Pith. sign in

REVIEW 3 major objections 6 minor 57 references

Dictionary-Based Reconstruction of Spatio-Temporal 3D Magnetic Field Images from Quantum Diamond Microscope

T0 review · 3 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper claims that a LASSO-based reconstruction using experimentally measured single-coil magnetic field maps enables dynamic three-dimensional localization and tracking of current sources in NV widefield microscopy at micrometre and…

desk verdict A careful experimental demo of dictionary-based LASSO reconstruction for NV microscopy, but the abstract overclaims generalized 3D localization when the evidence is in-sample template matching. read the letter →

arxiv 2506.05491 v1 pith:YMYCIV6V submitted 2025-06-05 physics.ins-det physics.comp-phquant-ph

Anuj Bathla , Madhur Parashar , Matthew Markham , Ajit Rajwade , Kasturi Saha This is my paper
classification physics.ins-detphysics.comp-phquant-ph
keywords nitrogen-vacancydiamondmicroscopymagneticfieldimagingcurrentsourcelocalizationLASSOsparseregressiondictionary-basedreconstruction3Dtemporalmicro-coilarray
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that a quantum diamond microscope—an NV-center widefield magnetometer—can image not just static two-dimensional current patterns but full three-dimensional, time-varying current sources. The authors' route is to treat each possible source's measured magnetic field map as a dictionary atom and to recover which sources are active by solving a sparsity-promoting LASSO problem frame by frame. They validate the claim on a custom three-layer array of eight micro-coils that mimics neuronal spiking, and the static reconstructions achieve high correlation with ground truth while the temporal ones preserve the main firing structure. If the claim holds, the same microscope hardware could map currents inside circuits or neural tissue without contact, in volumes rather than planes.

What carries the argument

The load-bearing object is the dictionary matrix $A$, whose columns are flattened, mean-centered, unit-variance-normalized magnetic field maps of each individual micro-coil, acquired experimentally under steady-state current. The reconstruction solves $\min_x \tfrac{1}{2}\|Ax-y\|_2^2 + \lambda\|x\|_1$ for each test map $y$ (static configuration or a single temporal frame), with $\lambda$ tuned to balance sparsity and stability. The dictionary encodes the depth-dependent Biot–Savart blur as a spatial prior, and the $L^1$ penalty is what separates overlapping sources that a least-squares pseudo-inverse would smear together.

What would settle it

Take a known configuration of two simultaneously driven coils and compare the measured field image against the pixelwise sum of the two individually measured single-coil basis images; any residual above the noise floor falsifies the linear superposition assumption. For the temporal claim, drive one coil with a current step that is faster than the 208 fps acquisition and check whether the reconstructed coefficient tracks the true on/off timing; if it does not, the static-dictionary assumption about transients fails.

Watch

Extended reading notes

Core claim

The central discovery claimed is that a dictionary of experimentally measured single-coil magnetic field maps, used as columns of the forward matrix $A$ in the model $y = Ax + \epsilon$, turns the ill-posed 3D magnetic inversion problem into a tractable sparse regression. Because current sources are sparse in time—few coils active at once, like neurons spiking—the $L^1$ penalty in LASSO selects the correct basis maps while suppressing noise and cross-talk between layers. The paper reports that this recovers static multi-coil configurations with Pearson correlation above 0.95 and structural similarity 0.57–0.82, and that frame-by-frame reconstruction of a 300 ms action-potential-like stimulus across all eight coils tracks activations with millisecond resolution, albeit with lower fidelity (SSIM up to 0.31). This is presented as the first demonstration of dynamic 3D current-source imaging with NV widefield microscopy using measured spatial priors.

Load-bearing premise

The reconstruction assumes every measured field pattern is exactly a linear combination of the same steady-state, single-coil field maps that make up the dictionary, so any transient or interaction effect not captured in those maps breaks the model.

