Electrical impedance tomography

from Wikipedia, the free encyclopedia
Figure 1: Electrode arrangement on the chest (thorax): The measuring current introduced via the red electrodes generates a potential distribution in the thorax, which is measured via the green electrodes.
Prototype for Applied Potential Tomography with 16 electrodes by Brian H. Brown from 1987.

The electrical impedance tomography ( EIT ) is a relatively new, non-invasive imaging method , the electrical conductivity measurements on the human body is based. This method is based on the observation that the electrical conductivities of biological tissues differ greatly depending on their nature ( absolute EIT ) and / or functional state ( functional or relative EIT ). In addition to the approaches of absolute and functional EIT , in which alternating currents of a single frequency are mostly used, alternating currents of different wavelengths can also be fed in, for example to address questions about the localization of pathological changes within a tissue type ( EIT spectroscopy ).

If you position several surface electrodes around a certain region of the body on the skin and let higher-frequency alternating currents with low amplitude flow between each pair of electrodes while simultaneously registering the electrical potential using the other electrons, a cross-sectional image (tomogram) is obtained by means of repeated measurements with any variation of the stimulation electrode pair from which one can draw conclusions about the tissue composition within the examined body region (Figure 1).

The main reason for the conductivity of a biological tissue is the content of free ions. This can differ significantly between different types of tissue or body fluids, which is why, for example, muscles and blood can conduct the fed-in measuring current better than fat, bone or lung tissue due to their relatively high content of unbound ions. If this property is used for the anatomical representation of a static state, one speaks of absolute EIT ( a-EIT ).

Since human lung tissue has a conductivity that is about five times lower than that of most other soft tissues inside the chest, the lungs are particularly suitable for imaging methods based on the EIT due to the associated high absolute contrast. In addition, the conductivity of the lungs fluctuates several times between inhalation and exhalation (dynamic contrast), which is why the EIT per se also appears to be suitable for clinical questions associated with inhomogeneities in lung ventilation. Since differential measurements between two or more physiological states are recorded here, one speaks of functional EIT ( f-EIT ).

One advantage of the functional EIT compared to the absolute EIT lies in the fact that inaccuracies due to individual anatomy, poorly conductive skin electrodes and other sources of artifacts can be significantly reduced by simply subtracting the images. These are decisive factors why the greatest advances in EIT development so far have been made in the area of ​​functional pulmonary EIT.

Further hopes for use within clinical routine have been made so far in tumor diagnostics (e.g. as an additional diagnostic tool for mammography), the optimized localization of epilepsy-causing areas of the brain or the early identification of abnormal areas of the cervix, as well as in the diagnosis of gastric emptying disorders (for example, narrowing of the stomach outlet) To localize pathologically suspicious changes within a tissue, mostly alternating currents of varying frequencies are fed in according to the approach of EIT spectroscopy (also referred to as multi-frequency EIT ( MF-EIT )).

The invention of the EIT as a medical imaging method is attributed to John G. Webster with his 1978 publication, but the first scientifically published practical implementation came later by David C. Barber and Brian H. Brown . One of the first tomograms created using EIT was published by them as early as 1983 and shows the cross-section of a human arm using absolute EIT . Since then, the absolute and functional EIT has been intensively developed - the majority of purely morphological applications using absolute EIT , however, are still at a more experimental stage. A further development of the a-EIT is the MF-EIT or electroimpedance spectroscopy (EIS), which registers tissue-typical impedance patterns at varying AC frequencies. Brian H. Brown is also heavily involved in the further development of this technology.

Aside from medical imaging, a principle similar to EIT is also used in geophysics for the representation of underground structures (electrical resistance tomography, ERT). and used in process technology for the quantitative determination of conductive liquids

Basics

As already described above, electrical impedance tomography (EIT) is based on the observation that the electrical conductivity of biological tissue differs greatly depending on its nature, which is primarily due to the different content of free ions.

