Pith.

Image and Video Processing · arXiv:2611.17225 v1 · submitted 12 Nov 2026 · 27 references

Data Archive for 2D and 3D Electrical Impedance Tomography with moderate conductivity contrast phantoms

S.J. Hamilton, A. Hauptmann, A. Hänninen, V. Kolehmainen, J. Niskanen, T. Savolainen, J. Toivanen

The paper says

“This document describes open Electrical Impedance Tomography (EIT) measurement data that were collected from biomedical imaging inspired phantoms with moderate conductivity contrasts.”

more of the abstract

The data were collected with the KIT4 and KIT5 EIT measurement devices at the University of Eastern Finland, and are intended for EIT image reconstruction algorithm development and testing. The data from thoracic imaging inspired cylindrical and chest-shaped phantoms are suitable for 2D algorithms, and the data from an intracerebral hemorrhage imaging inspired head-shaped phantom with resistive skull are suitable for 3D algorithms. In addition to the measurement data, computational meshes and example reconstruction codes are provided and sample reconstructions included.

from the abstract

Pith says

Machine review · deepseek-v4-flash · 13 Nov 2026 · what was read was not recorded · not checked by a person

It can be wrong. Challenge this review or reply to it.

The passes disagreed. Reader: holds up. Skeptic: agrees with the reader. Referee: major revision.

A solid data descriptor for moderate-contrast EIT phantoms; the conductivity uncertainty is real but the archive is still useful. This line is wrong

3 major objections · 6 minor

How often objections from reviews like this one are right: not yet measured

Are you an author? What you can do.

Not checked: figures, code, what the 27 references say.

The paper says, at §2.1–2.2, §5

The review did not quote this passage.

Pith says: Objection 1 of 3 · major

The ground-truth conductivity values for the agar targets and conductive cylinders are based on single measurements with no reported uncertainty, temperature control, or cross-check against tank conditions. As the skeptic notes, agar conductivity is temperature-sensitive and ion exchange with the saline bath could shift values over time.

more of objection 1

The only validation reported in §5 is that computed saline estimates 'closely matched' measured values, but no error bounds are given and this covers only the background, not the targets. Since the archive's value as a quantitative benchmark depends on these values being representative, the authors should either provide uncertainty estimates, describe temperature monitoring, or add a cross-check (e.g., measuring a target's conductivity before and after immersion). Without this, the 'approximately known' claim is unquantified, which limits the dataset's use for rigorous benchmarking.

The paper says, at §4, §5

The review did not quote this passage.

A classifier read this sentence of the paper as acknowledging the concern (probability 0.70):

“The conductivity changes are nominally in mS/cm; however, using 2D computations for 3D setups makes the values inexact.”

Pith says: Objection 2 of 3 · major

The paper states that using 2D computations for 3D setups makes the conductivity values 'inexact' (b101). This is an important limitation for the 2D data, but the authors do not quantify the expected error or discuss how users should interpret the reconstructed values. Since the archive is intended for algorithm development, a brief discussion of the magnitude of 3D effects (e.g., from simulations or prior work) would help users set realistic expectations.

The paper says, at §5

The review did not quote this passage.

Pith says: Objection 3 of 3 · major

The technical validation section lists publications that used the same measurement devices with similar phantoms, but does not provide direct evidence that the specific shared datasets are free of systematic errors (e.g., electrode contact issues, drift, or noise beyond the stated averaging).

more of objection 3

While the use in prior peer-reviewed work is supportive, it does not substitute for a dedicated validation of the archived data. The authors should consider adding a simple reproducibility check, such as comparing two independent measurements of the same phantom setup, or reporting the standard deviation across repeats.

6 minor objections

The paper says, at §2.1

Pith says: minor objection 1 of 6

The sentence 'The exact conductivities of the agar targets were measured in a conductivity measurement cell using a Hioki 3532-50 conductivity meter' uses 'exact' which is inconsistent with the later hedge 'approximately known'. Suggest rewording to 'measured' or 'estimated'.

The paper says, at §2.2

Pith says: minor objection 2 of 6

The conductivity of the saline for the head phantom is given as 4.00 mS/cm measured with an RS PRO meter, but no frequency or temperature is specified. Since conductivity meters are often temperature-compensated or require manual temperature input, this information should be included.

show 4 more

The paper says, at §3

Pith says: minor objection 3 of 6

The description of the data files is clear, but it would be helpful to explicitly state the dimensions of the current and voltage matrices (e.g., number of current patterns × number of electrodes) for each phantom, as this is not immediately obvious from the text.