Editorial extensions

If this is right

  • A single widefield magnetic image of a multilayer stack can be decomposed into the individual active sources without knowing the circuit layout in advance.
  • The same static dictionary reconstructs every frame of a dynamic recording, so no per-timestep calibration is needed as long as the source positions are fixed.
  • Sources in deeper layers, though more blurred, remain identifiable, so the method extends NV microscopy from surfaces to volumes.
  • The method keeps working as the number of simultaneously active sources grows from two to seven, indicating scalability to denser configurations.
  • Against pseudo-inverse reconstruction on simulated multilayer and bent-wire geometries, LASSO suppresses spurious activations of inactive sources even under high basis coherence.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because the dictionary is built from measured single-coil responses, applying the method to real nervous tissue would require a different route to the dictionary—simulated or learned forward models—since individual neurons cannot be calibrated one by one; the paper acknowledges this when it says complete dictionaries are impractical in some scenarios.
  • A natural testbed extension is to compare the static dictionary against a time-resolved or adaptive dictionary on the same frames; the paper's own numbers suggest temporal SSIM could rise from the current 0.31 ceiling toward static levels if transient effects are modeled.
  • The recovered LASSO coefficients are calibrated against the known 800 $\mu$A drive current, so coefficient magnitudes across frames can be read as quantitative current estimates; checking them against the true drive waveform would certify the method as a current metrology tool.
  • The main failure mode named inside the paper—strong coherence between basis fields from different layers—predicts that reconstruction error should track the condition number of $A$; reporting that number alongside each dataset would give users a built-in reliability gauge.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper describes a nitrogen-vacancy diamond wide-field magnetic microscope combined with a LASSO-based reconstruction algorithm that uses experimentally measured single-coil magnetic field maps as a dictionary. Static reconstructions are demonstrated on a purpose-built three-layer micro-coil array containing eight functional coils, using all 256 on/off combinations as test maps, with reported SSIM 0.57–0.82, RRMSE 0.20–0.29, and correlation >0.95. Temporal reconstructions of action-potential-like waveforms at 208 frames per second yield SSIM up to 0.31, RRMSE 0.36–0.54, and correlation between 0.29 and 0.76. The authors claim that this enables robust identification of active current sources across space and time and advances dynamic 3D current imaging.

Significance. If the method genuinely localizes arbitrary 3D current sources, it would be valuable for applications such as neuronal imaging and semiconductor diagnostics. The experimental apparatus is well described, the micro-coil testbed is carefully fabricated, and the static results provide a clean demonstration that magnetic fields from simultaneously active coils superpose linearly and can be decomposed into the measured dictionary. The temporal experiment is a useful step toward dynamic imaging. However, as presented, the evidence supports template selection among the eight known coils rather than generalized 3D source localization; the temporal metrics are moderate and are not scored against ground-truth activation coefficients. The paper's own stated limitations about the static dictionary and ill-conditioned inversion directly qualify the abstract's strong claims.

