The-Bodega: A Matlab Toolbox for Biologically Dynamic Microbubble Simulations on Realistic Hemodynamic Microvascular Graphs

Alexis Leconte; Alice Wu; Andreas Linninger; Jean Provost; Jonathan Poree; Joshua Kinugasa; Maxence Laplante-Berthier; Pierre-Olivier Bouchard; Samuel Mihelic; Simon Desrocher

REVIEW 5 major objections 6 minor 39 references

The-Bodega: A Matlab Toolbox for Biologically Dynamic Microbubble Simulations on Realistic Hemodynamic Microvascular Graphs

T0 review · 5 major / 6 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read The paper presents The-Bodega, an open-source toolbox that simulates thousands of microbubbles flowing through realistic brain and heart vascular networks, and converts their trajectories into ground-truth ultrasound localization microscopy

desk verdict A well-engineered ULM simulation toolbox with a couple of self-consistent demonstrations that are described as validations; deserves review once code and data ship. read the letter →

arxiv 2509.08149 v1 pith:AH3BDNAQ submitted 2025-09-09 physics.med-ph cs.SEphysics.app-ph

Stephen Alexander Lee , Alexis Leconte , Alice Wu , Jonathan Poree , Maxence Laplante-Berthier , Simon Desrocher , Pierre-Olivier Bouchard , Joshua Kinugasa

show 3 more authors

Samuel Mihelic Andreas Linninger Jean Provost

This is my paper · ORCID

classification physics.med-phcs.SEphysics.app-ph

keywords ultrasoundlocalizationmicroscopymicrobubblesimulationhemodynamicmodelingMonteCarlopulsatileflowvasculargraphground-truthbenchmarkingfunctional

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

The paper presents The-Bodega, an open-source Matlab toolbox whose aim is to generate ground-truth ultrasound localization microscopy (ULM) data: ultrasound images and raw radio-frequency signals in which the true positions, velocities, and pulsatile phases of every microbubble are known. It claims this closes a gap between simple synthetic benchmarks and in vivo ULM by letting researchers evaluate tracking, clutter filtering, motion correction, and functional imaging against a realistic digital phantom of the microvasculature. The pipeline operates on anatomically validated directed graphs of the mouse brain and human coronary circulation, propagates microbubbles with Monte Carlo sampling and pulsatile flow, and synthesizes acoustic data with a linear ultrasound simulator. Demonstrations include capillary mesh saturation times, SVD-filter-induced capillary ablation, beating-heart motion artifacts, and simulated neurovascular responses to whisker stimulation.

What carries the argument

The carrying mechanism is a directed vascular graph coupled to a sequential Monte Carlo microbubble simulator. The graph supplies all possible trajectories; each microbubble is propagated along a randomly weighted path under a Poiseuille flow profile modulated by pulse-decomposition-analysis (PDA) waveforms that are precomputed per edge and shifted in time. The resulting microbubble positions, velocities, and radii are passed to a linear acoustic simulator (SIMUS) that renders RF and IQ data, with tissue and skull clutter added as separate scatterer blocks. This design preserves exact ground truth at every stage while allowing arbitrary vascular architectures and transducer configurations.

What would settle it

Record microbubble tracks in a living mouse brain with ULM and compare individual capillary transit times, velocity pulsatility phase lags, and branch choices to simulations run on the same vascular graph; a systematic mismatch would falsify the claim that the simulated ground truth is biologically realistic.

Watch

Extended reading notes

Core claim

The central claim is that The-Bodega provides an end-to-end, modular simulation framework that produces ULM ground truth of sufficient hemodynamic and anatomical realism to stand in for in vivo acquisitions. The authors show that by starting from directed vascular graphs annotated with vessel radii, flow, and pulse pressure, they can simulate stochastic microbubble trajectories, encode them in HDF5 with full ground truth, and then generate RF/IQ ultrasound data with tissue clutter and motion. They demonstrate that this lets them quantify how long it takes to populate the capillary mesh, how SVD clutter filtering ablates slow capillary signals and degrades resolution, how cardiac tissue motio

Load-bearing premise

The realism rests on treating blood as a simple pipe flow and microbubbles as passive spheres that only slow down or vanish in narrow vessels; if real microbubbles stick, deform, or interact with sound, the simulated ground truth will not match in vivo.

Editorial extensions

If this is right

ULM algorithms can be benchmarked under controlled, known-truth conditions that separate the effects of microbubble concentration, tissue clutter, and tissue motion.
The capillary saturation analysis suggests that roughly three to five minutes of acquisition may suffice to sample capillary function, even though full capillary mesh reconstruction takes much longer.
SVD clutter filtering is shown to preferentially ablate slow capillary signals, and higher eigenvalue cutoffs degrade ULM resolution; this motivates alternative clutter-removal strategies.
In the beating heart, the simulations indicate that tissue clutter, not motion itself, is the dominant obstacle, because motion alone with correction preserves most vessels while motion plus clutter destroys them.
Simulated neurovascular responses could serve as test data for functional ULM analysis pipelines and for studying the link between blood-volume changes and ULM-derived activation maps.