The paper says, at §5

Pith says: minor objection 4 of 6

The sentence 'The conductivity measurements were validated against computed estimates of the phantoms’ saline solution conductivities, which closely matched the measured values' is vague. Please provide the numerical values or a reference to a figure/table showing this comparison.

The paper says, at §1.2

Pith says: minor objection 5 of 6

The phrase 'finite, approximately known conductivity values' is used twice. Consider defining 'finite' (as opposed to infinite contrast) earlier to avoid confusion.

The paper says, at §6

Pith says: minor objection 6 of 6

The note about CGO-based methods using only odd-numbered electrodes is helpful, but it would be clearer to state that the even-numbered electrodes are voltage-only measurement electrodes, as mentioned earlier.

What would settle it

Re-measure each of the four agar target types and the conductive cylinders under the data-collection session conditions: same preparation batch, same temperature, and matched immersion time in the respective saline, with at least three independently prepared samples and repeated cell readings, and report mean plus or minus standard uncertainty; verify that the stated 0.61, 0.67, 3.05, 3.23, and 4.70 mS/cm values are reproduced within a declared tolerance such as ten percent. A complementary end-to-end check is to simulate voltages on the provided electrode-modeled forward meshes with the stated conductivities, fitting contact impedances on the saline-only frames, and compare predicted versus measured voltages; systematic bias concentrated in target-containing frames would reveal ground-truth drift.

from the skeptic's reading

How the review went

4 readings by deepseek-v4-flash · 13 Nov 2026

  1. Reader holds up confidence moderate
  2. Skeptic agrees 1 min 58 s · 10,344 tokens read, 13,329 written
  3. Editor wrote the verdict above 9 s · 11,267 tokens read, 577 written
  4. Referee major revision: 3 major objections, 6 minor 11 s · 10,731 tokens read, 1,124 written

Not run on this paper: the simulated authors' defence.

Pith did not check this review's objections against the paper.

The review in full

The review in brief, machine-written

This is a straightforward data descriptor that fills a genuine gap: open EIT phantom data with moderate-contrast agar targets and a 3D head phantom with a resistive skull and partial electrode coverage. The earlier open archive [20] used near-perfect conductors/insulators, and [15] had full boundary coverage. This release gives the community something closer to biomedical contrast levels, with voltages measured on all electrodes including injectors, which matters for CGO-based methods. The archive includes meshes, precomputed Jacobians, and sample reconstruction code—everything a group needs to get started without building a tank.

The paper does what a data descriptor should: describes the phantoms, the measurement devices (KIT4 and KIT5), the current patterns, the file formats, and shows sample reconstructions. The validation section lists prior publications that used these or similar measurements, which is reasonable evidence that the data are usable. The authors are transparent about the 2D/3D mismatch (Section 4, b101) and about what code is and isn't provided (b100).

The soft spot is the one the stress-test flags: the ground-truth conductivities rest on single measurements with no quoted uncertainty, temperature control, or tank-condition cross-check. The head phantom contrast is only ~17%, so a modest error in either the target or saline conductivity changes the actual contrast materially. The paper's own hedge ('approximately known') is honest, but the tolerance is never quantified. The only cross-check in Section 5 covers saline, not the targets. This doesn't invalidate the archive—the data are still useful for algorithm development and relative comparisons—but it does limit how confidently one can treat the stated conductivities as ground truth for absolute imaging benchmarks.

Who this is for: EIT researchers who need real moderate-contrast phantom data to test reconstruction algorithms, especially those working on CGO-based or difference imaging methods. The archive is a practical resource, not a theoretical contribution.

Recommendation: Accept. The paper is competent, the data are real, and the limitations are acknowledged. The conductivity uncertainty is a real concern but not a fatal one—the archive's value for algorithm comparison and development survives it. I'd send it to a referee who knows EIT hardware to check the measurement description, but I wouldn't hold it up over the uncertainty issue if the authors add a sentence or two acknowledging it.