major comments (3)
  1. [Static Magnetic Field Imaging and Reconstruction; Methods (Reconstruction Algorithm Using Pseudo-Inverse and LASSO)] The validation is in-sample. The dictionary A is built from the measured fields of the same eight coils whose on/off combinations form all 256 test maps, so every test map lies in the span of the dictionary columns by construction (up to noise). The reported SSIM, RRMSE, and correlation therefore quantify how well a known linear combination can be fitted, not whether a source at an arbitrary position or depth outside the dictionary can be recovered. The abstract's claim of 'robust identification of active current sources across space and time' is stronger than what the static experiments demonstrate. Please add a validation case where the test source is not a dictionary element (for example, an intermediate in-plane position or an intermediate depth, using FEM-simulated data or a leave-one-coil-out analysis), or rephrase the claims to describe template-based decomposition of known sources.
  2. [Temporal Imaging and Reconstruction] The temporal metrics are modest and not assessed against ground truth. For the high-sparsity case, SSIM values are 0.16–0.22 and correlations can be as low as 0.29, yet the text states that reconstructions 'closely resemble experimental magnetic field frames' and 'effectively track individual coil activations.' No localization error (for example, center-of-mass error or layer-assignment error) is reported, and the reconstructed coefficients in Figures 3 and 4 are not scored against the known activation waveforms. In addition, the regularization parameter is tuned per case (λ=0.3–0.4 for high sparsity and λ=0.2–0.25 for low sparsity) without a principled selection rule; if λ is chosen to optimize the displayed metrics, the reported values are optimistic. Please report coefficient-recovery errors against ground truth and provide uncertainty estimates across frames or repeated acquisitions.
  3. [Outlook and Temporal Imaging and Reconstruction] The manuscript explicitly acknowledges that the static dictionary 'does not account for transient variations' and that strong coherence between basis fields causes an ill-conditioned inverse problem. These limitations directly constrain the central claim of dynamic 3D source localization and should be reconciled with the abstract's strength. The FEM benchmark involving multilayer arrays and bent microwires is mentioned only qualitatively in the Outlook and is not quantified in the main text; since it is the only evidence toward generalization beyond the eight-coil array, it should either be presented with quantitative metrics in the main text or clearly labeled as preliminary. Please temper the abstract and conclusion so that the stated claims match the demonstrated evidence.
minor comments (6)
  1. [Methods, Experimental Apparatus] The sentence 'The experimental setup is described in Figure 1(a) illustrates the configuration' is grammatically incomplete; revise to 'The experimental setup is shown in Figure 1(a), which illustrates the configuration.'
  2. [Methods, Reconstruction Algorithm Using Pseudo-Inverse and LASSO] The phrase 'if (𝑨has full coloumn Rank)' contains typos; it should read 'if A has full column rank.'
  3. [Table 1] No uncertainty estimates are provided for the static reconstruction metrics; if these values are obtained from multiple configurations, report standard deviations or confidence intervals.
  4. [Outlook] The term 'true 3D dynamic reconstruction' is not explicitly defined; clarify whether it refers to source localization in three spatial dimensions plus time or to volumetric magnetic field reconstruction.
  5. [Static Magnetic Field Imaging and Reconstruction] The main text compares LASSO only with itself in the static experiments; since the Outlook mentions a pseudo-inverse comparison, consider including a pseudo-inverse baseline in the main text (or in Table 1) to substantiate the claimed advantage.
  6. [References] Reference [42] is cited for matching pursuit, but the context could be clearer about why matching pursuit is mentioned and how it relates to the ill-conditioning discussion.

Circularity Check

2 steps flagged · score 6.0 of 10

In-sample validation: dictionary built from the same eight coils used to generate all static and temporal test maps, so reported SSIM/RRMSE measure fit within the span of known templates, not generalized 3D source localization.

  1. fitted input called prediction [Results: Static Magnetic Field Imaging and Reconstruction; Methods: Reconstruction Algorithm Using Pseudo-Inverse and LASSO]
    "To accurately localize the current sources in three-dimensions, we employed LASSO-based sparse regression, leveraging a predefined dictionary of basis maps obtained from individual micro-coil field measurements. ... We investigated 256 distinct magnetic field configurations, corresponding to all(28)possible on/off combinations of the eight working micro-coils across the three layers. ... A∈R M×N dictionary matrix, where each column is a flattened basis map."

    The dictionary columns are the measured field maps of the eight functional coils, and every static test map is a combination of exactly those same eight coils. Under the paper's linear model y=Ax+ε, the test map therefore lies in the span of the dictionary columns up to noise. LASSO returns coefficients x that minimize ||Ax−y||2^2+λ||x||1, and the 'reconstructed' image is Ax by definition. Comparing Ax with y via SSIM, RRMSE, and correlation quantifies in-sample fit error, not the ability to recover an unseen source at an unlisted position. No coefficient-recovery error against the known on/off ground truth is reported, so 'robust identification of active current sources' reduces to selecting among already-known dictionary templates.

