Pith. sign in

REVIEW 4 major objections 4 minor 29 references

Optimizing SPION Labeling for Single-Cell Magnetic Microscopy

T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper claims that a convolutional neural network trained only on simulated magnetic dipole data can reconstruct, from NV widefield magnetic images of single cells, the cell diameter, the cell-to-sensor distance, and the surface iron…

desk verdict Plausible CNN-based pipeline for per-cell SPION quantification from NV widefield images, but the headline numbers need error bars and the forward-model calibration needs a stronger independent check. read the letter →

arxiv 2505.20373 v2 pith:F6DFF57M submitted 2025-05-26 physics.bio-ph quant-ph

A. Pointner , D. Thalheim , S. Belasi , L. Heinen , C. Bonato , T. Luehmann , J. Meijer , R. Tietze
show 3 more authors
C. Alexiou R. Schneider-Stock R. Nagy
This is my paper · ORCID
classification physics.bio-phquant-ph
keywords quantumsensingNVcenterswidefieldmagnetometrysuperparamagneticironoxidenanoparticlesmagneticcelllabelingconvolutionalneuralnetworksingle-cellimagingmassquantification
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 convolutional neural network trained only on simulated magnetic dipole images can turn a nitrogen-vacancy (NV) widefield magnetic measurement of a single cell into three quantitative parameters: cell diameter, distance from the cell to the diamond sensor, and the mass of iron bound to the cell surface. The motivation is that raw magnetic field strength cannot be compared across cells because standoff distance and cell size vary; if the network's normalization works, those confounding variables are removed and labeling efficiency can be assessed cell by cell. Using this approach on SPION-labeled HT29 cells, the authors report that surface iron mass and magnetic field strength increase with SPION concentration and then saturate, and they identify 116.39 µg/ml as the practical optimum concentration, the half-maximum of the saturation curve. If correct, the method provides a quantitative, high-throughput way to characterize magnetically labeled cells for tracking and sensing applications.

What carries the argument

The load-bearing object is the simulation-to-experiment transfer pipeline: a forward model that places point magnetic dipoles on a spherical surface representing the cell, assigns each dipole a magnetic moment from the SPION batch's reported value (8.6 × 10⁻¹⁹ A m² at 40 mT, treated as saturated at 30 mT), adds noise and dipole overlap, and generates 32 × 32 pixel snippets of magnetic field. A convolutional neural network with residual connections, dilated convolutions, self-attention, and multi-head regression inverts these snippets into position, distance, diameter, iron mass, and a background term. The same forward model is then used to normalize predictions to a fixed reference cell, 12 µm diameter at 15 µm distance, which is what turns raw, standoff-dependent field values into a comparable labeling-efficiency metric.

What would settle it

Measure a phantom or a cell sample with an independently known number of SPIONs, for example by MP-AES or ICP-MS on the same cell suspension or by depositing a counted particle solution, at a known distance from the NV layer; run the CNN on its magnetic image and compare the predicted iron mass and distance to the known values, because a systematic offset beyond measurement noise would invalidate the forward model while agreement would confirm the calibration.

Watch

Extended reading notes

Core claim

The central claim is that the inverse problem of recovering physical labeling parameters from a scalar NV magnetic image is solvable: a CNN trained purely on synthetic dipole fields, with SPION clusters represented as point dipoles distributed on a sphere and saturated at 30 mT, predicts the cell diameter, the cell-to-NV-layer distance, the lateral position, and the iron mass for each measured dipole footprint. The authors validate the approach indirectly by comparing predicted cell diameters with brightfield images and by checking normalized field strengths and iron masses against independent sample-level measurements. With these reconstructions they construct a labeling-efficiency curve for HT29 cells showing a saturation plateau as SPION concentration rises, and they report that the maximum achievable normalized field for a 12 µm cell at 15 µm distance saturates at about 2.91 µT, with 2.71 µT reached at 500 µg/ml and the half-maximum, the practical optimum, at 116.39 µg/ml.

Load-bearing premise

The whole calibration depends on the simulated world matching the real one: the model treats SPION clusters on a cell as simple magnetic points spread over a sphere, uses a magnetic moment reported for one particle batch, and assumes the particles are fully magnetized at 30 mT; if any of these assumptions is off, every predicted iron mass and normalized field value is shifted even when the network fits its training data perfectly.