The strongest objection

The central claim is that the archive holds EIT datasets 'based on finite, approximately known conductivity values.' The load-bearing premise is that the target conductivities reported in Sections 2.1–2.2 (b55, b59, and the 4.70 mS/cm conductive cylinders) were representative while the KIT4 and KIT5 data were being collected. Each target was measured once in a conductivity cell: no uncertainty is reported, no temperature control or recording is described, no replicate measurements are mentioned, and nothing compares cell conditions to tank conditions. Agar conductivity varies strongly with temperature, and ion exchange between agar targets and the surrounding saline of differing ionic strength can shift both target and bath conductivity over the unstated time targets spend in the tank. The head phantom is the sharpest case: the 4.70 versus 4.00 mS/cm contrast is only about 17%, so a modest error or drift in either value materially changes the contrast actually measured by KIT5. The only quantitative cross-check in Section 5 is that computed saline estimates 'closely matched' the measured values, with no error bounds and covering the background rather than the targets. The paper's own hedge ('approximately known') means the claim survives small deviations, but the advertised tolerance is never stated, so the archive's quantitative benchmark value is conditional on an unquantified accuracy assumption.

This is a competent, modestly scoped data descriptor, and the reader's verdict reflects that fairly. The central claim is checkable: an open archive of 2D and 3D phantom EIT data with documented setups, meshes, Jacobians, and example code. The internal description is consistent (electrode counts, current patterns, conductivity values, and device capabilities all align), the authors disclose the main model limitation (sentence b101, the 2D-versus-3D inexactness), and the data have prior use in cited published works, which is real supporting evidence. The omission of the total-variation code (b100) is acknowledged and does not undercut the sample-code claim, since linear difference codes are included. My stress-test pass lands on the same place the reader did: the reported target conductivities are the load-bearing ground truth for the 'approximately known' claim, and they rest on single cell measurements without uncertainty, temperature control, or replicate samples. The concern is genuine but bounded. The claim is hedged in the paper's own wording, the moderate-contrast regime characterization would survive modest drift, and the archive's primary stated purpose (algorithm development and testing, including difference imaging that is largely insensitive to absolute ground-truth error) does not collapse even if the values are imperfect. The head phantom's low 17% contrast is where the concern bites hardest, because the effective target-to-background ratio is exactly what a benchmarking user must trust. Because the reader already priced this into MODERATE confidence and the concern does not falsify the modest central claim, I recommend no change to the ACCEPT verdict. The proposed metrology check, or the complementary forward-model residual test, would convert the concern from a caveat into a quantified property of the archive. I found no internal inconsistencies, no hidden circularity in the validation, and no mis-cited external result; the remaining risk of repository accessibility is a verification step rather than an argument-based defect.

The referee's report

If the data are reliable and well-documented, this archive fills a gap in the field by providing moderate-contrast phantom data with realistic geometries and partial boundary coverage, which is valuable for testing and comparing EIT reconstruction algorithms. The inclusion of both 2D and 3D data, multiple current patterns, and reference saline-only measurements enhances its utility. The paper's main contribution is the dataset itself, not a new method or result.

Pith's reading of the paper (a machine's summary, not the authors')

This paper presents an open-access data archive of Electrical Impedance Tomography (EIT) measurements from phantoms with moderate conductivity contrasts, designed to fill a gap in the field where most publicly available datasets use nearly perfect conductors or insulators. The archive includes 2D data from cylindrical and chest-shaped tanks with agar targets of known conductivities (0.61–3.23 mS/cm in 1.35 mS/cm saline) and 3D data from a head-shaped phantom with conductive plastic cylinders (4.70 mS/cm in 4.00 mS/cm saline) and a resistive skull. Alongside the measurements, the authors provide computational meshes, precomputed Jacobian matrices, example reconstruction code, and photographs of every setup, enabling independent benchmarking of both absolute and time-difference EIT reconstruction algorithms on realistic, moderate-contrast targets.

What carries the argument. The data archive itself is the central object: a set of .mat files containing current injection matrices, single-ended voltage measurements from all electrodes (including current-injecting electrodes), computational meshes (forward and inverse), and precomputed Jacobian matrices. The key property that carries the argument is the use of finite, approximately known moderate conductivity contrasts—agar targets at 0.61, 0.67, 3.05, and 3.23 mS/cm in 1.35 mS/cm saline for 2D setups, and conductive plastic cylinders at 4.70 mS/cm in 4.00 mS/cm saline for the 3D head phantom—which are measured in a conductivity cell before the experiments. The data format supports both absolute (static) and time-difference imaging, and the inclusion of current-injecting electrode voltages enables methods like CGO-based reconstruction that require the Neumann-to-Dirichlet map approximation.