  2. fitted input called prediction [Results: Temporal Imaging and Reconstruction; Methods: Temporal Field Reconstruction]
    "For both scenarios, four representative time frames were selected during periods of micro-coil activity, and frame-wise reconstruction was performed using the static dictionary. ... This dictionary, constructed from steady-state magnetic field maps of individual micro-coils, does not account for transient variations that may occur during rapid current switching or simultaneous activations. ... The time-dependent coefficients, x_t, are used to linearly combine basis maps, reconstructing the test map for each time point."

    The temporal frames are magnetic fields from the same eight micro-coils driven by a known AP-like waveform, while the dictionary contains those same coils' steady-state maps normalized to 800 μA. Frame-wise 'reconstruction' is therefore a sparse fit within the same template span. The known waveform matrix provides ground-truth coefficients, but the paper never scores x_t against it; all reported temporal metrics compare y_t to A x_t. Since the dictionary explicitly excludes transient effects, even the in-span fit is weak (SSIM up to 0.31, correlation as low as 0.29), so claims of 'track[ing] individual micro-coil activations' rest on comparing the fit to its own inputs rather than on validated source-localization accuracy.

full rationale

The central derivation chain is the linear forward model y=Ax+ε with A assembled from experimentally measured single-coil basis maps. All static test images are the 2^8 on/off combinations of the same eight coils that define A, and all temporal test frames come from those same coils; no held-out coil, arbitrary source position, or source outside the dictionary is tested in the main text. Because the reconstructed image is by definition Ax, the reported SSIM (0.57–0.82), RRMSE (0.20–0.29), and correlation (>0.95) quantify how well the fitted coefficients reproduce data inside the training span, not how well the method generalizes to unseen current-source geometries. The temporal validation is weaker still: the static dictionary is acknowledged not to capture transient variations, yet the paper reports only field-map fit metrics and never compares the recovered coefficient vectors with the known drive waveform. The FEM benchmark mentioned in the Outlook could in principle supply out-of-dictionary evidence, but it is not quantified in the main text and so cannot break the in-sample character of the reported claims. There is no load-bearing uniqueness-theorem self-citation: references [27] and [42] are earlier work by the authors but only support the lock-in instrumentation and the 2D context, not the dictionary-construction step. Score 6 reflects the partial circularity: the central validation metrics reduce by construction to in-span fits, while the LASSO optimization itself is a genuine numerical problem and the low temporal scores show the result is not entirely forced.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The central claim rests on empirically measured dictionary maps and a tuned regularization parameter, not on a derivation. The model assumes linear superposition and dictionary completeness, which are the main untested premises.

free parameters (1)
  • lambda (LASSO regularization parameter) = 0.2 to 0.4 depending on case
    Tuned per reconstruction case (static and temporal) to balance sparsity and stability; reported metrics depend on this choice.
assumptions (3)
  • domain assumption Magnetic fields of simultaneously active coils are the linear superposition of individual coil fields (y=Ax).
    Used in the forward model in Methods; not independently validated for simultaneous activation or transient switching.
  • domain assumption The static dictionary of eight measured single-coil maps is complete and stationary for all test configurations, including temporal frames.
    Used in temporal reconstruction; the paper states the same static basis maps were used 'without introducing time-resolved dictionaries' and acknowledges transient variations are not captured.
  • domain assumption Only a small subset of sources is active at any time (sparsity).
    Motivates the L1 regularization; holds by design of the test waveforms but not for arbitrary current distributions.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Dictionary-Based Reconstruction of Spatio-Temporal 3D Magnetic Field Images from Quantum Diamond Microscope." pith.science (2026). https://pith.science/paper/YMYCIV6V

@misc{pith2026250605491,
  author       = {Pith},
  title        = {Pith review of: Dictionary-Based Reconstruction of Spatio-Temporal 3D Magnetic Field Images from Quantum Diamond Microscope},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YMYCIV6V}},
  note         = {Machine review of arXiv:2506.05491}
}
read the original abstract