Reading between the lines

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

The same pipeline could be used to stress-test adaptive or spatially-varying SVD filters, nonlinear contrast imaging, and motion-compensation algorithms before deploying them in vivo.
Because the simulator accepts arbitrary directed graphs and user-defined pulse waveforms, it could generate pathological hemodynamic patterns to identify ULM-visible biomarkers of vascular disease.
A direct experimental comparison between simulated microbubble trajectories and in vivo microbubble tracks in the same vascular network would be a natural next step; the paper currently validates pulsatility against its own imposed waveforms.
The ground-truth labels and raw data could be used to train deep learning models for localization, tracking, and denoising without manual annotation, an application the authors note but leave largely undeveloped.

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.

Desk Editor's Note

read the letter

Here's my take on The-Bodega. Read it as a software toolbox preprint, not a physiological validation study. The architecture is genuinely modular: arbitrary directed vascular graphs, Monte Carlo microbubble propagation, PDA-derived pulsatile waveforms, cardiac motion imported from in vivo echo, neurovascular augmentation, and SIMUS-based ultrasound generation. The engineering work is real, and the computational benchmarks show serious attention to scaling—the allpaths bottleneck, GPU/parallel speedups. The capillary saturation times cross-check against independent estimates, which is the strongest external evidence in the paper. The SVD ablation demonstration is also useful: it quantifies how aggressive eigen-truncation kills slow capillary tracks, a real concern for ULM.

The main soft spot is the same one flagged in the stress test: Sec 3.2 is self-consistent rather than externally validated. The PDA waveforms are imposed inputs in Sec 2.2; the dULM pipeline in Sec 2.5 uses the same cardiac-cycle timing for realignment; then Sec 3.2 reports that dULM recovers the imposed waveforms and calls it validation. That's a good integration test, but it doesn't validate the hemodynamic model. The authors do include a caveat in the Discussion that flow should be validated before simulation, but the phrasing in Sec 3.2 goes further than the evidence supports. Similarly, the fULM activation follows the injected stimulus by construction; the authors are honest about the model being relatively simple, and the correlation values matching in vivo ranges is a sanity check, not a validation.

More minor issues: the node importance score is a linear combination of betweenness and PageRank, but no weights are given. Several quantitative results (Dice, Jaccard, PI) are reported without error bars or repeats. And the code and data are still to-be-published, which blocks independent inspection. That's the single most important fix for a software paper.

On balance, I think the stress-test note overreaches if it says the central claim is unsupported. The paper's central claim is that this is a flexible, open-source benchmarking platform, and the design and demonstrations support that. The claim that it substitutes for in vivo is not something the authors lean on heavily—they explicitly flag the need for flow validation. I'd send this to peer review with a request for full code/data release, explicit node-importance weights, and a rewrite of Sec 3.2's validation language. A serious referee group would add value, especially on the software engineering side. I'd bring it to reading group, and I would cite it once the code is out.

Referee Report

5 major / 6 minor

Summary. The paper introduces The-Bodega, a Matlab toolbox for simulating microbubble trajectories on anatomically derived vascular graphs and generating synthetic Ultrasound Localization Microscopy (ULM) datasets. The pipeline combines sequential Monte Carlo particle simulation with Poiseuille flow profiles and PDA-derived pulsatile waveforms (Algorithm 1), then feeds resulting trajectories into a SIMUS-based linear ultrasound simulator with optional tissue clutter and motion. Demonstrations cover computational benchmarking, 3D pulsatility via dynamic ULM, capillary saturation and SVD clutter-filter ablation, cardiac motion artifacts, and simulated neurovascular responses for functional ULM. The authors position the toolbox as an open-source resource for generating realistic ground-truth ULM data for algorithm benchmarking and deep learning training.

Significance. If released with working code and data, The-Bodega would be a valuable community resource, complementing existing simulators such as PALA, BUFF, and PROTEUS. Its modular design, support for arbitrary vascular graphs, HDF5 output, and CPU/GPU parallelization are practical strengths. The capillary saturation times reported in Fig. 9D appear consistent with independent estimates [5,35], and the controlled SVD ablation and cardiac-motion experiments are plausible demonstrations of the kind of studies such a simulator enables. However, the central claim of hemodynamic realism is not yet externally validated; the current evidence supports a self-consistent digital phantom rather than a substitute for in vivo acquisitions. The computational scaling results (Fig. 7, Table 1) are useful and well presented.

major comments (5)