Core claim. The paper's central contribution is a curated, openly archived EIT dataset that uses agar targets with measured moderate conductivities rather than the metal or insulating plastic targets common in prior public archives. This allows algorithm developers to test reconstruction methods on targets whose conductivity contrast (approximately 2:1 to 4:1 relative to the saline background) is closer to biomedical tissue contrasts than the extreme contrasts of perfect conductors or insulators. The archive covers both 2D cross-sectional setups (cylinder and chest phantoms) and a 3D partial-boundary head phantom with a resistive skull, all with documented measurement protocols, multiple current patterns, and repeat measurements. If the archive is used as intended, an independent group can reproduce the sample reconstructions and compare their own algorithms on the same data, which have already been used in several published studies.

The premise it rests on. The load-bearing premise is that the conductivity values measured once in a conductivity cell for each agar target and conductive plastic cylinder remain representative of the targets during the entire EIT measurement session, without accounting for temperature drift, measurement uncertainty, or possible differences between the cell and the tank environment.

If the paper is right (Pith's inferences, not the authors')

Researchers can directly compare linear, nonlinear, and deep-learning-based EIT reconstruction algorithms on a common benchmark with moderate-contrast targets, reducing the reproducibility gap caused by proprietary or extreme-contrast datasets.

The 3D head phantom data with a resistive skull enable testing of stroke-monitoring algorithms on realistic partial-boundary measurements, potentially accelerating translation of EIT to clinical intracranial hemorrhage monitoring.

The provided meshes, Jacobians, and example code lower the barrier for entering EIT research, allowing groups without access to measurement hardware to develop and validate reconstruction methods.

The archive supports both absolute and time-difference imaging, so it can serve as a testbed for algorithm robustness to the choice of reference frame and to the presence of contact impedance effects on current-injecting electrodes.

Because the archive includes multiple current patterns (adjacent, skip-1, skip-2, skip-3, all-against-one) and repeat measurements, it could be used to systematically study the effect of current pattern choice on reconstruction quality for moderate-contrast targets, a question the paper does not explicitly address.

The lack of reported uncertainty or temperature control for the conductivity measurements means that users who require precise ground truth for quantitative validation (e.g., for Bayesian inversion with known priors) may need to supplement the archive with their own conductivity cell calibrations; the archive is best suited for algorithmic comparison rather than metrological validation.

The vertical non-uniformity of the 2D setups (electrodes 7 cm high in a cylinder of 14 cm radius) creates a known mismatch between the 3D physics and the 2D reconstruction model; the archive could be used to quantify the resulting approximation error, which the authors note but do not quantify.

Check this review

Pith stored no record of this review's passes and took no fingerprint of it. Reviews made by code from 13 Nov 2026 on carry both.

How often reviews like this one are right

Reviews like this one: written by deepseek-v4-flash, where what was read was not recorded.

How often objections from reviews like this one are right: not yet measured

Every value above, in the review's public record (JSON)

Reviews by people

This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.

Request a human review

How Pith checks papers · The full audit of this paper

The paper

Abstract

This document describes open Electrical Impedance Tomography (EIT) measurement data that were collected from biomedical imaging inspired phantoms with moderate conductivity contrasts. The data were collected with the KIT4 and KIT5 EIT measurement devices at the University of Eastern Finland, and are intended for EIT image reconstruction algorithm development and testing. The data from thoracic imaging inspired cylindrical and chest-shaped phantoms are suitable for 2D algorithms, and the data from an intracerebral hemorrhage imaging inspired head-shaped phantom with resistive skull are suitable for 3D algorithms. In addition to the measurement data, computational meshes and example reconstruction codes are provided and sample reconstructions included.

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. No sentence in this record was written by a model.

abstract compiled · 2026-11-14

The authors' main claim, as the classifier picked it out 0.57

“The data from thoracic imaging inspired cylindrical and chest-shaped phantoms are suitable for 2D algorithms, and the data from an intracerebral hemorrhage imaging inspired head-shaped phantom with resistive skull are suitable for 3D algorithms.”