Editorial extensions

If this is right

  • The same measured field map yields per-cell iron mass and distance, so different samples, adhesive batches, and tilt conditions become quantitatively comparable without additional calibration.
  • The labeling curve shows that EpCAM-mediated SPION binding saturates, placing a physical ceiling on per-cell magnetic moment; the practical optimum concentration is 116.39 µg/ml.
  • At the highest tested concentration of 500 µg/ml, a normalized 12 µm cell at 15 µm from the sensor produces about 2.71 µT, which the authors estimate allows single-cell detection with SNR greater than 6 at up to 20 µm distance.
  • Because the network outputs derive from the same physical forward model, iron mass estimates can be checked against independent bulk iron measurements, providing a path to validation.

Reading between the lines

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

  • If the forward model is accurate, the same simulation-trained network should transfer to other cell lines and receptor targets without retraining, as long as the SPION magnetic moment and labeling geometry are updated; the saturation plateau specifically points to EpCAM site density as the bottleneck, so a reader could test whether co-targeting EGFR or HER2 lifts the ceiling above 2.91 µT.
  • An immediate testable extension is to compare per-cell CNN iron masses against single-cell ICP-MS or fluorescence-calibrated iron assays; agreement would convert the method from a relative to an absolute assay.
  • The simulation-based training also suggests that sensitivity limits could be probed purely in silico: adding noise and dipole overlap lets one predict the smallest iron mass or largest standoff distance at which the network's predictions remain unbiased, a question the paper only partially addresses.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. The manuscript presents a workflow for quantifying magnetic labeling of individual HT29 cells with SPIONs using NV-center widefield magnetic microscopy. A convolutional neural network is trained on synthetic magnetic-field snippets generated from a forward model of point dipoles distributed on a spherical cell surface; from measured dipole footprints the network predicts cell diameter, cell-to-sensor distance, and surface iron mass. The authors use these predictions to produce a labeling-efficiency curve versus SPION concentration and report a saturation plateau near 2.91 µT and a half-maximum concentration of 116.39 µg/ml. The paper argues that the CNN-based normalization removes distance and cell-size variability, enabling cross-sample comparison and a practical cost-benefit optimum.

Significance. If the quantitative calibration is trusted, the method would fill a genuine gap in magnetic cell labeling: per-cell, non-optical iron-mass quantification with a quantum sensor. The simulation-based training strategy is a reasonable approach to an ill-posed inverse problem, and the use of scalar NV-image formation with known ground truth is a strength. The paper also makes a falsifiable prediction (saturation of labeling with concentration) that is directly testable. However, the central quantitative outputs currently depend on a literature magnetic-moment value from a different SPION batch, on the assumption of full saturation at 30 mT, and on validation data relegated to the SI. These issues must be resolved before the reported absolute iron masses and field strengths can be accepted.