EIT takes advantage of this by attaching surface electrodes to a specific region of the body on the skin (e.g. using adhesive electrodes, electrode belt or conductive electrode vest) and between 2 (mostly adjacent) electrodes, higher-frequency alternating currents (10 - 100 kHz) with a low amplitude in the single-digit miliampère range. These spread out three-dimensionally in the body and are measured by the remaining electrodes, which are usually arranged in a circle around the examination plane. This process is then repeated, for example, for the closest pair of electrodes, until a complete circle analogous to a complete measuring cycle has taken place. The registered data of such a measurement cycle can be digitally processed further using relatively complex mathematical algorithms to form an image similar to a tomogram.

Figure 2: Visualization of the current flow (shown in blue) and the corresponding equipotential (shown in black) after feeding in via two measuring electrodes that are not directly adjacent using a CT of the human chest. Note the organ-dependent curved current flow according to the respective conductivity.

In the case of the absolute EIT , the morphology (anatomy) of the examined body region should generally be shown. The problem with this form of EIT, however, is the characteristic of current that it is preferably distributed in three-dimensional space according to the lowest resistance (Figure 2) and thus not only within, but also outside the corresponding examination level (impedance transfer). Therefore, the digital creation of the actual "slice image" by means of EIT is also significantly more complicated than with the X-ray-based computed tomography (CT) method , in which linear X-rays penetrate the examination plane to be displayed rotationally from different angles. In contrast, the raw data of an EIT measurement cycle measured by means of absolute EIT provides several options for how the two-dimensional representation correlate could look, since a. due to the variable impedance transfer, cannot draw conclusions about a single and unambiguous possibility of the image plane to be reconstructed. Seen in this way, the EIT by definition does not actually correspond to a "real" tomography method, which also represents a two-dimensional virtual body section in two dimensions, but rather a tomography-like method, which projects a three-dimensional body area onto a two-dimensional plane.

Mathematically, this phenomenon is called the inverse problem , which at first appears difficult or impossible to solve. It is considered incorrectly posed because it does not correspond to Jacques Hadamard's definition of a correctly posed problem ( existence, uniqueness, stability ). Another problem of the absolute EIT is also the different skin conductivities of individual electrodes of a test person, as well as inter-individual differences in the skin conductance of different test persons. Both can cause distorted images or artifacts. Ultimately, the plane to be examined is rarely a circular body, so that intra- and inter-individual differences in electrode positioning contribute to further distortions of the anatomy to be represented (e.g. the human chest). Using active electrodes directly on the patient can significantly improve the signal-to-noise ratio and greatly reduce the likelihood of artifacts occurring. In order to do better justice to the patient's individual chest anatomy, it is also useful to consider a priori data sets on patient size, weight and gender in the image reconstruction. There are now also EIT systems that identify and visualize poorly conductive electrodes directly or exclude them from the image reconstruction.

The functional EIT circumvents this problem largely, by performing in a particular subject, measurements under different experimental conditions, which are accompanied by changes in the electrical impedance. An example is the representation of the regional lung action between inhalation and exhalation, since the electrical conductivity changes linearly many times over between the two examination conditions due to the insulating properties of the air inhaled and exhaled. If, for example, one of the electrodes conducts less well than the other skin electrodes, there will be no significant distortion or artefact formation, since it can be assumed that the relation of the impedance change between inhalation and exhalation will also remain the same on this electrode. Nevertheless, in the functional EIT it is also helpful to consider anatomical a priori data sets in order to be able to merge the most likely organ delimitation depending on patient size, weight and gender with the functional imaging.

EIDORS is a program package for GNU Octave and Matlab published under a GNU / GPL license , which among other things enables the reconstruction and display of EIT measurement data.

The Open Innovation EIT Research Initiative is aimed at the international EIT research community. An experimental EIT package with hardware can be purchased from Swisstom at cost price. The associated EIT open source software enables the further processing of stored raw data with EIDORS and can be downloaded from the Open Innovation EIT Research Initiative .

properties

In contrast to many other tomography methods, EIT does not use ionizing radiation. Since higher-frequency alternating currents in the range between 10 and 100 kHz with currents in the single-digit miliampère range are used, warming effects and nerve stimulation within the examination region can be avoided. EIT can therefore be used continuously on humans. The equipment required by the EIT is also much smaller and cheaper than with conventional tomography methods, so that the EIT is suitable for functional real-time visualization directly at the patient's bed, depending on the issue. The main disadvantage, however, is the lower maximum spatial resolution of the EIT compared to other tomography methods. However, this can be optimized, for example, by using 32 instead of just 16 electrodes. If the EIT system is also constructed with active surface electrodes, the quality of the resulting images can be significantly improved, since this can greatly reduce signal losses, artifacts and interference due to cables, cable length and handling.