Three-dimensional magnetic imaging with high spatio-temporal resolution is critical for probing current paths in various systems, from biosensing to microelectronics. Conventional 2D Fourier-based current source localization methods are ill-posed in multilayer or dynamic systems due to signal overlap and noise. In this work, we demonstrate an innovative nitrogen-vacancy (NV) center-based wide-field magnetic microscopy technique for dynamic three-dimensional imaging and localization of current sources. Using custom-fabricated multilayer micro-coil platform to emulate localized, time-varying currents similar to neuronal activity, we acquire magnetic field maps with micrometre-scale spatial and millisecond-scale temporal resolution using per-pixel lock-in-based detection. Source localization and image reconstruction are achieved using a Least Absolute Shrinkage and Selection Operator (LASSO)-based reconstruction framework that incorporates experimentally measured basis maps as spatial priors. Our method enables robust identification of active current sources across space and time, and significantly advances the accuracy of dynamic 3D current imaging and NV-based magnetometry for complex systems.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

57 extracted references · 47 canonical work pages

  1. [1]

    The basis maps are loaded from experimental datasets, where each basis map corresponds to the field contribution of an individual micro-coil

    work page

  2. [2]

    The test map is computed as the difference between the magnetic field maps in the I-on and I-off states for a given coil combination

    work page

  3. [3]

    Both the basis maps and test maps are mean-centered and standardized to have unit variance to ensure numerical stability

    work page

  4. [4]

    For pseudo-inverse: coefficients are estimated using the pseudo-inverse of the basis matrix

    work page

  5. [5]

    For LASSO: The regression model is initialized with a chosen value𝝀(typically tuned) and the model is trained to find sparse coefficients𝒙by solving the optimization problem

    work page

  6. [6]

    The estimated coefficients are used to linearly combine the basis maps, forming the recon- structed test map

    work page

  7. [7]

    Temporal Field Reconstruction The temporal reconstruction follows the same steps as the static case but is applied to individual frames at different time points: 18

    The reconstruction is then rescaled back to the original scale of the experimental data. Temporal Field Reconstruction The temporal reconstruction follows the same steps as the static case but is applied to individual frames at different time points: 18

    work page

  8. [8]

    Experimental time-resolved data from 300 frames is loaded, using the same dictionary of basis maps for all time points

    work page

Show all 57 references
  1. [9]

    Each frame at a specific time point is treated as an independent test map𝒚 𝒕 and baseline correction is applied by subtracting the mean field from frames with no current waveform applied

  2. [10]

    Normalization and saturation correction are performed to mitigate intensity variations and oversaturated pixels

  3. [11]

    Basis and test maps are mean-centered and standardized for numerical stability

  4. [12]

    LASSO regression is applied with fine-tuning of the regularization parameter to balance sparsity and stability

  5. [13]

    The time-dependent coefficients,𝒙 𝒕, are used to linearly combine basis maps, reconstructing the test map for each time point

  6. [14]

    Acknowledgments The authors are grateful to D

    The reconstructed map is then rescaled back to the original scale of the experimental data. Acknowledgments The authors are grateful to D. Shishir, Prof. Udayan Ganguly, Prof. Pradeep Sarin, Prof Kantimay Dasgupta and Prof. Swaroop Ganguly for their valuable insights on experi...

    2023

  7. [15]

    Tetienne,et al., Quantum imaging of current flow in graphene.Science Advances3(4), e1602429 (2017)

    J.-P. Tetienne,et al., Quantum imaging of current flow in graphene.Science Advances3(4), e1602429 (2017)

    2017

  8. [16]

    M. J. Turner,et al., Magnetic Field Fingerprinting of Integrated-Circuit Activity with a Quantum Diamond Microscope.Physical Review Applied14(1), 014097 (2020)

    2020

  9. [17]

    Kehayias,et al., Measurement and Simulation of the Magnetic Fields from a 555 Timer Integrated Circuit Using a Quantum Diamond Microscope and Finite-Element Analysis

    P. Kehayias,et al., Measurement and Simulation of the Magnetic Fields from a 555 Timer Integrated Circuit Using a Quantum Diamond Microscope and Finite-Element Analysis. Physical Review Applied17(1), 014021 (2022)

    2022

  10. [18]