[Sec. 3.2 with Algorithm 1 and Sec. 2.5] The pulsatility validation is a round-trip rather than an independent test. The PDA waveforms are an input (Algorithm 1), and dULM realignment uses the same simulated cardiac-phase information from the simulation (§2.5). The recovered pulsatility index is then described as validating "a model of pulsatility that mimics in vivo pulsatile flow." This only checks internal consistency. Please add external validation against in vivo microbubble trajectory or Doppler velocity waveforms (e.g., in the same mouse-brain or human-heart context), or explicitly reframe the demonstration as self-consistency, not hemodynamic validation.
[Sec. 2.2] The microbubble-vessel interaction rule is an ad hoc heuristic: if a bubble's diameter exceeds vessel diameter, it decelerates until it shrinks or disappears. No experimental reference or validation is provided, and there is no sensitivity analysis to the parameters of this rule. Since this rule directly controls capillary occupancy and disappearance, it is load-bearing for the capillary saturation and SVD ablation conclusions. Please validate against measured microbubble trajectories or, at minimum, provide a sensitivity analysis and describe the rule as a heuristic whose impact on downstream metrics is quantified.
[Sec. 2.6 and Fig. 9] The node-importance weighting is under-specified. The text states the importance is "a linear combination of the betweenness centrality and PageRank estimator for node connectivity," but the weights, normalization, and directed/undirected treatment are not given. The definition of the track "score" used in Fig. 9E is also missing. Without these details, the claim that 95% of highly influential capillary nodes are populated within 2-3 minutes cannot be reproduced or evaluated. Please provide the exact formula and parameter values.
[Sec. 3 (Figs. 8-12)] Quantitative results are reported without repeats, error bars, or confidence intervals. Because the simulator is stochastic (random sampling of trajectories, bubble distributions, and subsets in Secs. 2.2-2.3), all reported curves and metrics depend on random seeds. This applies to Fig. 9 saturation curves, Fig. 10F Dice/Jaccard/sensitivity/specificity, Fig. 8G pulsatility index distributions, and Fig. 12 fULM correlations. The absence of variability estimates undermines quantitative conclusions such as "3-5 minute scan time may be sufficient" and the FRC resolution values. Please include multiple independent runs with corresponding statistics.
[Intro, Sec. 2.1, Sec. 4] The paper repeatedly describes The-Bodega as "fully open-source" and its datasets as "openly available," but the code repository and data links are listed as "to-be-published" (Intro and Sec. 2.1). For a software/toolbox paper, this availability is load-bearing and should be resolved with a permanent DOI or repository before publication. Additionally, Sec. 4 states that "the flow should be validated before simulation for accurate representation"; this is a significant limitation that should be elevated to a clearly stated caveat with quantitative evidence for the default graphs, rather than only a remark in the discussion.

minor comments (6)

[Eq. (1)] ΔV is described as "stroke volume (mL)" but Eq. (1) represents the compliance relation ΔV = C·ΔP, where ΔV is the volume change. Please clarify whether stroke volume or pulse-induced volume change is intended, as these differ in general.
[Sec. 2.4] Typo: "for in vivomice" should be "for in vivo mice." Also, the sentence beginning "To draw parallels, in the human heart" could be reworded for clarity.
[Sec. 3.2] Typo: "arterioes" should be "arterioles." The sentence "Here, In the contrast-enhanced Doppler volume..." has an awkward capitalization and comma structure.
[Fig. 10 and text] In the text describing Figure 10, "Zoomed image of microbubble-only tracking from F" appears to be an error: Figure 10F contains quantitative metrics, not an image. The reference should likely be to panel A or B. Please correct.
[Table 1] The table lists a "Toy" network. In the main text only the whole-brain, half-brain, heart, and synthetic capillary networks are mentioned. Please define "Toy" in the text or rename the entry.
[Sec. 2.8] "This assumption rapidly fails in moving organs..." is missing a comma and the antecedent "This" is ambiguous (it refers to a multi-sentence prior statement). Rephrase for clarity.

Circularity Check

2 steps flagged · score 4.0 of 10

Demonstration-level circularity: dULM and fULM recover their own injected waveforms/activations, but the core simulator pipeline is not circular.

self definitional [Section 2.5 (dULM processing) and Section 3.2 (3D Pulsatility); Algorithm 1]
"Here, in lieu of ECG signal, the tissue Doppler used to spatiotemporally align microbubble phase in the cardiac cycle during the simulation was used for realignment. … Importantly, this characterizes the use of the PDA algorithm for propagating the pulse wave through these microvascular network and validates a model of pulsatility that mimics in vivo pulsatile flow."