major comments (4)
  1. [Eq. (2) and surrounding calibration text] The calibration of the forward model is not self-consistent. Equation (2) uses the reported single-particle moment mP = 8.6e-19 A m^2 from Glenn et al. at 40 mT and sets the Langevin factor L to 1 at the 30 mT bias field. For this moment, the Langevin argument gives L(30 mT) ≈ 0.84 (Langevin argument x ≈ 6.2), not 1; if mP is instead treated as a saturation moment, the field is only about 84% of saturation. The SHS-20 batch used here is not magnetically characterized in the manuscript. Because the CNN outputs (iron mass m_j, normalized field, and hence the 2.71/2.91 µT values) scale with mP, this is a load-bearing systematic bias, not a cosmetic detail.
  2. [Statements around 'The validity of this approach...' and 'predicted mean iron masses align well...'] The only independent checks of the absolute iron-mass scale are mentioned in the main text as being 'qualitatively confirmed' and 'align well' with MP-AES measurements, with details deferred to the SI, which is not part of the reviewed submission. The central quantitative claim of the paper (iron mass per cell and the saturation curve in Fig. 3c) therefore rests on simulation self-consistency. The authors should either include the MP-AES comparison in the main text with per-condition statistics, or provide a calibrated phantom measurement with known iron mass to demonstrate that the CNN outputs are not systematically biased.
  3. [Fig. 3c] The saturation curve, the extrapolated maximum 2.91 µT, and the half-maximum concentration 116.39 µg/ml are presented without error bars, confidence intervals, or the number of cells per concentration. The 500 µg/ml data point was obtained with a different adhesive batch, which shifts the measured distance distribution; although the CNN is designed to compensate this, the combination of an uncontrolled mounting variable and absent statistical uncertainty makes the optimization claim unverifiable as presented. Please provide per-concentration means with cell counts and uncertainties, and a sensitivity analysis excluding the 500 µg/ml batch.
  4. [Forward model and CNN training section (from 'We therefore turned to numerical simulations...' to Fig. 2d)] The network is trained and evaluated on the same forward model that is later used to compute normalized field strengths. This is not circular, but it means that any systematic mismatch between the point-dipole-on-sphere model and the true SPION-cluster field (e.g., finite cluster size, demagnetization, inter-particle interactions, or non-spherical cell surfaces) can be absorbed by the network into biased predictions of distance, diameter, or iron mass. A phantom test with a known dipole source, or a comparison against an independent reconstruction method, is needed to establish simulation-to-experiment transfer. As it stands, the agreement in Fig. 2d is a self-consistency check, not a validation.
minor comments (4)
  1. [Fig. 2d caption vs main text] The text reports a mean normalized field of 1.34 µT for 100 µg/ml, while the Fig. 2d caption quotes a Voigt peak at 1.18 µT with width 0.69 µT; please reconcile or clarify which statistic is used.
  2. [SNR statement in the Results section] The phrase 'SNR ¿ 6' contains a non-standard symbol; if '>6' is intended, please specify the noise band and the distance at which this SNR holds.
  3. [Fig. 3 overall] The absence of any visual uncertainty representation (error bars, shaded bands, or per-point cell counts) makes the saturation trend difficult to assess; adding this information would substantially strengthen the quantitative claims.
  4. [General proofreading] There are several typographical issues (e.g., 'indivual' in the concluding paragraph, 'convolution neural network' in the methods section) and a few places where the sentence structure is convoluted; a careful proofread is recommended.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the CNN inversion is trained on an external forward model and checked against independent MP-AES; the normalized-field curve is a labeled simulation, not an independent prediction.

full rationale

The derivation chain is self-contained rather than circular. The forward model (Eqs. 1-2) is a physical dipole superposition with the single-particle moment taken from an external source (Glenn et al., ref. 23), not from the present paper. The CNN is trained on simulations of that forward model with known ground truth, then applied to measured widefield magnetic images to predict cell diameter, distance, and iron mass. The paper explicitly states an independent check: "The predicted mean iron masses align well with experimentally determined iron contents within prepared samples (see SI)" - a comparison to bulk MP-AES measurements, which is external to the network's training data. The normalized magnetic-field curve (Figure 3c, green graph) is presented transparently as a simulation: "Simulating the magnetic field with our simulation framework for a fixed distance and cell diameter allows semi-normalization of the magnetic field strengths expected by a given SPION concentration." This is a defined transformation of the predicted iron mass through the same forward model, not an independent measurement, but the paper labels it as "simulative normalization" rather than as a new empirical result. The saturation plateau (2.91 uT) and half-maximum concentration (116.39 ug/ml) are descriptive fits to these outputs, not claimed to be derived from first principles independent of the data. No load-bearing self-citation appears: the cited prior work by the same authors (refs. 7, 14) is not used to justify the forward model or the network architecture. The main risk is the unverified transfer of the Glenn et al. SPION moment to the SHS-20 batch and the assumption L -> 1 at 30 mT, but that is a validity/calibration concern, not circularity. Therefore the paper's central reconstruction is not equivalent to its inputs by construction.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central claims rest on a calibration chain: the particle moment from a different experiment, a spherical cell model, the saturation assumption, and the generalization of a learned inverse model. The first three are domain assumptions from prior literature; the fourth is a methodological assumption. The fitted saturation curve parameters are extracted from the data and carry no independent evidence.