    L. T. Hall,et al., High spatial and temporal resolution wide-field imaging of neuron activity using quantum NV-diamond.Scientific Reports2, 401 (2012)

    2012

  11. [19]

    Le Sage,et al., Optical magnetic imaging of living cells.Nature496, 486–489 (2013)

    D. Le Sage,et al., Optical magnetic imaging of living cells.Nature496, 486–489 (2013)

    2013

  12. [20]

    Schirhagl, K

    R. Schirhagl, K. Chang, M. Loretz, C. L. Degen, Nitrogen-Vacancy Centers in Diamond: Nanoscale Sensors for Physics and Biology.Annual Review of Physical Chemistry65, 83–105 (2014)

    2014

  13. [21]

    D. R. Glenn,et al., Single-cell magnetic imaging using a quantum diamond microscope. Nature Methods12(8), 736–738 (2015). 20

    2015

  14. [22]

    J. F. Barry,et al., Optical magnetic detection of single-neuron action potentials using quantum defects in diamond.Proceedings of the National Academy of Sciences113(49), 14133–14138 (2016), doi:10.1073/pnas.1601513113

    2016 doi

  15. [23]

    J. C. Price, R. Mesquita-Ribeiro, F. Dajas-Bailador, M. L. Mather, Widefield, Spatiotemporal Mapping of Spontaneous Activity of Mouse Cultured Neuronal Networks Using Quantum Diamond Sensors.Frontiers in Physics8, 255 (2020), doi:10.3389/fphy.2020.00255

    2020

  16. [24]

    Ariyaratne, D

    A. Ariyaratne, D. Bluvstein, B. A. Myers, A. C. B. Jayich, Nanoscale electrical conductivity imaging using a nitrogen-vacancy center in diamond.Nature Communications9(1), 2406 (2018), doi:10.1038/s41467-018-04853-4

    2018 doi

  17. [25]

    S. E. Lillie,et al., Imaging Graphene Field-Effect Transistors on Diamond Using Nitrogen- Vacancy Microscopy.Physical Review Applied12(2), 024018 (2019), doi:10.1103/ PhysRevApplied.12.024018

    2019

  18. [26]

    M. J. H. Ku,et al., Imaging viscous flow of the Dirac fluid in graphene.Nature583, 537–541 (2020), doi:10.1038/s41586-020-2507-2

    2020 doi

  19. [27]

    S. C. Scholten,et al., Imaging current paths in silicon photovoltaic devices with a quantum diamond microscope.Physical Review Applied18(1), 014041 (2022), doi: 10.1103/PhysRevApplied.18.014041

    2022 doi

  20. [28]

    Li,et al., Wide-field tomography imaging of a double circuit using NV center ensembles in a diamond.Opt

    Z. Li,et al., Wide-field tomography imaging of a double circuit using NV center ensembles in a diamond.Opt. Express30(22), 39877–39890 (2022), doi:10.1364/OE.469077,https: //opg.optica.org/oe/abstract.cfm?URI=oe-30-22-39877

    2022 doi

  21. [29]

    C. Zhong,et al., High Spatial Resolution 2D Imaging of Current Density and Pressure for Graphene Devices under High Pressure Using Nitrogen-Vacancy Centers in Diamond.Nano Letters24(16), 4993–5001 (2024)

    2024

  22. [30]

    J. M. Taylor,et al., High-sensitivity diamond magnetometer with nanoscale resolution.Nature Physics4(10), 810–816 (2008), doi:10.1038/nphys1075. 21

    2008 doi

  23. [31]

    Steinert,et al., High sensitivity magnetic imaging using an array of spins in diamond

    S. Steinert,et al., High sensitivity magnetic imaging using an array of spins in diamond. Review of Scientific Instruments81(4), 043705 (2010), doi:10.1063/1.3385689

    2010 doi

  24. [32]

    L. M. Pham,et al., Magnetic field imaging with nitrogen-vacancy ensembles.New Journal of Physics13(4), 045021 (2011), doi:10.1088/1367-2630/13/4/045021

    2011 doi

  25. [33]