The PDA waveform is not an independent physiological measurement: Algorithm 1 takes pulse parameters as input and emits edge-wise pulsatile waveforms, which are scaled by compliance (Eq. 1) and propagated through the graph. The dULM realignment then reuses the simulation's own cardiac-cycle phase information, so the pulsatility index reported in §3.2 is the imposed waveform re-detected after simulation and tracking. Calling this a 'validation' of an in-vivo-like pulsatility model is therefore a round-trip check, not a test against independent data.
self definitional [Section 2.7 (Simulating Neurovascular responses) and Section 3.5 (functional ULM)]
"To dynamically simulate functional hemodynamic responses to stimulus-evoked impulses, two microbubble distributions were simulated: steady-state and full whisker activation … the trajectory selection probability distribution was augmented to preferentially increase vascular paths that lead to the BF cortex … we calculated the Pearson correlation coefficient between the ULM images (sliding window size of 4 s, 0.4 s stride) and the pulse train (Figure 5C)."

The activation is inserted into the simulation by augmenting flow and trajectory-selection probabilities in the left S1BF barrel-field ROI, and then fULM correlates the resulting density images with the very same pulse train that drove the sampling between steady-state and activated distributions. Positive correlation in the barrel field is thus guaranteed by construction; the demonstration does not independently validate the neurovascular-coupling model. The paper concedes in §4 that the sampling method is 'relatively simple' and 'may not entirely reflect microvascular flow changes in vivo.'

full rationale

This is a software/toolbox paper, not a closed-form derivation chain, so the classic Eq-X-equals-Eq-Y circularity does not arise. The core pipeline—vascular graph inputs, Monte Carlo bubble trajectories, SIMUS RF/IQ synthesis, and ULM reconstruction—is self-contained and gives direct access to ground truth; benchmarking SVD filtering, FRC, and capillary saturation against that ground truth is legitimate, and the capillary-population results are cross-checked against independent reports [5,35]. The two circular elements are in the demonstration/validation sections. Section 3.2 labels the PDA pulsatility demonstration as a 'validation' of a model 'that mimics in vivo pulsatile flow,' but the PDA waveform is a user-specified input and dULM realignment reuses the simulation's own cardiac-cycle phase; the recovered pulsatility index is therefore the input replayed through the pipeline. Section 3.5 similarly injects the neurovascular response by augmenting flow in the S1BF region and then computes fULM correlation against the same pulse train, so recovery of activation is by construction. The paper itself concedes (§4) that flow 'should be validated before simulation' and that the fULM sampling model 'may not entirely reflect microvascular flow changes in vivo.' Those concessions are external-validity caveats, not evidence of a circular derivation. Prior-work citations ([14], [18], [34]) are published external results rather than unverified self-referential premises, and they do not force the core simulator's outputs. Overall, the circularity is partial and demonstration-level; the central toolbox contribution remains independently meaningful, though its in-vivo realism is not independently established.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The central claim depends on a chain of domain assumptions in hemodynamics, acoustics, and imaging physics, plus a number of user-specified parameters. The most consequential are the input vascular graphs (from prior in vivo-validated models) and the pulsatile flow construction, which is neither fitted nor independently validated here. No new physical entities are introduced.

free parameters (6)

Microbubble size distribution (mean, std) = mean = 2 um, std = 3 um
Assumed Definity distribution (Sec 2.2), controls which vessels a bubble can enter and when it disappears; from manufacturer or literature rather than fitted in this paper.
Vascular compliance C = user-specified
Eq. 1 scales stroke volume from pulse pressure; user parameter that sets pulsatility amplitude across the graph (Sec 2.2).
PDA pulsatile waveform parameters = heart rate, pulse wave velocity, 5 Gaussian components
Pulse Decomposition Analysis parameters adapted from [20]; determine waveform shape and transit, hence all pulsatile velocity traces and dULM results (Sec 2.2, Algorithm 1).
Poiseuille radial scaling factor p = not given
Radial offset with normalization 1/sqrt(1-p) sets cross-sectional distribution of bubbles along vessel (Sec 2.2); value not reported.
Hemodynamic response function parameters = FWHM = 4 s, delay = 1 s
Gamma-variate HRF parameters convolved with neural response (Sec 2.7) determine the timing of functional ULM activation; from literature [33].
Capillary node importance weighting = unspecified
Linear combination of betweenness centrality and PageRank (Sec 2.6) without the combination coefficient, so the node saturation curves in Fig 9E are not exactly reproducible.