free parameters (4)
  • Particle magnetic moment mP = 8.6e-19 A m^2
    Taken from Glenn et al. 2015 for a different SPION system and assumed valid for the Ocean Nanotech SHS-20 batch. All iron mass estimates scale linearly with this value.
  • Saturation maximum field = 2.91 uT
    Obtained by extrapolating the concentration series in Figure 3c, not derived or independently predicted.
  • Half-maximum SPION concentration = 116.39 ug/ml
    Fitted from the same concentration series as the maximum; quoted as a cost-benefit trade-off without error bars.
  • Simulation hyperparameters (N point dipoles, noise levels, snippet size) = see SI
    The number of dipoles per cell, noise levels, and the 32x32 pixel snippet size are chosen by the authors and detailed only in the SI.
assumptions (4)
  • domain assumption SPIONs are magnetically saturated at the 30 mT bias field, so the Langevin function L approaches 1 and the cluster moment scales linearly with particle number.
    This is stated in Eq. 2 and the surrounding text; if L is significantly below 1 at 30 mT, the linear scaling and the inferred iron mass are wrong.
  • domain assumption The magnetic moment per SPION is the same as that reported by Glenn et al. 2015 for their particles.
    The paper explicitly relies on the value 8.6e-19 A m^2 from Ref. 23 to convert the measured fields to iron mass; this is a calibration input from prior literature.
  • domain assumption A spherical cell surface with N point dipoles is an adequate forward model for a real HT29 cell with clustered SPIONs.
    The entire training dataset is generated from this model. Deviations in real cell shape or SPION clustering are not accounted for in the main text.
  • domain assumption The CNN trained on simulated data generalizes to experimental magnetic field maps, including any systematic differences in background, noise, and adhesion.
    The paper uses simulations as ground truth and applies the network to experimental data, claiming distances and iron masses are captured; the SI is said to provide qualitative validation, but the main text does not.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Optimizing SPION Labeling for Single-Cell Magnetic Microscopy." pith.science (2026). https://pith.science/paper/F6DFF57M

@misc{pith2026250520373,
  author       = {Pith},
  title        = {Pith review of: Optimizing SPION Labeling for Single-Cell Magnetic Microscopy},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/F6DFF57M}},
  note         = {Machine review of arXiv:2505.20373}
}
read the original abstract

This study explores the correlation between iron mass on cell surfaces and the resultant magnetic field. Human colorectal cancer cells (HT29 line) were labeled with varying concentrations of SPIONs and imaged via a NV center widefield magnetic microscope. To assess the labeling efficacy, a convolutional neural network trained on simulated magnetic dipole data was utilized to reconstruct key labeling parameters on a cell-by-cell basis, including cell diameter, sensor proximity, and the iron mass associated with surface-bound SPIONs. Our analysis provided quantitative metrics for these parameters across a range of labeling concentrations. The findings indicated that increasing SPION concentration enhances both the cell-surface iron mass and magnetic field strength, demonstrating a saturation effect. This methodology offers a coherent framework for the quantitative, high-throughput characterization of magnetically labeled cells, presenting significant implications for the fields of cell biology and magnetic sensing applications.

Figures

Figures reproduced from arXiv: 2505.20373 by the authors.

Figure 1
Figure 1. NV Center based Widefield Magnetometry. (a) Widefield magnetic field microscope to observe magnetically labeled cells. The sample under investigation is mounted to the NV layer of the diamond substrate with a thin layer of UV curing optical adhesive. The CW-ODMR measurement is performed by exciting the NV centers in the center of the FoV with a 532 nm laser and simultaneously sweeping the frequency of a microwave si… view at source ↗
Figure 2
Figure 2. Field strength normalization of magnetically labeled cells. (a) Brightfield microscopy image show￾ing circular cell structures. Small black spots are dust accumulation on the camera sensor. Corresponding magnetic field map displaying dipole signatures with color-coded magnitude or the reconstructed magnetic field strength along the chosen NV axis. (b) Schematic representation of the dipole snippet selection process … view at source ↗
Figure 3
Figure 3. Evaluation of the magnetic labeling efficiency with respect to the SPION concentration based on NN predictions. (a) Cell diameter distribution prediction for all observed samples. A distribution centered around a mean of approximately 12 µm is observed. This matches visual confirmation in brightfield images taken prior to a magnetic field measurement (compare Figure 2a). (b) Cell distance distribution for all observ… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 29 canonical work pages

