The effect of serial data collection on the accuracy of electrical impedance tomography images

Yerworth, Rebecca J.; Bayford, Richard

doi:10.1088/0967-3334/34/6/659

Cited by 12 publications

(21 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The variability may be mitigated by choosing appropriate electrode plane, body position and ROIs and/or by adjusting stimulation and measurement patterns [10]. Errors due to hardware imperfections (e.g., electrode movements) may also be accounted by EIT image reconstruction algorithms [20-23]. …”

Section: Discussionmentioning

confidence: 99%

Variability in EIT Images of Lung Ventilation as a Function of Electrode Planes and Body Positions

Zhang¹,

Patterson²

2014

TOBEJ

View full text Add to dashboard Cite

This study is aimed at investigating the variability in resistivity changes in the lung region as a function of air volume, electrode plane and body position. Six normal subjects (33.8 ± 4.7 years, range from 26 to 37 years) were studied using the Sheffield Electrical Impedance Tomography (EIT) portable system. Three transverse planes at the level of second intercostal space, the level of the xiphisternal joint, and midway between upper and lower locations were chosen for measurements. For each plane, sixteen electrodes were uniformly positioned around the thorax. Data were collected with the breath held at end expiration and after inspiring 0.5, 1.0, or 1.5 liters of air from end expiration, with the subject in both the supine and sitting position. The average resistivity change in five regions, two 8x8 pixel local regions in the right lung, entire right, entire left and total lung regions, were calculated. The results show the resistivity change averaged over electrode positions and subject positions was 7-9% per liter of air, with a slightly larger resistivity change of 10 % per liter air in the lower electrode plane. There was no significant difference (p>0.05) between supine and sitting. The two 8x8 regions show a larger inter individual variability (coefficient of variation, CV, is from 30% to 382%) compared to the entire left, entire right and total lung (CV is from 11% to 51%). The results for the global regions are more consistent. The large inter individual variability appears to be a problem for clinical applications of EIT, such as regional ventilation. The variability may be mitigated by choosing appropriate electrode plane, body position and region of interest for the analysis.

show abstract

Section: Discussionmentioning

confidence: 99%

Variability in EIT Images of Lung Ventilation as a Function of Electrode Planes and Body Positions

Zhang¹,

Patterson²

2014

TOBEJ

View full text Add to dashboard Cite

show abstract

“…Three approaches have been proposed: linear interpolation, phase correction in the frequency domain (both on the boundary voltage data) [ 6 ], and use of a regularization prior within the reconstruction algorithm which takes into account the temporal and spatial effects in the sensitivity matrix used [ 7 ]. The intention of this work was to create a standalone tool, which researchers and clinicians could use with their existing data collection, image reconstruction and analysis software.…”

Section: Data Correctionmentioning

confidence: 99%

“…If we consider a typical system, with 16 electrodes, operating at 13 frames per second, the following consequences arise: for a neonate with a breathing rate of up to 60 breaths per minute and an even faster heart rate of 80–150 beats per minute, physiological changes will occur during the time it takes to collect one frame of data. 1 It has been demonstrated in a single case report [ 6 ] that this introduces error of up to 4 % in the reconstructed images, and that these errors are not uniformly distributed. It was calculated that a frame rate 50 times more than the frequency of interest would be needed to reduce this effect to less than the smallest difference the system could measure [ 6 ].…”

Section: Introductionmentioning

confidence: 99%

“…1 It has been demonstrated in a single case report [ 6 ] that this introduces error of up to 4 % in the reconstructed images, and that these errors are not uniformly distributed. It was calculated that a frame rate 50 times more than the frequency of interest would be needed to reduce this effect to less than the smallest difference the system could measure [ 6 ]. Alternatively, due to the systematic nature of the errors, they can be reduced using a mathematical correction [ 6 ].…”

Section: Introductionmentioning

confidence: 99%

“…It was calculated that a frame rate 50 times more than the frequency of interest would be needed to reduce this effect to less than the smallest difference the system could measure [ 6 ]. Alternatively, due to the systematic nature of the errors, they can be reduced using a mathematical correction [ 6 ]. Such a correction would enable the continued use of existing data collection hardware, and an improvement in the accuracy of existing data sets.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Analysis and compensation for errors in electrical impedance tomography images and ventilation-related measures due to serial data collection

Yerworth

Frerichs²,

Bayford

2016

J Clin Monit Comput

Self Cite

View full text Add to dashboard Cite

Electrical impedance tomography (EIT) is increasingly being used as a bedside tool for monitoring regional lung ventilation. However, most clinical systems use serial data collection which, if uncorrected, results in image distortion, particularly at high breathing rates. The objective of this study was to determine the extent to which this affects derived parameters. Raw EIT data were acquired with the GOE-MF II EIT device (CareFusion, Höchberg, Germany) at a scan rate of 13 images/s during both spontaneous breathing and mechanical ventilation. Boundary data for periods of undisturbed tidal breathing were corrected for serial data collection errors using a Fourier based algorithm. Images were reconstructed for both the corrected and original data using the GREIT algorithm, and parameters describing the filling characteristics of the right and left lung derived on a breath by breath basis. Values from the original and corrected data were compared using paired t-tests. Of the 33 data sets, 23 showed significant differences in filling index for at least one region, 11 had significant differences in calculated tidal impedance change and 12 had significantly different filling fractions (p = 0.05). We conclude that serial collection errors should be corrected before image reconstruction to avoid clinically misleading results.Electronic supplementary materialThe online version of this article (doi:10.1007/s10877-016-9920-y) contains supplementary material, which is available to authorized users.

show abstract

Unilateral empyema impacts the assessment of regional lung ventilation by electrical impedance tomography

Bläser¹,

Pulletz

Becher³

et al. 2014

Physiol. Meas.

View full text Add to dashboard Cite

Several studies have shown the ability of electrical impedance tomography (EIT) to assess regional ventilation distribution in human lungs. Fluid accumulation in the pleural space as in empyema, typically occurring on one chest side, may influence the distribution of ventilation and the corresponding EIT findings. The aim of our study was to examine this effect on the assessment of regional ventilation by EIT. Six patients suffering from unilateral empyema and intubated with a double-lumen endotracheal tube were studied. EIT data were acquired during volume-controlled ventilation with bilateral (tidal volume (V(T)): 800 ml) and unilateral ventilation (V(T): 400 ml) of the right and left lungs. Mean tidal amplitudes of the EIT signal were calculated in all image pixels. The sums of these values, expressed as relative impedance change (rel. ΔZ), were then determined in whole images and functionally defined regions-of-interest (ROI). The sums of rel. ΔZ calculated during the two cases of one-lung ventilation either on the affected or unaffected side were significantly smaller than during bilateral ventilation. However, in contrast to previous findings in patients with no pleural pathology, very low values of rel. ΔZ were found when the lung on the affected side was ventilated. ROI-based analysis rendered higher values than the whole-image analysis in this case, nonetheless, the values were significantly smaller than when the unaffected side was ventilated in spite of identical VT. In conclusion, our results indicate that the presence of empyema may affect the quantitative evaluation of regional lung ventilation by EIT.

show abstract

The effect of serial data collection on the accuracy of electrical impedance tomography images

Cited by 12 publications

References 16 publications

Variability in EIT Images of Lung Ventilation as a Function of Electrode Planes and Body Positions

Variability in EIT Images of Lung Ventilation as a Function of Electrode Planes and Body Positions

Analysis and compensation for errors in electrical impedance tomography images and ventilation-related measures due to serial data collection

Unilateral empyema impacts the assessment of regional lung ventilation by electrical impedance tomography

Contact Info

Product

Resources

About