assumptions (6)

domain assumption Poiseuille flow approximates all microvascular flow profiles
Used to compute radial bubble positions and forward propagation (Sec 2.2).
domain assumption PDA pulse shape propagated over graph edges achieves steady-state pulsatility
Algorithm 1 treats the arterial pulse as 5 Gaussian components and accumulates shifted waveforms; no in vivo validation provided (Sec 2.2).
domain assumption Linear ultrasound simulation (SIMUS) is sufficient for ULM ground truth
Sec 2.4 uses SIMUS, a linear particle-based simulator; nonlinear bubble echoes are ignored.
domain assumption Microbubble dynamics reduce to size-checked passive advection
Bubbles either pass, decelerate, or disappear based on diameter vs vessel diameter (Sec 2.2); no deformation or acoustic force modeling.
domain assumption The vascular graphs from [18] and [19] accurately represent in vivo anatomy and flow
Entire simulation runs on these graphs; flow velocities were validated in the source publications, not re-validated in this paper (Sec 2.1).
domain assumption Buxton neurovascular coupling model and gamma-variate HRF describe whisker-stimulation hemodynamics
Sec 2.7 derives the activation sampling distribution from [33] with tuned FWHM and delay; authors note flow redistribution is omitted.

how reviews work

0 comments

Cite this review

Pith. "Pith review of The-Bodega: A Matlab Toolbox for Biologically Dynamic Microbubble Simulations on Realistic Hemodynamic Microvascular Graphs." pith.science (2026). https://pith.science/paper/AH3BDNAQ

@misc{pith2026250908149,
  author       = {Pith},
  title        = {Pith review of: The-Bodega: A Matlab Toolbox for Biologically Dynamic Microbubble Simulations on Realistic Hemodynamic Microvascular Graphs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AH3BDNAQ}},
  note         = {Machine review of arXiv:2509.08149}
}

read the original abstract

The-Bodega is a Matlab-based toolbox for simulating ground-truth datasets for Ultrasound Localization Microscopy (ULM)-a super resolution imaging technique that resolves microvessels by systematically tracking microbubbles flowing through the microvasculature. The-Bodega enables open-source simulation of stochastic microbubble dynamics through anatomically complex vascular graphs and features a quasi-automated pipeline for generating ground-truth ultrasound data from simple vascular inputs. It incorporates sequential Monte Carlo simulations augmented with Poiseuille flow distributions and dynamic pulsatile flow. A key novelty of our framework is its flexibility to accommodate arbitrary vascular architectures and benchmark common ULM algorithms, such as Fourier Ring Correlation and Singular Value Decomposition (SVD) spatiotemporal filtering, on realistic hemodynamic digital phantoms. The-Bodega supports consistent microbubble-to-ultrasound simulations across domains ranging from mouse brains to human hearts and automatically leverages available CPU/GPU parallelization to improve computational efficiency. We demonstrate its versatility in applications including image quality assessment, motion artifact analysis, and the simulation of novel ULM modalities, such as capillary imaging, myocardial reconstruction under beating heart motion, and simulating neurovascular evoked responses.

Figures

Figures reproduced from arXiv: 2509.08149 by the authors.

**Figure 2.** Mouse brain and human heart microvascular graphs with two sample cross [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗

**Figure 3.** Microbubble populations are simulated through precomputed vascular paths. [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗

Figures from the paper (9 more)

**Figure 4.** Figure 4: Resultant combination of microbubbles populated for one mouse whole brain [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]

**Figure 5.** Figure 5: Simulating microbubbles flowing through neurovascular-activated barrel field [PITH_FULL_IMAGE:figures/full_fig_p015_5.png]

**Figure 6.** Figure 6: Schematic representation of the simulation pipeline designed to integrate dy [PITH_FULL_IMAGE:figures/full_fig_p017_6.png]

**Figure 7.** Figure 7: Computational time for the ParticleSim class and the mouse whole brain cortex [PITH_FULL_IMAGE:figures/full_fig_p020_7.png]

**Figure 8.** Figure 8: 3D pulsatile flow in dynamic ULM via Row-Column Arrays in the Mouse Whole [PITH_FULL_IMAGE:figures/full_fig_p021_8.png]

**Figure 9.** Figure 9: Capillary mesh ULM analysis. (A) Ground truth 2D projection of the vascular [PITH_FULL_IMAGE:figures/full_fig_p023_9.png]

**Figure 10.** Figure 10: Capillary ablation via spatiotemporal SVD filtering. (A) Representative ULM [PITH_FULL_IMAGE:figures/full_fig_p025_10.png]

**Figure 11.** Figure 11: Impact of motion on representative ULM maps. (A) Ground truth representa [PITH_FULL_IMAGE:figures/full_fig_p026_11.png]

**Figure 12.** Figure 12: Functional ultrasound localization microscopy reveals simulated neurovascular [PITH_FULL_IMAGE:figures/full_fig_p027_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 35 canonical work pages

[1]

S. S. Martin, A. W. Aday, Z. I. Almarzooq, C. A. Anderson, P. Arora, C. L. Avery, C. M. Baker-Smith, B. B. Gibbs, A. Z. Beaton, A. K. Boehme, Y. Commodore-Mensah, M. E. Currie, M. S. Elkind, K. R. Evenson, G. Generoso, D. G. Heard, S. Hiremath, M. C. Johansen, R. Kalani, D. S. Kazi, D. Ko, J. Liu, J. W. Magnani, E. D. Michos, M. E. Mussolino, S. D. Navane...