    M. W. Doherty,et al., The nitrogen-vacancy colour centre in diamond.Physics Reports528(1), 1–45 (2013), doi:10.1016/j.physrep.2013.02.001

    2013 doi

  26. [34]

    Steinert,et al., Magnetic spin imaging under ambient conditions with sub-cellular resolu- tion.Nature Communications4, 1607 (2013), doi:10.1038/ncomms2604

    S. Steinert,et al., Magnetic spin imaging under ambient conditions with sub-cellular resolu- tion.Nature Communications4, 1607 (2013), doi:10.1038/ncomms2604

    2013 doi

  27. [35]

    Rondin,et al., Magnetometry with nitrogen-vacancy defects in diamond.Reports on Progress in Physics77(5), 056503 (2014), doi:10.1088/0034-4885/77/5/056503

    L. Rondin,et al., Magnetometry with nitrogen-vacancy defects in diamond.Reports on Progress in Physics77(5), 056503 (2014), doi:10.1088/0034-4885/77/5/056503

    2014 doi

  28. [36]

    L. A. Knauss,et al., Scanning SQUID Microscopy for Current Imaging.Microelectronics Reliability41(8), 1211–1229 (2001), doi:10.1016/S0026-2714(01)00067-7

    2001 doi

  29. [37]

    Wang,et al., Sensing of Arbitrary-Frequency Fields Using a Quantum Mixer.Phys

    G. Wang,et al., Sensing of Arbitrary-Frequency Fields Using a Quantum Mixer.Phys. Rev. X 12, 021061 (2022), doi:10.1103/PhysRevX.12.021061,https://link.aps.org/doi/10. 1103/PhysRevX.12.021061

    2022 doi

  30. [38]

    S. J. Karlson,et al., Quantum frequency mixing using an N-V diamond microscope.Physical Review Applied22(6), 064051 (2024)

    2024

  31. [39]

    E. V. Levine,et al., Principles and techniques of the quantum diamond microscope.Nanopho- tonics8(11), 1945–1973 (2019), doi:10.1515/nanoph-2019-0206

    2019 doi

  32. [40]

    S. C. Scholten,et al., Widefield quantum microscopy with nitrogen-vacancy centers in di- amond: Strengths, limitations, and prospects.Journal of Applied Physics130(15), 150902 (2021), doi:10.1063/5.0061364

    2021 doi

  33. [41]

    Parashar,et al., Sub-second temporal magnetic field microscopy using quantum defects in diamond.Scientific Reports12, 8743 (2022)

    M. Parashar,et al., Sub-second temporal magnetic field microscopy using quantum defects in diamond.Scientific Reports12, 8743 (2022). 22

    2022

  34. [42]

    J. L. Webb,et al., High-Speed Wide-Field Imaging of Microcircuitry Using Nitrogen Va- cancies in Diamond.Phys. Rev. Appl.17, 064051 (2022), doi:10.1103/PhysRevApplied.17. 064051,https://link.aps.org/doi/10.1103/PhysRevApplied.17.064051

    2022 doi

  35. [43]

    Casola, T

    F. Casola, T. Van Der Sar, A. Yacoby, Probing Condensed Matter Physics with Magnetometry Based on Nitrogen-Vacancy Centres in Diamond.Nature Reviews Materials3(1), 1–13 (2018), doi:10.1038/natrevmats.2017.88

    2018 doi

  36. [44]

    Shishir, K

    D. Shishir, K. Saha, Nitrogen vacancy centre-based diamond microscope for investigat- ing quantum materials.Bulletin of Materials Science44(4), 276 (2021), doi:10.1007/ s12034-021-02548-6,https://doi.org/10.1007/s12034-021-02548-6

    2021 doi

  37. [45]

    B. J. Roth, N. G. Sepulveda, J. P. Wikswo Jr, Using a magnetometer to image a two-dimensional current distribution.Journal of Applied Physics65(1), 361–372 (1989)

    1989

  38. [46]