How the main claim is stated: asserted 0.94

What the abstract itself states, as probabilities: claims first 0.03 · states limitation 0.21 · human participants 0.05 · controlled comparison 0.02 · randomized assignment 0.01 · states quantitative result 0.04 · compares with prior results 0.04 · mentions code or data availability 0.98

The abstract, sentence by sentence

This table labels the abstract only (arXiv listing abstract v1). The number is the probability the classifier gave the label.

SentenceRoleThe authors' own workStates the main conclusionStrength
This document describes open Electrical Impedance Tomography (EIT) measurement data that were collected from biomedical imaging inspired phantoms with moderate conductivity contrasts. contribution 0.99 0.83 0.12 asserted 1.00
The data were collected with the KIT4 and KIT5 EIT measurement devices at the University of Eastern Finland, and are intended for EIT image reconstruction algorithm development and testing. method 0.97 0.87 0.07 asserted 1.00
The data from thoracic imaging inspired cylindrical and chest-shaped phantoms are suitable for 2D algorithms, and the data from an intracerebral hemorrhage imaging inspired head-shaped phantom with resistive skull are suitable for 3D algorithms. contribution 0.52 0.77 0.11 asserted 0.94
In addition to the measurement data, computational meshes and example reconstruction codes are provided and sample reconstructions included. contribution 0.99 0.81 0.07 asserted 1.00

What kind of paper the classifier reports this is: dataset or benchmark 1.00 · evidence from observation or experiment 0.95

What was examined, and what was not

Sources read

  • arXiv listing abstract v1

Not examined

  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers

Levels run

  • P abstract

Not run

  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

How the labels were made

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-11-14: 1 request, 3,679 input tokens.

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. This record as JSON

Figures (6)
Figure 1 of the paper
Figure 1. Agar phantom measurement setups for the cylinder phantom (top row) and the chest phantom (bottom row). in all measurement setups, allowing for reasonable reconstructions using 2D computational geometries. AgarChestPhantom The data for the AgarChestPhantom was taken using a tank mimicking a thoracic cross￾section and used agar targets in a saline bath of 1.35 mS/cm. The conductivities of the targets were measured at 0.61 mS/cm for the large white agar pieces simulating lungs and 3.23 mS/cm for the pink agar pieces simulating the heart and the fluid-filled part of the lung
Figure 2 of the paper
Figure 2. Head phantom measurement setups separated based on the loca￾tion of the conductive cylinder. 2.3. EIT Devices: KIT4 and KIT5. The agar phantom data were measured with the KIT4 (Kuopio Impedance Tomography 4) device [21] designed for use in industrial envi￾ronments for monitoring of industrial processes. The device can use up to 80 measurement channels and up to 16 current injection channels. The current injection frequency range is 1–120 kHz and the amplitude range is 0–5 mA. The KIT4 measurement device is shown in
Figure 3 of the paper
Figure 3. The KIT4 measurement device is shown on the left, and the KIT5 on the right. 3. Data Records The data were uploaded to Zenodo (https://zenodo.org) and are available at [23]. The data for each tank is stored in its own zip file for the user’s convenience. Each zip con￾tains the current and voltage data measurements as well as the corresponding computa￾tional meshes and Jacobian matrices for computing sample reconstructions. Sample lin￾ear difference reconstruction code is provided for the agar phantom setups. The data are stored in Matlab version 7.3 file format. The repository…
Figure 4 of the paper
Figure 4. Sample linear difference imaging reconstructions (left) and smoothness regularized absolute reconstructions (right) using the AgarChestPhantom data
Figure 5 of the paper
Figure 5. Sample linear difference imaging reconstructions using a growth case from the HeadPhantom data. A 3D visualization is displayed as well as a 2D slice of that reconstruction
Figure 6 of the paper
Figure 6. Computational meshes with (top row) and without (middle row) electrodes for the three different phantoms (columns) as well as their dimen￾sions (bottom row). 7. Data Availability Experimental data collected for 2D cross-sectional EIT imaging and 3D EIT imaging are available via Zenodo at [23]. The AgarCylinderPhantom and AgarChestPhantom data are suitable for use with 2D cross-sectional imaging algorithms and use moderate contrast agar targets with known conductivity values (0.61 mS/cm, 0.67 mS/cm, 3.05 mS/cm, and 3.23 mS/cm) placed in a saline bath of 1.35 mS/cm inside their…

Figures are the authors', reproduced from the paper.