  1. [1]

    Z.; Takahashi, M.; Ishida, Y.; Hata, J.; Shimozono, S.; Namiki, K.; Fukano, T.; Kiyama, M.; Okano, H.; Kizaka-Kondoh, S.; McHugh, T

    Iwano, S.; Sugiyama, M.; Hama, H.; Watakabe, A.; Hasegawa, N.; Kuchimaru, T.; Tanaka, K. Z.; Takahashi, M.; Ishida, Y.; Hata, J.; Shimozono, S.; Namiki, K.; Fukano, T.; Kiyama, M.; Okano, H.; Kizaka-Kondoh, S.; McHugh, T. J.; Yamamori, T.; Hioki, H.; Maki, S. et al. Single-cell bioluminescence imaging of deep tissue in freely moving animals. Science 2018,...

    work page 2018

  2. [2]

    Multi-color live- cell super-resolution volume imaging with multi-angle interference microscopy

    Chen, Y.; Liu, W.; Zhang, Z.; Zheng, C.; Huang, Y.; Cao, R.; Zhu, D.; Xu, L.; Zhang, M.; Zhang, Y.-H.; Fan, J.; Jin, L.; Xu, Y.; Kuang, C.; Liu, X. Multi-color live- cell super-resolution volume imaging with multi-angle interference microscopy. Nature Communications 2018, 9, 4818

    work page 2018

  3. [3]

    Microscope-based techniques to study cell adhesion and migration

    Roy, P.; Rajfur, Z.; Pomorski, P.; Jacobson, K. Microscope-based techniques to study cell adhesion and migration. Nature Cell Biology 2002, 4, E91–E96

    work page 2002

  4. [4]

    Advances in Magnetoresistive Biosensors

    Su, D.; Wu, K.; Saha, R.; Peng, C.; Wang, J.-P. Advances in Magnetoresistive Biosensors. Micromachines 2020, 11, 34. 15

    work page 2020

  5. [5]

    I.; Shao, H.; Min, C.; Lee, K.; Liong, M.; Weissleder, R.; Lee, H

    Issadore, D.; Park, Y. I.; Shao, H.; Min, C.; Lee, K.; Liong, M.; Weissleder, R.; Lee, H. Magnetic sensing technology for molecular analyses. Lab on a Chip 2014, 14, 2385–2397

    work page 2014

  6. [6]

    Nanodiamond-Based Optical-Fiber Quantum Probe for Magnetic Field and Biological Sensing

    Chen, Y.; Lin, Q.; Cheng, H.; Huang, H.; Shao, J.; Ye, Y.; Liu, G.-S.; Chen, L.; Luo, Y.; Chen, Z. Nanodiamond-Based Optical-Fiber Quantum Probe for Magnetic Field and Biological Sensing. ACS Sensors 2022, 7, 3660–3670

    work page 2022

  7. [7]

    K.; Dasari, D

    Hollendonner, M.; Sharma, S.; Parthasarathy, S. K.; Dasari, D. B. R.; Finkler, A.; Kusminskiy, S. V.; Nagy, R. Quantum sensing of electric field distributions of liquid electrolytes with NV-centers in nanodiamonds.New Journal of Physics 2023, 25, 093008

    work page 2023

  8. [8]

    Compact and Fully Integrated LED Quantum Sensor Based on NV Cen- ters in Diamond

    Pogorzelski, J.; Horsthemke, L.; Homrighausen, J.; Stiegek¨ otter, D.; Gregor, M.; Gl¨ osek¨ otter, P. Compact and Fully Integrated LED Quantum Sensor Based on NV Cen- ters in Diamond. Sensors 2024, 24, 743

    work page 2024

Show all 29 references
  1. [9]