work page 2024
[2]

L. B. Eisenmenger, A. Peret, B. M. Famakin, A. Spahic, G. S. Roberts, J. H. Bockholt, K. M. Johnson, J. S. Paulsen, Vascular contributions to alzheimer’s disease, Translational Research 254 (2023) 41–53

work page 2023
[3]

R. A. NISHIMURA, F. A. MILLER Jr, M. J. CALLAHAN, R. C. BE- NASSI, J. B. SEWARD, A. J. TAJIK, Doppler echocardiography: the- ory, instrumentation, technique, and application, in: Mayo Clinic Pro- ceedings, Vol. 60, Elsevier, 1985, pp. 321–343

work page 1985
[4]

O.Couture, V.Hingot, B.Heiles, P.Muleki-Seya, M.Tanter, Ultrasound localization microscopy and super-resolution: A state of the art, IEEE transactions on ultrasonics, ferroelectrics, and frequency control 65 (8) (2018) 1304–1320

work page 2018
[5]

Hingot, A

V. Hingot, A. Chavignon, B. Heiles, O. Couture, Measuring image reso- lution in ultrasound localization microscopy, IEEE transactions on med- ical imaging 40 (12) (2021) 3812–3819

work page 2021
[6]

J. Yan, B. Huang, J. Tonko, M. Toulemonde, J. Hansen-Shearer, Q. Tan, K. Riemer, K. Ntagiantas, R. A. Chowdhury, P. D. Lambiase, et al., Transthoracic ultrasound localization microscopy of myocardial vascu- lature in patients, Nature biomedical engineering 8 (6) (2024) 689–700

work page 2024
[7]

Denis, S

L. Denis, S. Bodard, V. Hingot, A. Chavignon, J. Battaglia, G. Renault, F. Lager, A. Aissani, O. Hélénon, J.-M. Correas, et al., Sensing ultra- sound localization microscopy for the visualization of glomeruli in living rats and humans, EBioMedicine 91 (2023)

work page 2023
[8]

S. A. Niederer, J. Lumens, N. A. Trayanova, Computational models in cardiology, Nature reviews cardiology 16 (2) (2019) 100–111

work page 2019

Show all 39 references

[9]

Heiles, A

B. Heiles, A. Chavignon, V. Hingot, P. Lopez, E. Teston, O. Couture, Performance benchmarking of microbubble-localization algorithms for ultrasoundlocalizationmicroscopy, NatureBiomedicalEngineering6(5) (2022) 605–616. 32

2022
[10]

Lerendegui, K

M. Lerendegui, K. Riemer, B. Wang, C. Dunsby, M.-X. Tang, Bubble flow field: a simulation framework for evaluating ultrasound localization microscopy algorithms, arXiv preprint arXiv:2211.00754 (2022)

2022 arXiv
[11]

Blanken, B

N. Blanken, B. Heiles, A. Kuliesh, M. Versuis, K. Jain, D. Maresca, G. Lajoinie, Proteus: A physically realistic contrast-enhanced ultra- sound simulator—part i: Numerical methods, IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control (2024)

2024
[12]

Heiles, N

B. Heiles, N. Blanken, A. Kuliesh, M. Versuis, K. Jain, G. La- joinie, D. Maresca, Proteus: A physically realistic contrast-enhanced ultrasound simulator—part ii: Imaging applications, IEEE Transac- tions on Ultrasonics, Ferroelectrics, and Frequency Control (2025) 1– 1doi:10....

2025
[13]

Stevenson, M

C. Stevenson, M. Brookes, J. D. López, L. Troebinger, J. Mattout, W. Penny, P. Morris, A. Hillebrand, R. Henson, G. Barnes, Does func- tion fit structure? a ground truth for non-invasive neuroimaging, Neu- roImage 94 (2014) 89–95

2014
[14]

Belgharbi, J

H. Belgharbi, J. Porée, R. Damseh, V. Perrot, L. Milecki, P. Delafontaine-Martel, F. Lesage, J. Provost, An anatomically realistic simulation framework for 3d ultrasound localization microscopy, IEEE Open Journal of Ultrasonics, Ferroelectrics, and Frequency Control 3 (2023) 1–13

2023
[15]

Garcia, F

D. Garcia, F. Varray, Simus3: An open-source simulator for 3-d ultra- sound imaging, Computer Methods and Programs in Biomedicine 250 (2024) 108169

2024
[16]

B. E. Treeby, B. T. Cox, k-wave: Matlab toolbox for the simulation and reconstruction of photoacoustic wave fields, Journal of biomedical optics 15 (2) (2010) 021314–021314

2010
[17]