    N. R. Reed,et al., Machine learning for improved current-density reconstruction from two- dimensional vector magnetic images.Phys. Rev. Appl.23, 034035 (2025), doi:10.1103/ PhysRevApplied.23.034035,https://link.aps.org/doi/10.1103/PhysRevApplied. 23.034035

    2025 doi

  39. [47]

    Zhang, W

    E. Zhang, W. Zhang, X. Jiang, X. Qin, Deep-neural-network-based magnetic-field acquisition method for quantum sensing with nitrogen-vacancy centers.Phys. Rev. A110, 052417 (2024), doi:10.1103/PhysRevA.110.052417,https://link.aps.org/doi/10.1103/PhysRevA. 110.052417

    2024 doi

  40. [48]

    Midha,et al., Optimized current-density reconstruction from wide-field quantum diamond magnetic field maps.Phys

    S. Midha,et al., Optimized current-density reconstruction from wide-field quantum diamond magnetic field maps.Phys. Rev. Appl.22, 014015 (2024), doi:10.1103/PhysRevApplied.22. 014015,https://link.aps.org/doi/10.1103/PhysRevApplied.22.014015

    2024 doi

  41. [49]

    Garsi,et al., Three-dimensional imaging of integrated-circuit activity using quantum defects in diamond.Phys

    M. Garsi,et al., Three-dimensional imaging of integrated-circuit activity using quantum defects in diamond.Phys. Rev. Appl.21, 014055 (2024), doi:10.1103/PhysRevApplied.21. 014055,https://link.aps.org/doi/10.1103/PhysRevApplied.21.014055

    2024 doi

  42. [50]

    Utsugi, 3-D inversion of magnetic data based on the L1–L2 norm regularization.Earth, Planets and Space71(73) (2019), doi:https://doi.org/10.1186/s40623-019-1052-4

    M. Utsugi, 3-D inversion of magnetic data based on the L1–L2 norm regularization.Earth, Planets and Space71(73) (2019), doi:https://doi.org/10.1186/s40623-019-1052-4. 23

    2019 doi

  43. [51]

    V. Kolehmainen,et al., Tikhonov regularization and prior information in electrical impedance tomography.IEEE Transactions on Medical Imaging13(2), 326–332 (1994), doi:10.1109/ 42.293895

    1994

  44. [52]

    S. S. Chen, D. L. Donoho, M. A. Saunders, Atomic Decomposition by Basis Pursuit.SIAM Review43(1), 129–159 (2001), doi:10.1137/S003614450037906X

    2001 doi

  45. [53]

    A. Nowodzinski,et al., Nitrogen-vacancy centers in diamond for current imaging at the redistributive layer level of integrated circuits.Microelectronics Reliability55(9-10), 1549– 1553 (2015)

    2015

  46. [54]

    Acheson, A

    K. Acheson, A. Kirrander, Robust Inversion of Time-Resolved Data via Forward-Optimization in a Trajectory Basis.Journal of Chemical Theory and Computation19(10), 2721–2734 (2023), doi:10.1021/acs.jctc.2c01113,https://doi.org/10.1021/acs.jctc.2c01113

    2023 doi

  47. [55]

    H. C. Davis,et al., Mapping the Microscale Origins of Magnetic Resonance Image Contrast with Subcellular Diamond Magnetometry.Nature Communications9(1), 131 (2018), doi: 10.1038/s41467-017-02471-7

    2018 doi

  48. [56]

    Parashar, K

    M. Parashar, K. Saha, S. Bandyopadhyay, Axon hillock currents enable single-neuron- resolved 3D reconstruction using diamond nitrogen-vacancy magnetometry.Communications Physics3(1), 174 (2020), doi:10.1038/s42005-020-00439-6,https://doi.org/10.1038/ s42005-020-00439-6

    2020 doi

  49. [57]

    Z. Wang, A. C. Bovik, Mean Squared Error: Love It or Leave It? A New Look at Signal Fidelity Measures.IEEE Signal Processing Magazine26(1), 98–117 (2009), doi:10.1109/ MSP.2008.930649. 24

    2009

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.