Applications

Lung EIT (f-EIT)

Superimposition of a CT image with EIT data using the example of a COPD patient. The reduced ventilation in the ventral area of ​​the left lung (shown in the picture above right) can be clearly seen.

The fact that the medical breakthrough of EIT technology is beginning or taking place in the field of lung function diagnostics is due, on the one hand, to the fact that human lung tissue has about five times less conductivity than other soft tissue in the chest (high contrast), but also to the fact that the electrical conductivity of the lungs fluctuates many times over between maximum exhalation and inhalation. Certain clinical questions can therefore be addressed particularly well with the EIT, especially if they are associated with an inhomogeneity of lung ventilation (e.g. insufficient ventilation or overinflation of individual lung areas, lung collapse, etc.), since intrathoracic impedance changes correlate strongly with changes in regional lung ventilation . Differences in the individual skin conductivity or electrode positioning, which cause difficulties in the purely morphological a-EIT, can be more or less neglected with this form of the relative f-EIT, as these factors vary only slightly between inhalation and exhalation and thus potentially resulting artifacts eliminate yourself. Thanks to the latest advances in digital processing of the raw data obtained, it is now possible for the intensive care physician to visualize the regional lung action directly at the patient's bedside and in real time. After years of prototypes, most of which have not exceeded the research stage (Maltron: Sheffield Mark 3.5 , Timpel SA: Enlight , CareFusion: Goe MF II ), the first series models of intensive care lung function monitors have recently been commercially marketed on a large scale ( Dräger Medical GmbH : Pulmovista 500 or Swisstom AG: Swisstom BB 2 ). In focus centers and larger clinics, these monitors are already used occasionally within the clinical routine, for example in the context of intensive medical treatment of patients with acute progressive lung failure ( Acute Respiratory Distress Syndrome, ARDS ). The increasing spread of these commercial EIT systems will show whether the promising results of animal studies (identification of the optimal PEEP level, avoidance of ventilator-associated lung damage (VILI), detection of a pneumothorax etc.) can also be transferred to humans. The first prospective outcome study on EIT-adapted mechanical ventilation was only recently able to show in an animal model that this is associated with a significant improvement in respiratory mechanics or gas exchange and a significant reduction in histological evidence for VILI.

In addition to the applicability of EIT in intensive care medicine, initial studies with spontaneously breathing patients reveal further possible applications. Especially in patients with obstructive pulmonary diseases (e.g. COPD , cystic fibrosis ), the high temporal resolution of EIT (up to 50 Hz) also allows a regional assessment of time-dependent measurement parameters of the lung function diagnostics (e.g. one-second capacity ). With this group of patients in particular , a comprehensive insight into the pathophysiology of the lungs is expected by overlaying EIT data with morphological image sources such as CT or MRI .

In addition to visual information (e.g. regional distribution of the tidal volume), abstract parameters can also be calculated from the recorded raw data (e.g. change in the intrathoracic gas volume during the intensive care stay) - the latter, however, must still be evaluated and validated. As part of the thoracic EIT, heartbeat-related signals of perfusion (blood flow) can also be filtered out and recorded - the analysis of these data is currently not considered to be fully developed. Should a breakthrough be achieved here, regional imbalances between lung ventilation (ventilation) and lung perfusion (blood circulation) could be mapped in parallel. A corresponding ventilation / perfusion imbalance is often the cause of inadequate oxygen enrichment in the blood (oxygenation) - therapeutic countermeasures can be initiated through detection and localization (e.g. positioning measures, ventilation pressure optimization, etc.).