G. Pinton, A fullwave model of the nonlinear wave equation with mul- tiple relaxations and relaxing perfectly matched layers for high-order numerical finite-difference solutions, arXiv preprint arXiv:2106.11476 (2021)

2021 arXiv
[18]

Linninger, G

A. Linninger, G. Hartung, S. Badr, R. Morley, Mathematical synthesis of the cortical circulation for the whole mouse brain-part i. theory and 33 image integration, Computers in biology and medicine 110 (2019) 265– 275

2019
[19]

J. C. Schwarz, M. G. van Lier, J. P. van den Wijngaard, M. Siebes, E. VanBavel, Topologic and hemodynamic characteristics of the human coronary arterial circulation, Frontiers in physiology 10 (2020) 1611

2020
[20]

M. C. Baruch, D. E. Warburton, S. S. Bredin, A. Cote, D. W. Gerdt, C. M. Adkins, Pulse decomposition analysis of the digital arterial pulse during hemorrhage simulation, Nonlinear biomedical physics 5 (2011) 1–15

2011
[21]

W. Wang, B. Jüttler, D. Zheng, Y. Liu, Computation of rota- tion minimizing frames, ACM Trans. Graph. 27 (1) (Mar. 2008). doi:10.1145/1330511.1330513

2008
[22]

Flesch, M

M. Flesch, M. Pernot, J. Provost, G. Ferin, A. Nguyen-Dinh, M. Tanter, T. Deffieux, 4d in vivo ultrafast ultrasound imaging using a row-column addressed matrix and coherently-compounded orthogonal plane waves, Physics in Medicine & Biology 62 (11) (2017) 4571

2017
[23]

C. E. Morton, G. R. Lockwood, Theoretical assessment of a crossed electrode2-darrayfor3-dimaging, in: IEEESymposiumonUltrasonics, 2003, Vol. 1, IEEE, 2003, pp. 968–971

2003
[24]

B. H. Ophir Klein, Richard A. Spritz, Micro-ct images of adult mouse skulls, collaborative cross nod x nod., FaceBase Consortium (2016)

2016
[25]

Leconte, J

A. Leconte, J. Porée, B. Rauby, A. Wu, N. Ghigo, P. Xing, S. Lee, C. Bourquin, G. Ramos-Palacios, A. F. Sadikot, et al., A tracking prior to localization workflow for ultrasound localization microscopy, IEEE Transactions on Medical Imaging (2024)

2024
[26]

H. W. Kuhn, The hungarian method for the assignment problem, Naval Research Logistics (NRL) 52 (1) (2004) 7–21

2004
[27]

Wu, J.Porée, G.Ramos-Palacios, N.Ghigo, C

A. Wu, J.Porée, G.Ramos-Palacios, N.Ghigo, C. Bourquin, A.Leconte, P. Xing, A. F. Sadikot, M. Chassé, J. Provost, 3d transcranial dy- namic ultrasound localization microscopy in the mouse brain using a row-column array, IEEE Transactions on Biomedical Engineering (2025) 1–12do...

2025
[28]

Ghigo, G

N. Ghigo, G. Ramos-Palacios, C. Bourquin, P. Xing, A. Wu, N. Cortés, H. Ladret, L. Ikan, C. Casanova, J. Porée, et al., Dynamic ultrasound localization microscopy without ecg-gating, Ultrasound in Medicine & Biology 50 (9) (2024) 1436–1448

2024
[29]

Renaudin, C

N. Renaudin, C. Demené, A. Dizeux, N. Ialy-Radio, S. Pezet, M. Tan- ter, Functional ultrasound localization microscopy reveals brain-wide neurovascular activity on a microscopic scale, Nature methods 19 (8) (2022) 1004–1012

2022
[30]

Wang, S.-L

Q. Wang, S.-L. Ding, Y. Li, J. Royall, D. Feng, P. Lesnar, N. Graddis, M. Naeemi, B. Facer, A. Ho, et al., The allen mouse brain common coordinate framework: a 3d reference atlas, Cell 181 (4) (2020) 936– 953

2020
[31]

E. M. Hillman, Coupling mechanism and significance of the bold signal: a status report, Annual review of neuroscience 37 (1) (2014) 161–181

2014
[32]

V. J. Srinivasan, D. N. Atochin, H. Radhakrishnan, J. Y. Jiang, S. Ru- vinskaya, W. Wu, S. Barry, A. E. Cable, C. Ayata, P. L. Huang, et al., Optical coherence tomography for the quantitative study of cerebrovas- cular physiology, Journal of Cerebral Blood Flow & Metabolism 31...