References (27)
  1. [20] A. Hauptmann, V. Kolehmainen, N. M. Mach, T. Savolainen, A. Sepp¨ anen, and S. Siltanen. Open 2D electrical impedance tomography data archive. arXiv preprint arXiv:1704.01178, 2017 on PithPith's reviewarXiv
  2. [15] S. J. Hamilton, P. A. Muller, D. Isaacson, V. Kolehmainen, J. Newell, O. Rajabi Shishvan, G. Saulnier, and J. Toivanen. Fast absolute 3D CGO-based electrical impedance tomography on experimental tank data.Physiological Measurement, 43(12):124001, 2022.https://doi.org/10.1088/1... on Pithdoi:10.1088/1361-6579/
  3. [1] M. Cheney, D. Isaacson, and J. C. Newell. Electrical impedance tomography.SIAM Review, 41(1):85– 101, 1999.https://doi.org/10.1137/S0036144598333613 on Pithdoi:10.1137/s0036144598333613
  4. [2] J. L. Mueller and S. Siltanen.Linear and Nonlinear Inverse Problems with Practical Applications. SIAM, 2012 on Pith
  5. [3] D. S. Holder, editor.Electrical Impedance Tomography: Methods, History and Applications. IOP Pub- lishing, 2005 on Pith
  6. [4] L. Borcea. Electrical impedance tomography.Inverse Problems, 18(6):R99–R136, 2002.https://doi. org/10.1088/0266-5611/18/6/201 on Pithdoi:10.1088/0266-5611/18/6/201
  7. [5] A. Denker, F. Margotti, J. Ning, K. Knudsen, D. Nganyu Tanyu, B. Jin, A. Hauptmann, and P. Maass. Chapter 3 - deep learning based reconstruction methods for electrical impedance tomog- raphy. In A. Hauptmann, B. Jin, and C.-B. Sch¨ onlieb, editors,Machine Learning Solutions fo... on Pithdoi:10.1016/bs.hna.2025.09.003
  8. [6] D. Nganyu Tanyu, J. Ning, A. Hauptmann, B. Jin, and P. Maass. Electrical impedance tomog- raphy: A fair comparative study on deep learning and analytic-based approaches. arXiv preprint arXiv:2310.18636, 2023 on PithPith's reviewarXiv
  9. [7] S. J. Hamilton, A. H¨ anninen, A. Hauptmann, and V. Kolehmainen. Beltrami-net: domain-independent deep D-bar learning for absolute imaging with electrical impedance tomography (a-EIT).Physiological Measurement, 40(7):074002, 2019.https://doi.org/10.1088/1361-6579/ab21b2 on Pithdoi:10.1088/1361-6579/ab21b2
  10. [8] J. P. Agnelli, V. Kolehmainen, M. J. Lassas, P. Ola, and S. Siltanen. Simultaneous reconstruction of conductivity, boundary shape, and contact impedances in electrical impedance tomography.SIAM Journal on Imaging Sciences, 14(4):1407–1438, 2021.https://doi.org/10.1137/21M1407975 on Pithdoi:10.1137/21m1407975
  11. [9] W. Herzberg, D. B. Rowe, A. Hauptmann, and S. J. Hamilton. Graph convolutional networks for model-based learning in nonlinear inverse problems.IEEE Transactions on Computational Imaging, 7:1341–1353, 2021.https://doi.org/10.1109/TCI.2021.3132190. 2D/3D UEF EIT DATA ARCHIVE 11 on Pithdoi:10.1109/tci.2021.3132190
  12. [10] W. Herzberg, A. Hauptmann, and S. J. Hamilton. Domain independent post-processing with graph U-nets: applications to electrical impedance tomographic imaging.Physiological Measurement, 44(12):125008, 2023.https://doi.org/10.1088/1361-6579/ad0b3d on Pithdoi:10.1088/1361-6579/ad0b3d
  13. [11] W. Herzberg.Graph Neural Networks for Inverse Problems with Flexible Meshes. PhD thesis, Marquette University, Milwaukee, Wisconsin, USA, 2022.https://epublications.marquette.edu/ dissertations_mu/2053 on Pith
  14. [12] J. Avery, K. Aristovich, B. Low, and D. Holder. Reproducible 3D printed head tanks for electrical impedance tomography with realistic shape and conductivity distribution.Physiological Measurement, 38(6):1116–1131, 2017.https://doi.org/10.1088/1361-6579/aa6586 on Pithdoi:10.1088/1361-6579/aa6586