Additional diagnostics for mammography (MF-EIT)

Due to the relatively low specificity of mammography and magnetic resonance imaging (MRI), false positive screening results are relatively common in routine breast cancer screening, which are associated with high psychological stress for the affected patients and significant costs for the health system, which is why there is a need here for additional or alternative diagnostic examination methods. Supplementary methods can increase the specificity, while alternative preventive methods could reduce or eliminate potential risks and complications from exposure to ionizing radiation (mammography) or the contrast agent gadolinium (MRT). Since the electrical conductivities between normal and malignant breast tissue differ at different frequencies, the potential suitability of MF-EIT for breast cancer screening was investigated. According to a study submitted to the Food and Drug Administration (FDA) (n = 504), the sensitivity and specificity of breast cancer screening were higher when mammography was combined with breast MF-EIT using T-Scan 2000 (TransScan) than when screening using mammography or Breast EIT (sensitivity of 88% versus 82% and 62% respectively; specificity of 51% versus 39% and 47% respectively).

Additional diagnostics for gynecological cancer screening (MF-EIT)

Prof. Brian H. Brown is not only credited with a pioneering role in the development and improvement of the first Sheffield EIT systems, he is still actively involved in research and development of an electro-impedance spectroscope (EIS) based on the MF-EIT. In 2000 he published an experimental study according to which MF-EIT predicts cervical intraepithelial neoplasias (CIN) of type 2 and 3 in the Pap test with a sensitivity ("sensitivity") and specificity of 92% each ("predict") ) to let. Whether this should ultimately be used as an alternative to the smear or as an additional diagnostic agent for better atypical localization has not yet been finally clarified. Brown is co-founder of Zilico Limited, which sells the corresponding spectroscope under the name ZedScan I.

Neuroscientific applications (tumors, epilepsy, ischemia)

While the EIT is significantly inferior to classical imaging methods such as CT and MRT in terms of its spatial resolution in applications for structural imaging of the brain (EIT: approx. 15% of the electrode diameter; CT and MRI: approx. 1 mm), it is CT in terms of its temporal resolution and MRI clearly superior (0.1 milliseconds versus 0.1 seconds). Possible applications would be the intensive medical monitoring of brain activity in adults and children, telemetric long-term measurements in patients for preoperative epilepsy focus localization, as well as the imaging of structural brain pathologies, which occur with clear impedance changes due to pronounced cell swelling on the basis of a disturbed cerebral energy supply, e.g. in the context of cerebral bleeding, Ischemia, oxygen starvation, or hypoglycemia. Despite the limited selection of EIT systems at the time, Holder was able to show in 1992 that changes in cerebral impedance can be measured non-invasively using surface electrodes through the skull. In animal experiments, increases in impedance of up to 100% could be observed in the stroke model, about 10% during an artificially induced seizure. In the meantime, the selection of offered EIT systems is somewhat larger, so that the applied measurement current can also be fed in from electrodes that are not adjacent to one another. Corresponding EITs have not yet been used in routine clinical practice, but clinical studies on stroke and epilepsy are currently being carried out.

Organ blood flow (perfusion)

Due to the good electrical conductivity of blood, the functional EIT could also be suitable for the display of the pulsatile blood flow in tissues of higher impedance, for example for the visualization of the regional pulmonary blood flow. This is possible against the background that the impedances in the regions under consideration between systole and diastole differ significantly depending on the vessel filling, especially when physiological saline solution is injected as a contrast medium.

Sports medicine and homecare sector

In the context of the application in people with healthy lungs or in contrast to the visual representation of regional inhomogeneities in people with lung disease, fewer electrodes are required for the global measurement of abstract parameters. A further development of electro-impedance technology for the sports medicine sector (e.g. determination of VO2) or the homecare sector (e.g. non-invasive measurement of arterial blood pressure) therefore also appears possible and interesting.

