Pith. sign in

REVIEW 4 major objections 5 minor 92 references

Accelerated Patient-Specific Calibration via Differentiable Hemodynamics Simulations

T0 review · 4 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read A differentiable 1D-0D hemodynamics solver enables gradient-based inference of patient-specific vessel and Windkessel parameters from pressure data, with forward predictions matching an established solver within 1% error.

desk verdict A genuinely useful differentiable 1D hemodynamics solver with solid forward validation, but the calibration claims rest on synthetic self-consistency and an unexamined transient-optimization shortcut. read the letter →

arxiv 2412.14572 v1 pith:XE5BVGME submitted 2024-12-19 physics.med-ph cs.LGcs.MSphysics.comp-phq-bio.QM

Diego Renner , Georgios Kissas This is my paper · ORCID
classification physics.med-phcs.LGcs.MSphysics.comp-phq-bio.QM MSC 65M0876Z0592C35 PACS 47.63.Cb87.19.Uv
keywords differentiablehemodynamics1DNavier-Stokesreduced-ordermodelpatient-specificcalibrationWindkesselparameterinferenceJAXautomaticdifferentiationMUSCLfinitevolumeschemeHamiltonianMonteCarlo
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 reduced-order (0D-1D) model of blood flow can be made fully differentiable without sacrificing the interpretability of the underlying physics, and that this differentiability turns patient-specific calibration into fast optimization. The solver, written in the JAX library, computes pressure waveforms through a finite-volume discretization of the 1D Navier-Stokes equations and at the same time returns exact gradients of those waveforms with respect to vessel lengths, stiffnesses, resistances, and compliances. With those gradients, the paper recovers Windkessel parameters from synthetic pressure measurements by gradient descent in about a minute and by Hamiltonian Monte Carlo in a few minutes, and it validates the solver's forward outputs against the open-source openBF code on four anatomies, finding less than 1% relative error and a speedup that approaches two times on the largest network. If correct, this gives a middle path between slow gradient-free calibration and uninterpretable deep-learning surrogates: fast, physics-grounded inference of parameters that cannot be measured noninvasively, with sensitivity analysis available from the same gradients.

What carries the argument

The load-bearing object is the differentiable solver itself, which is a finite-volume discretization of the dimension-reduced Navier-Stokes equations for a compliant tube: conservation laws for cross-sectional area $A$ and flow rate $Q$, closed by the tube-law pressure relation $P = P_{\mathrm{ext}} + \beta(\sqrt{A/A_0}-1)$ with elasticity coefficient $\beta$. A MUSCL scheme with total-variation-diminishing slope limiting and Lax-Friedrichs fluxes advances the hyperbolic system; inlet values come from Riemann-invariant extrapolation of given pressure or flow data, and outlets are coupled to three-element Windkessel (RCR) models whose roots are found by a frozen Newton method. Because the whole chain is implemented in JAX with padding and masking to vectorize across vessels, automatic differentiation yields gradients of any simulated quantity with respect to any parameter, and it is these gradients that both the Adafactor and the Hamiltonian Monte Carlo inference procedures consume.

What would settle it

Repeat the four-parameter resistance inference on the same bifurcation geometry, but this time iterate the solver to the paper's own convergence criterion (pressure changing by less than 0.1 mmHg between cardiac cycles) before taking gradients, and compare the recovered $R_1$ and $R_2$ values with those obtained from the truncated unconverged window; if any parameter differs by more than the optimizer's tolerance, the accelerated protocol is fitting a transient that the converged periodic state would contradict.

Watch

Extended reading notes

Core claim

Stated on the paper's own terms, the discovery is that the full pipeline — cross-sectionally averaged 1D Navier-Stokes equations, tube-law pressure closure, MUSCL finite-volume time stepping, Riemann-invariant boundary conditions, and Windkessel outlets solved by Newton's method — can be written in JAX so that automatic differentiation flows through every step. The paper demonstrates two consequences of this differentiability. First, calibration: minimizing a normalized pressure error with the Adafactor optimizer recovers a single terminal resistance close to its ground-truth value within about 500 iterations (under a minute), and recovers four Windkessel resistances of a two-outlet bifurcation within 1000 iterations (under two minutes), while Hamiltonian Monte Carlo yields a posterior concentrated at the same values after 500 sampling steps in less than four minutes. Second, fidelity: pressure waveforms from jaxFlowSim overlap those of openBF on aortic, abdominal, cerebral, and full-body ADAN56 anatomies with relative errors between $2.5 \times 10^{-4}$ and $5.5 \times 10^{-3}$, with the JAX implementation nearly twice as fast on the 77-segment network.

Load-bearing premise

The optimization experiments stop the solver before the flow has settled into its repeating heartbeat cycle, and the claim of fast calibration assumes that gradients computed from that truncated, not-yet-converged window still point toward the parameters of the true periodic state.

Editorial extensions

If this is right

  • Windkessel resistances and compliances — parameters that cannot be measured noninvasively — can be recovered from pressure data by gradient descent in roughly a minute for a single parameter and under two minutes for four parameters.
  • Hamiltonian Monte Carlo inference over four parameters is close to ground truth after 500 sampling steps, completing in less than four minutes, which makes posterior uncertainty estimates practical in a calibration setting.
  • The solver's forward predictions agree with openBF to within 1% relative error on networks from 9 to 77 vessels, and wall-clock time scales almost linearly with vessel count while compile time stays near constant.
  • Sensitivity analysis of any simulated biomarker with respect to vessel and outlet parameters becomes a single backward pass, which the paper positions as a step toward deciding which patient-specific measurements are informative.
  • JAX's device-agnostic execution means the same solver can run batched on a GPU, so many patient models could be calibrated in parallel.

