Distinguishability in EIT using a hypothesis-testing model

Adler, Andy; Gaggero, Pascal; Maimaitijiang, Yasheng

doi:10.1088/1742-6596/224/1/012056

Cited by 6 publications

(7 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The distinguishability is the ability to separate two adjacent targets from one while detectability is to distinguish the presence of a target from null (homogenous tank) [7]. Potential data are collected for single and multiple inhomogeneities placed at different locations in the phantom domain.…”

Section: Snr V Sdmentioning

confidence: 99%

Performance Evaluation of a Digital Electrical Impedance Tomography System

Ayati¹,

Bouazza-Marouf²,

Kerr³

et al. 2012

Imaging and Signal Processing in Health Care and Technology / 772: Human-Computer Interaction / 773: Communication, Internet An

View full text Add to dashboard Cite

Performance evaluation of a portable digital multifrequency electrical impedance tomography system is presented. The instrumentation hardware and image reconstruction are assessed according to a systematic methodology using a practical phantom. The phantom is equipped with eight electrodes in a ring configuration and a sinusoidal current of constant amplitude is injected using an adjacent current injection protocol. Artificial anomalies are introduced as inhomogeneity targets and the boundary potential data is collected. The images are reconstructed from the boundary data using Comsol Multiphysics and Matlab. Signal to noise ratio (SNR) and accuracy of the measurements are calculated. The limits of detectability and distinguishability of contrasts are measured from the collected potential data set for single and double inhomogeneities. The conductivity of the targets is successfully reconstructed from the potential data measurements. The detectability value is found to be high when a single target is close to the electrodes, while the values are less for the target in the centre. Also, the value of distinguishability increases when the targets move further away from each other.

show abstract

Section: Snr V Sdmentioning

confidence: 99%

Performance Evaluation of a Digital Electrical Impedance Tomography System

Ayati¹,

Bouazza-Marouf²,

Kerr³

et al. 2012

Imaging and Signal Processing in Health Care and Technology / 772: Human-Computer Interaction / 773: Communication, Internet An

View full text Add to dashboard Cite

show abstract

“…where v is the measurement set, F() characterizes the EIT system, σ is the impedance distribution, and n is system noise [13]. The measurement set v results from directing current through the body with impedance distribution σ and measuring the voltage at the boundary.…”

Section: What Is the Inverse Problem?mentioning

confidence: 99%

“…First, the stimulation and measurement patterns are varied by adjusting the electrode pair separation. Second, the criterion of 'Distinguishability', described in [13], is applied to analyze the pattern performance. Distinguishability uses a hypothesis-testing model where the nullhypothesis is that the impedance change estimate between two conductivity distributions is 0.…”

Section: Distinguishabilitymentioning

confidence: 99%

“…The work aimed to use the distinguishability measure described in [13] to determine optimal measurement strategies. This involved first running simulations on a hemispherical phantom The 10-20 system had the highest distinguishability.…”

Section: Eit and Brain Imagingmentioning

confidence: 99%

“…13 shows a set of images taken from a subject in the Standing posture using the Planar measurement pattern. In the first set, every 10 th image from a full inhalation and exhalation is displayed and represent one full VC breathing cycle.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Electrical Impedance Tomography in 3D: Characterization and Evaluation

Wagenaar¹

View full text Add to dashboard Cite

Electrical Impedance Tomography (EIT) uses body surface electrical stimulation and measurements to create images of the conductivity contrasting fluids. Identifying and monitoring regions of inhomogeneous behaviour in the lungs can be a vital part of the treatment of lung injuries. Traditionally, EIT uses a placement of electrodes in a 2D ring around the thorax, and thus produces cross-sectional images. 3D EIT has a potential to image regional inhomogeneities with 3D spatial extent and to improve planar resolution over 2D EIT. However, while 3D reconstruction algorithms are available, little evaluation has been done to understand the performance of 3D EIT in terms of the measurement configurations available. This thesis focuses on this evaluation task. It has three main objectives. First, to examine different measurement patterns, and determine the best-suited pattern for in vivo 3D lung imaging of regional inhomogeneities.Second, to capture regional inhomogeneities in the lungs caused by gravitational effects by recording EIT on human subjects in the standing, sitting, supping, and decline postures. This method is used in order to cause and measure regional inhomogeneities in the lungs. Third, to verify the ability of 3D EIT to measure global lung volume changes in volunteers by comparing reconstructed images to spirometry lung volume measurements.To address these objectives, the thesis develops an analysis methodology, and applies it to data from simulation, phantom, and measurements on 8 healthy volunteers. Results indicate that 3D EIT can provide meaningful and stable reading of the global change in lung volume. Functional EIT images are created from inhalation curve features to analyze the effect of posture on regional lung behaviour with the greater lung activity moving upwards in the thorax when moving from the standing postures to a declined posture. The Planar and Zigzag Offset patterns show statistically significant difference in vertical lung activity between the decline and standing postures and the supine and standing postures for several fEIT features. While inconsistent, the positive results indicate that 3D EIT is capable of capturing region lung inhomogeneities in 3D space. Overall, this thesis presents analysis methodologies (simulation, phantom, and experimental) to characterize and optimize 3D EIT imaging.iii

show abstract

Deep learning with ExtendeD Exponential Linear Unit (DELU)

Çatalbaş

Morgül

2023

Neural Comput & Applic

View full text Add to dashboard Cite

Distinguishability in EIT using a hypothesis-testing model

Cited by 6 publications

References 3 publications

Performance Evaluation of a Digital Electrical Impedance Tomography System

Performance Evaluation of a Digital Electrical Impedance Tomography System

Electrical Impedance Tomography in 3D: Characterization and Evaluation

Deep learning with ExtendeD Exponential Linear Unit (DELU)

Contact Info

Product

Resources

About