supporting documents

  1. ↑ a b c d e f g h i B. H. Brown: Electrical Impedance Tomography (EIT) - A Review. In: J. Med. Eng. Technol. 27 (3), 2003, pp. 97-108.
  2. ↑ a b c d e f g h M. Bodenstein, M. David, K. Markstaller: Principles of electrical impedance tomography and its clinical application. In: Crit. Care Med. 37 (2), 2009, pp. 713-724.
  3. ^ A b c d E. L. Costa, RG Lima, MB Amato: Electrical impedance tomography. In: Curr. Opon. Crit. Care. 15 (1), 2009, pp. 18-24.
  4. ↑ a b c d David S. Holder: Electrical Impedance Tomography. Methods, History and Applications. Institute of Physics, Bristol / Philadelphia 2005, Part 3 Applications .
  5. ↑ OV Trokhanova, YA Chijova, MB Okhapkin include: Possibilities of electrical impedance tomography in gynecology. In: J. Phys. Conference Series. 434, 2013, 012038.
  6. ^ RP Henderson, JG Webster: An Impedance Camera for Spatially Specific Measurements of the Thorax. In: IEEE Trans. Biomed. Closely. 25, 1978, pp. 250-254.
  7. ↑ DC Barber, BH Brown: Applied Potential Tomography "(Review Article). In: J. Phys. E: Sci. Instrum. 17, 1984, pp. 723-733.
  8. ↑ DC Barber, BH Brown, IL Freeston: Imaging Spatial distributions of resistivity using Applied Potential Tomography. In: Electronics Letters. 19, 1983, pp. 93-95.
  9. ↑ M. Schreiner, K. Kreysing: Handbook for exploring the subsurface of landfills and contaminated sites. 1998, ISBN 3-540-59461-2 , pp. 220-226.
  10. ^ MS Beck, R. Williams: Process Tomography: Principles, Techniques and Applications, Butterworth-Heinemann. 1995, ISBN 0-7506-0744-0 .
  11. ↑ a b c d David S. Holder: Electrical Impedance Tomography. Methods, History and Applications. Institute of Physics, Bristol / Philadelphia 2005, Part 1 Algorithms
  12. ↑ a b c W. RB Lionheart: EIT reconstruction algorithms: pitfalls, challenges and recent developments. Review Article. In: Physiol. Meas. 25, 2004, pp. 125-143.
  13. ↑ EIDORS documentation
  14. ↑ A. Boyle, A. Adler: The impact of electrode area, contact impedance and boundary shape on EIT images. In: Physiol. Meas. 32 (7), 2011, pp. 745-754.
  15. ↑ a b B. Rigaud, Y. Shi, N. Chauveau, JP Morucci: Experimental acquisition system for impedance tomography with active electrode approach. In: Med. Biol. Eng. Comput. 31 (6), 1993, pp. 593-599.
  16. ↑ a b P. O. Gaggero, A. Adler, J. Brunner, P. Seitz: Electrical impedance tomography system based on active electrodes. In: Physiol. Meas. 33 (5), 2012, pp. 831-847.
  17. ↑ a b D. Ferrario, B. Grychtol, A. Adler, J. Solà, SH Böhm, M. Bodenstein: Toward morphological thoracic EIT: major signal sources correspond to respective organ locations in CT. In: IEEE Trans. Biomed. Closely. 59 (11), 2012, pp. 3000-3008.
  18. ↑ EIDORS: Electrical Impedance Tomography and Diffuse Optical Tomography Reconstruction Software . Retrieved September 13, 2013 .
  19. ^ RG Cook, GJ Saulnier, DG Gisser: ACT3: A high-speed, high-precision electrical impedance tomograph. In: IEEE Trans Biomed Eng. 41, 1994, pp. 713-722.
  20. ^ T. Lücke, F. Corradi, P. Pelosi: Lung imaging for titration of mechanical ventilation. In: Lashing. Opin. Anaesthesiol. 25 (2), 2012, pp. 131-140.
  21. ↑ Maltron International: The Maltron Sheffield MK 3.5. Retrieved September 13, 2013 .
  22. ^ SA Timpel: Enlight. Retrieved September 13, 2013 .