2011
[33]

R. B. Buxton, K. Uludağ, D. J. Dubowitz, T. T. Liu, Modeling the hemodynamic response to brain activation, Neu- roImage 23 (2004) S220–S233, mathematics in Brain Imaging. doi:https://doi.org/10.1016/j.neuroimage.2004.07.013

2004 doi
[34]

Cormier, J

P. Cormier, J. Porée, C. Bourquin, J. Provost, Dynamic myocardial ul- trasound localization angiography, IEEE Transactions on Medical Imag- ing 40 (12) (2021) 3379–3388. doi:10.1109/TMI.2021.3086115

2021
[35]

M. R. Lowerison, C. Huang, Y. Kim, F. Lucien, S. Chen, P. Song, In vivo confocal imaging of fluorescently labeled microbubbles: Implications for ultrasound localization microscopy, IEEE transactions on ultrasonics, ferroelectrics, and frequency control 67 (9) (2020) 1811–1819

2020
[36]

Zhang, M

N. Zhang, M. B. Nguyen, L. Mertens, D. J. Barron, O. Villemain, J. Baranger, Improving coronary ultrafast doppler angiography using 35 fractional moving blood volume and motion-adaptive ensemble length, Physics in Medicine & Biology 67 (12) (2022) 125021. doi:10.1088/1361- 656...

2022 doi
[37]

Zhuang, H

D.-Y. Zhuang, H. Huang, W.-T. Chang, C.-C. Huang, In vivo ultrasound dynamic coronary blood flow imaging through adap- tive frame selection method, IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control 72 (8) (2025) 1108–1118. doi:10.1109/TUFFC.2025.3582154

2025
[38]

Heiles, F

B. Heiles, F. Nelissen, R. Waasdorp, D. Terwiel, B. M. Park, E. M. Ibarra, A. Matalliotakis, T. Ara, P. Barturen-Larrea, M. Duan, et al., Nonlinear sound-sheet microscopy: Imaging opaque organs at the cap- illary and cellular scale, Science 388 (6742) (2025) eads1325

2025
[39]

J. A. Jensen, Simulation of advanced ultrasound systems using field ii, in: 2004 2nd IEEE International Symposium on Biomedical Imaging: Nano to Macro (IEEE Cat No. 04EX821), IEEE, 2004, pp. 636–639. 36

2004

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [1]

S. S. Martin, A. W. Aday, Z. I. Almarzooq, C. A. Anderson, P. Arora, C. L. Avery, C. M. Baker-Smith, B. B. Gibbs, A. Z. Beaton, A. K. Boehme, Y. Commodore-Mensah, M. E. Currie, M. S. Elkind, K. R. Evenson, G. Generoso, D. G. Heard, S. Hiremath, M. C. Johansen, R. Kalani, D. S. Kazi, D. Ko, J. Liu, J. W. Magnani, E. D. Michos, M. E. Mussolino, S. D. Navane...

work page 2024

[2] [2]

L. B. Eisenmenger, A. Peret, B. M. Famakin, A. Spahic, G. S. Roberts, J. H. Bockholt, K. M. Johnson, J. S. Paulsen, Vascular contributions to alzheimer’s disease, Translational Research 254 (2023) 41–53

work page 2023

[3] [3]

R. A. NISHIMURA, F. A. MILLER Jr, M. J. CALLAHAN, R. C. BE- NASSI, J. B. SEWARD, A. J. TAJIK, Doppler echocardiography: the- ory, instrumentation, technique, and application, in: Mayo Clinic Pro- ceedings, Vol. 60, Elsevier, 1985, pp. 321–343

work page 1985

[4] [4]

O.Couture, V.Hingot, B.Heiles, P.Muleki-Seya, M.Tanter, Ultrasound localization microscopy and super-resolution: A state of the art, IEEE transactions on ultrasonics, ferroelectrics, and frequency control 65 (8) (2018) 1304–1320

work page 2018

[5] [5]

Hingot, A

V. Hingot, A. Chavignon, B. Heiles, O. Couture, Measuring image reso- lution in ultrasound localization microscopy, IEEE transactions on med- ical imaging 40 (12) (2021) 3812–3819

work page 2021

[6] [6]

J. Yan, B. Huang, J. Tonko, M. Toulemonde, J. Hansen-Shearer, Q. Tan, K. Riemer, K. Ntagiantas, R. A. Chowdhury, P. D. Lambiase, et al., Transthoracic ultrasound localization microscopy of myocardial vascu- lature in patients, Nature biomedical engineering 8 (6) (2024) 689–700

work page 2024

[7] [7]

Denis, S

L. Denis, S. Bodard, V. Hingot, A. Chavignon, J. Battaglia, G. Renault, F. Lager, A. Aissani, O. Hélénon, J.-M. Correas, et al., Sensing ultra- sound localization microscopy for the visualization of glomeruli in living rats and humans, EBioMedicine 91 (2023)

work page 2023

[8] [8]

S. A. Niederer, J. Lumens, N. A. Trayanova, Computational models in cardiology, Nature reviews cardiology 16 (2) (2019) 100–111

work page 2019