  15. [13] J. Toivanen, V. Kolehmainen, A. Paldanius, A. H¨ anninen, A. Hauptmann, and S. J. Hamilton. Graph convolutional networks enable fast hemorrhagic stroke monitoring with electrical impedance tomogra- phy.IEEE Transactions on Biomedical Engineering, 73(2):851–861, 2026.https://do... on Pith
  16. [14] S. J. Hamilton, P. A. Muller, V. Kolehmainen, and J. Toivanen. Fast 3D partial boundary data EIT reconstructions using direct inversion CGO-based methods.IEEE Transactions on Biomedical Engi- neering, 73(7):2435–2446, 2026.https://doi.org/10.1109/TBME.2025.3633560 on Pithdoi:10.1109/tbme.2025.3633560
  17. [16] D. Isaacson, J. L. Mueller, J. C. Newell, and S. Siltanen. Imaging cardiac activity by the D-bar method for electrical impedance tomography.Physiological Measurement, 27:S43–S50, 2006 on Pith
  18. [17] A. I. Nachman. Global uniqueness for a two-dimensional inverse boundary value problem.Annals of Mathematics, 143:71–96, 1996 on Pith
  19. [18] S. J. Hamilton, D. Isaacson, V. Kolehmainen, P. A. Muller, J. Toivanen, and P. F. Bray. 3D Electrical Impedance Tomography reconstructions from simulated electrode data using direct inversiont exp and Calder´ on methods.Inverse Problems and Imaging, 15(5):1135–1169, 2021.https... on Pith
  20. [19] A. I. Nachman. Reconstructions from boundary measurements.Annals of Mathematics, 128:531–576, 1988 on Pith
show 7 more
  1. [21] J. Kourunen, T. Savolainen, A. Lehikoinen, M. Vauhkonen, and L. M. Heikkinen. Suitability of a PXI platform for an electrical impedance tomography system.Measurement Science and Technology, 20(1):015503, 2008.https://doi.org/10.1088/0957-0233/20/1/015503 on Pithdoi:10.1088/0957-0233/20/1/015503
  2. [22] J. Toivanen, A. H¨ anninen, T. Savolainen, N. Forss, and V. Kolehmainen. Monitoring hemorrhagic strokes using EIT. InBioimpedance and Spectroscopy, pages 271–298. Elsevier, 2021.https://doi. org/10.1016/B978-0-12-818614-5.00007-2 on Pithdoi:10.1016/b978-0-12-818614-5.00007-2
  3. [23] S. J. Hamilton, A. Hauptmann, A. H¨ anninen, V. Kolehmainen, J. Niskanen, T. Savolainen, and J. Toivanen. Open Electrical Impedance Tomography data from phantoms with moderate conduc- tivity contrast for 2D and 3D reconstruction (KIT4 and KIT5). Zenodo, Version 1.0, 2026.https... on Pithdoi:10.5281/zenodo.24120439
  4. [24] A. Nissinen, V. Kolehmainen, and J. P. Kaipio. Compensation of modelling errors due to unknown do- main boundary in electrical impedance tomography.IEEE Transactions on Medical Imaging, 30(2):231– 242, 2011 on Pith
  5. [25] A. Nissinen, V. Kolehmainen, and J. P. Kaipio. Reconstruction of domain boundary and conductivity in electrical impedance tomography using the approximation error approach.International Journal for Uncertainty Quantification, 1(3):203–222, 2011 on Pith
  6. [26] V. Kolehmainen, M. J. Ehrhardt, and S. R. Arridge. Incorporating structural prior information and sparsity into EIT using parallel level sets.Inverse Problems and Imaging, 13(2):285–307, 2019 on Pith
  7. [27] A. J¨ a¨ askel¨ ainen, A. Vavilov, J. Toivanen, A. H¨ anninen, V. Kolehmainen, and N. Hyv¨ onen. Local elec- trical impedance tomography via projections.Inverse Problems, 2026. arXiv:2601.14198 on PitharXiv

Challenges and discussion

Disagree with the review, or know something it missed? Leave a comment, or challenge this review formally: a challenge becomes a permanent entry in this paper's record, and its decision is published here.

No comments yet.

Sign in to comment