    I.; Foy, C.; Trusheim, M

    Kim, D.; Ibrahim, M. I.; Foy, C.; Trusheim, M. E.; Han, R.; Englund, D. A CMOS- integrated quantum sensor based on nitrogen–vacancy centres. Nature Electronics 2019, 2, 284–289

    2019

  2. [10]

    Toward Quantitative Bio-sensing with Nitrogen-Vacancy Center in Diamond

    Zhang, T.; Pramanik, G.; Zhang, K.; Gulka, M.; Wang, L.; Jing, J.; Xu, F.; Li, Z.; Wei, Q.; C ´ ıgler, P.; Chu, Z. Toward Quantitative Bio-sensing with Nitrogen-Vacancy Center in Diamond. ACS Sensors 2021, 6, 2077–2107

    2021

  3. [11]

    S.; Degen, C.; Abendroth, J

    Janitz, E.; Herb, K.; V¨ olker, L.; Huxter, W. S.; Degen, C.; Abendroth, J. Diamond surface engineering for molecular sensing with nitrogen—vacancy centers. Journal of Materials Chemistry C 2022, 10, 13533–13569

    2022

  4. [12]

    Nanoscale electric-field imaging based on a quantum sensor and its charge-state control under ambient condition

    Bian, K.; Zheng, W.; Zeng, X.; Chen, X.; St¨ ohr, R.; Denisenko, A.; Yang, S.; Wrachtrup, J.; Jiang, Y. Nanoscale electric-field imaging based on a quantum sensor and its charge-state control under ambient condition. Nature Communications 2021, 12, 2457. 16

    2021

  5. [13]

    P.; Petracic, O.; Feoktystov, A

    Nasser, A.; Qdemat, A.; Unterweger, H.; Tietze, R.; Sun, X.; Landers, J.; Kopp, J.; Wu, B.; Appavou, M.-S.; Murmiliuk, A.; Gilbert, E. P.; Petracic, O.; Feoktystov, A. Impact of coating type on structure and magnetic properties of biocompatible iron ox- ide nanoparticles: insi...

    2024

  6. [14]

    P.; Kappes, M.; Cicha, I.; Tietze, R.; Braun, C.; Schneider-Stock, R.; Nagy, R.; Alexiou, C.; Janko, C

    Friedrich, R. P.; Kappes, M.; Cicha, I.; Tietze, R.; Braun, C.; Schneider-Stock, R.; Nagy, R.; Alexiou, C.; Janko, C. Optical microscopy systems for the detection of unla- beled nanoparticles. International Journal of Nanomedicine 2022, 17, 2139–2163

    2022

  7. [15]

    N.; Stanicki, D.; Laurent, S

    Vangijzegem, T.; Lecomte, V.; Ternad, I.; Van Leuven, L.; Muller, R. N.; Stanicki, D.; Laurent, S. Superparamagnetic iron oxide nanoparticles (SPION): From fundamentals to state-of-the-art innovative applications for cancer therapy. Pharmaceutics 2023, 15, 236

    2023

  8. [16]

    Superparamagnetic Iron Oxide Nanoparticles-current and prospective med- ical applications

    Duli´ nska-Litewka, J.; Lazarczyk, A.; Ha lubiec, P.; Szafra´ nski, O.; Karnas, K.; Karewicz, A. Superparamagnetic Iron Oxide Nanoparticles-current and prospective med- ical applications. Materials 2019, 12, 617

    2019

  9. [17]

    B.; Malheiros, J

    Mamani, J. B.; Malheiros, J. M.; Cardoso, E. F.; Tann´ us, A.; Silveira, P. H.; Gamarra, L. F. In vivo magnetic resonance imaging tracking of C6 glioma cells la- beled with superparamagnetic iron oxide nanoparticles. Einstein (Sao Paulo) 2012, 10, 164–170

    2012

  10. [18]

    Y.; Sheth, R.; Seifalian, A.; Green, M.; Seifalian, A

    Bull, E.; Madani, S. Y.; Sheth, R.; Seifalian, A.; Green, M.; Seifalian, A. M. Stem cell tracking using iron oxide nanoparticles. International Journal of Nanomedicine 2014, 9, 1641–1653

    2014

  11. [19]