  23. ^ A. Adler, MB Amato, JH Arnold, R. Bayford, M. Bodenstein, SH Böhm, BH Brown, I. Frerichs, O. Stenqvist, N. Weiler, GK Wolf: Whither lung EIT: where are we, where do we want to go and what do we need to get there? In: Physiol. Meas. 33 (5), 2012, pp. 679-694.
  24. ↑ G. Wolf, C. Gomez-Laberge, J. Rettig, S. Vargas, C. Smallwood, S. Prabhu, S. Vitali, D. Zurakowski, J. Arnold: Mechanical ventilation guided by electrical impedance tomography in experimental acute lung injury. In: Crit. Care. Med. 41 (5), 2013, pp. 1296-1304.
  25. ↑ Bo Gong, Sabine Krueger-Ziolek, Knut Moeller, Benjamin Schullcke, Zhanqi Zhao: Electrical impedance tomography: functional lung imaging on its way to clinical practice? In: Expert Review of Respiratory Medicine . tape 9 , no. 6 , November 2, 2015, ISSN  1747-6348 , p. 721-737 , doi : 10.1586 / 17476348.2015.1103650 , PMID 26488464 .
  26. ↑ Sabine Krueger-Ziolek, Benjamin Schullcke, Zhanqi Zhao, Bo Gong, Susanne Naehrig: Multi-layer ventilation inhomogeneity in cystic fibrosis . In: Respiratory Physiology & Neurobiology . tape 233 , p. 25-32 , doi : 10.1016 / y . or 2016.07.010 .
  27. ↑ Benjamin Schullcke, Bo Gong, Sabine Krueger Ziolek, Manuchehr Soleimani, Ullrich Mueller-Lisse: Structural-functional lung imaging using a combined CT EIT and a Discrete Cosine Transform reconstruction method . In: Scientific Reports . tape 6 , no. 1 , May 16, 2016, ISSN  2045-2322 , doi : 10.1038 / srep25951 , PMID 27181695 ( nature.com [accessed April 26, 2017]).
  28. ^ PT Huynh, AM Jarolimek, S. Daye: The false-negative mammogram. In: Radiographics. 18 (5), 1998, pp. 1137-1154.
  29. ^ CW Piccoli: Contrast-enhanced breast MRI: factors affecting sensitivity and specificity. In: Eur. Radiol. 7 Suppl 5, 1997, pp. 281-288.
  30. ^ PH Kuo, E. Kanal, AK Abu-Alfa, SE Cowper: Gadolinium-based MR contrast agents and nephrogenic systemic fibrosis. In: Radiology. 242 (3), 2007, pp. 647-649.
  31. ↑ J. Jossinet: The impedivity of freshly excised human breast tissue. In: Physiol. Meas. 19 (1), 1998, pp. 61-75.
  32. ↑ TransScan T-Scan 2000 - P970033 , April 24, 2002, Food and Drug Administration .
  33. ^ BH Brown, JA Tidy, K. Boston, AD Blackett, RH Smallwood, F. Sharp: Relation between tissue structure and imposed electrical current flow in cervical neoplasia. In: Lancet. 355 (9207), 2000, pp. 892-895.
  34. ^ Zilico Limited: new standard in colposcopy and cervical cancer diagnostics ... ZedScan I. (No longer available online.) Archived from the original on October 4, 2013 ; Retrieved September 13, 2013 .
  35. ^ PW Kunst, A. Vonk-Noordegraaf, OS Hoekstra, PE Postmus, PM de Vries: Ventilation and perfusion imaging by electrical impedance tomography: a comparison with radionuclide scanning. In: Physiol. Meas. 19 (4), 1998, pp. 481-490.
  36. ^ J. Sola, A. Adler, A. Santos, FS Sipmann, SH Bohm: Non-invasive monitoring of central blood pressure by electrical impedance tomography: first experimental evidence. In: Med. Biol. Eng. Comput. 49 (4), 2011, pp. 409-415.
  37. ^ J. Solà, A. Adler, A. Santos, G. Tusman, FS Sipmann, SH Bohm: Non-invasive monitoring of central blood pressure by electrical impedance tomography: first experimental evidence. In: Med. Biol. Eng. Comput. 49 (4), 2011, pp. 409-415.

Web links