ensemble averaging and parametric estimation techniques to provide absolute measurements without prior anatomical knowledge, allowing detection of small anatomical structures like the aorta and pulmonary arteries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If EIT methods use relative impedance changes for monitoring, then the monitoring function is achieved, but the ability to detect small anatomical structures like aorta and pulmonary arteries is limited
Solution Approach 1:
The patent transforms the measurement parameter from relative impedance changes to absolute pulse arrival time values. By estimating PAT from impedance time series and creating time-based images where pixel values represent absolute time measurements, the method achieves superior detection precision for small anatomical structures without increasing device complexity
Solution Approach 2:
The patent replaces the conventional approach of analyzing impedance magnitude changes with a time-domain analysis approach. By substituting the measurement paradigm from spatial impedance distribution to temporal pulse arrival time, the method enhances detection capability while maintaining the same EIT hardware system
2Measurement precision
If EIT methods provide qualitative information only, then the processing is simpler, but absolute physiological measurements cannot be obtained
Solution Approach 1:
The patent implements automatic and unsupervised processing where the system self-calibrates and extracts absolute PAT values without requiring external calibration procedures or manual intervention. The method autonomously processes EIT images to generate absolute physiological measurements, eliminating the need for operator involvement while achieving quantitative results
Solution Approach 2:
The patent performs preliminary ensemble averaging of EIT images to create a reference waveform before extracting PAT values. This preparatory step establishes the temporal baseline needed for accurate absolute measurements, enabling the system to automatically provide quantitative physiological data without additional calibration steps
3Reliability
If conventional EIT methods are used, then the monitoring of blood movement is achieved, but respiratory and cardiac cycle interference limits detection accuracy
Solution Approach 1:
The patent extracts the pulse arrival time information as a separate temporal parameter from the composite EIT signals. By isolating PAT values from the mixed respiratory and cardiac impedance variations, the method achieves reliable detection accuracy while eliminating the confounding effects of respiratory and cardiac cycle interference
Solution Approach 2:
The patent introduces time-based PAT estimation as an intermediary measurement that translates impedance changes into temporal delay values. This intermediary approach converts the affected impedance signals into robust time measurements that are less susceptible to respiratory and cardiac interference, improving overall detection reliability
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 the detection and monitoring of small anatomical structures and provides absolute physiological quantities in measurement units, overcoming the limitations of existing EIT methods by offering automatic and unsupervised processing of EIT images to determine PAT values accurately.
Implementation Method 1
measuring a sequence of temporally discrete electrical impedance tomography (EIT) images during a predetermined measuring time period in a measurement region
Implementation Method 2
estimating the pulse arrival time value from said time series
Data Source
Figure 1(a)~1(d)
Figure 2a~2f
Figure 3
AI summary
Method and apparatus for estimating an arrival time (PAT) value of a subject from a sequence of electrical impedance tomography (EIT) images (10). The method comprises: providing an EIT imaging device adapted to record impedance signal distribution within a measurement region (2) of the subject; measuring a sequence of temporally discrete EIT images (10) during a predetermined measuring time period in the measurement region using the EIT imaging device, each EIT image comprising one or a plurality of EIT pixel subsets, each of said one or a plurality of EIT pixel subset representing an impedance value; generating one or a plurality of time series (11), each of said one or a plurality of time series (11) representing a variation of the impedance value of the sequence of EIT images (10); and estimating the PAT value from each of said one or a plurality of time series (11). The method allows for estimating the PAT value in an automatic and unsupervised fashion.