    Targeting strategies for superparamagnetic iron oxide nanoparticles in cancer therapy

    Zhi, D.; Yang, T.; Yang, J.; Fu, S.; Zhang, S. Targeting strategies for superparamagnetic iron oxide nanoparticles in cancer therapy. Acta Biomaterialia 2020, 102, 13–34. 17

    2020

  12. [20]

    W´ ang, Y. X. J.; Id´ ee, J.-M. A comprehensive literatures update of clinical researches of superparamagnetic resonance iron oxide nanoparticles for magnetic resonance imaging. Quantitative Imaging in Medicine and Surgery 2017, 7, 88–122

    2017

  13. [21]

    Iron oxide nanoparticles: An alternative for positive contrast in magnetic resonance imaging

    Fern´ andez-Barahona, I.; Mu˜ noz-Hernando, M.; Ruiz-Cabello, J.; Herranz, F.; Pellico, J. Iron oxide nanoparticles: An alternative for positive contrast in magnetic resonance imaging. Inorganics 2020, 8, 28

    2020

  14. [22]

    Liu, Y.; Ji, X.; Tong, W. W. L.; Askhatova, D.; Yang, T.; Cheng, H.; Wang, Y.; Shi, J. Engineering multifunctional RNAi nanomedicine to concurrently target cancer hallmarks for combinatorial therapy. Angewandte Chemie International Edition 2018, 57, 1510– 1513

    2018

  15. [23]

    R.; Lee, K.; Park, H.; Weissleder, R.; Yacoby, A.; Lukin, M

    Glenn, D. R.; Lee, K.; Park, H.; Weissleder, R.; Yacoby, A.; Lukin, M. D.; Lee, H.; Walsworth, R. L.; Connolly, C. B. Single-cell magnetic imaging using a quantum diamond microscope. Nature Methods 2015, 12, 736–738

    2015

  16. [24]

    R.; DeVience, S

    Le Sage, D.; Arai, K.; Glenn, D. R.; DeVience, S. J.; Pham, L. M.; Rahn-Lee, L.; Lukin, M. D.; Yacoby, A.; Komeili, A.; Walsworth, R. L. Optical magnetic imaging of living cells. Nature 2013, 496, 486–489

    2013

  17. [25]

    Immunomagnetic microscopy of tumor tissues using quantum sensors in diamond

    Chen, S.; Li, W.; Zheng, X.; Yu, P.; Wang, P.; Sun, Z.; Xu, Y.; Jiao, D.; Ye, X.; Cai, M.; Shen, M.; Wang, M.; Zhang, Q.; Kong, F.; Wang, Y.; He, J.; Wei, H.; Shi, F.; Du, J. Immunomagnetic microscopy of tumor tissues using quantum sensors in diamond. Proceedings of the Nation...

    2022

  18. [26]

    A.; Vidal, X

    Mathes, N.; Comas, M.; Bleul, R.; Everaert, K.; Hermle, T.; Wiekhorst, F.; Knittel, P.; Sperling, R. A.; Vidal, X. Nitrogen-vacancy center magnetic imaging of Fe3O4 nanopar- ticles inside the gastrointestinal tract of Drosophila melanogaster. Nanoscale Advances 2023, 6, 247–255. 18

    2023

  19. [27]

    2025; https://oceannanotech.com/web/, Accessed: January 2025

    OCEAN NANOTECH, LLC 7964 Arjons Drive, Suite G, San Diego, CA 92126. 2025; https://oceannanotech.com/web/, Accessed: January 2025

    2025

  20. [28]

    2025; https: //www.qd-st.com/, Accessed: January 2025

    QuantumDiamonds GmbH Friedenstraße 18, 81671 M¨ unchen, Germany. 2025; https: //www.qd-st.com/, Accessed: January 2025

    2025

  21. [29]

    High sensitivity magnetic imaging using an array of spins in diamond

    Steinert, S.; Dolde, F.; Neumann, P.; Aird, A.; Naydenov, B.; Balasubramanian, G.; Jelezko, F.; Wrachtrup, J. High sensitivity magnetic imaging using an array of spins in diamond. Review of Scientific Instruments 2010, 81, 043705. 19

    2010

Pith tools

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