Reading between the lines

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

  • The differentiability the paper builds for parameter calibration also invites learning the tube law itself from data: the authors note that the pressure–area relation is empirical and contested, and the solver-in-a-loop idea they sketch would let pressure measurements select among or refine such constitutive laws.
  • The authors deliberately optimize on simulations that have not reached periodic steady state; a natural extension is to test how few cardiac cycles suffice for reliable gradients, since each skipped cycle multiplies the calibration speedup.
  • The junction solves, flagged as not readily GPU-parallelizable, are the likely scaling bottleneck; reformulating all junctions as one sparse system is the concrete step that would let GPU batching of many patients pay off.
  • Gradients of pressure with respect to measurement location could guide experimental design: placing sensors where gradient norms are largest would yield the most informative data for identifying RCR parameters.
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

4 major / 5 minor

Summary. The paper presents jaxFlowSim, a differentiable 0D-1D reduced-order hemodynamics solver implemented in JAX using a MUSCL finite-volume scheme, Windkessel outlet models, and junction treatments. The forward solver is validated against the open-source openBF solver on four vascular anatomies, with relative L1 pressure errors below 1%. Wall-clock timings show speedups over openBF for anatomies with more than roughly 15 vessel segments. The authors then use gradients through the solver to perform deterministic (Adafactor) and probabilistic (Hamiltonian Monte Carlo) inference of Windkessel RCR parameters on a single-bifurcation synthetic geometry, reporting convergence to known ground-truth values within one to four minutes.

Significance. If the parameter-inference claims are substantiated, the differentiable solver would be a genuinely useful tool for interpretable, gradient-based calibration of reduced-order hemodynamic models. The forward validation against an independent solver is a concrete strength, as is the engineering achievement of near-constant compile time and roughly linear compute scaling in the number of vessel segments. The current inference experiments, however, are synthetic self-consistency checks on truncated transients with no comparison to gradient-free baselines, so the central 'accelerated patient-specific calibration' claim is not yet established by the evidence presented.

major comments (4)
  1. [§3.5.1, Tables 5–7] The loss function L(ξ) defined in §3.5.1 is a normalized sum of squared errors and is therefore nonnegative, yet all loss values in Tables 5, 6, and 7 are negative (e.g., −5.77, −10.25, −13.41). This inconsistency makes the reported convergence behavior uninterpretable from the stated objective. Please either correct the reported values or clarify that a different objective, such as a negative log-likelihood, is being tabulated, and define that objective explicitly.
  2. [§3.5.1, paragraph before Fig. 9] The optimization is deliberately run on non-converged, truncated simulations: the text states 'we simulated for time-steps despite our simulation not having converged yet.' The paper provides no argument or experiment showing that gradients computed from this transient window point toward the parameters that reproduce the converged periodic pressure waveform. Because patient calibration ultimately targets the periodic steady state, the recovered R1 and R2 values could match the truncated transient while being incorrect for the actual periodic state. Please demonstrate, for the same test cases, that the parameters obtained from the truncated loss also reproduce the converged periodic solution, e.g., by re-running the forward solver with the optimized parameters until convergence and reporting the periodic-state pressure error.
  3. [§3.5, Tables 5–7] The abstract and introduction claim 'faster' and 'accelerated' parameter inference, but no comparison against gradient-free calibration methods is provided for the inverse problem. The wall-clock times in Tables 5–7 are absolute and are not contrasted with, for example, Nelder–Mead, CMA-ES, or Bayesian optimization on the same calibration task. Figure 2 compares only forward-solve timings. To support the acceleration claim, please add inverse-problem wall-clock times and iteration counts for at least one standard gradient-free baseline.
  4. [§3.5.2, paragraph after Fig. 10] The HMC section states 'We set N = 1000, implying that only the last time step that we simulated is used in our loss function.' This contradicts the earlier usage of N as the number of simulated time steps and the statement in §3.5.1 that N = 1000 means all simulated time steps are used in the loss. Please clarify which time steps enter the HMC likelihood and correct the inconsistency; the current text leaves the exact likelihood ambiguous and hampers reproducibility of the posterior results.
minor comments (5)
  1. [§3.5.1, first case] The text says measurements were evaluated at M = 2 locations 'at the end of each outlet vessel,' but Figure 9 is described as showing pressure in the middle of the first outlet vessel. Please clarify the measurement locations and match them to the plotted waveforms.
  2. [§3.5.1, initialization] The initialization is described as ξ0 = {2 × log(1 + e^{s0}) × ξ̂}, and the table column is labeled 'scaling factors' but lists values that equal softplus(s0), not s0 itself. Please define and label the tabulated quantity consistently.
  3. [§2.2, Eq. (10)] The density ρ is written as a function of z and t in Eq. (10), although the reduction assumed blood density to be constant. Please use a constant ρ or explain the retained dependence.
  4. [§3.1, Windkessel parameter distribution] The parallel-circuit distribution uses ξ_i = Σ_j A_{0,j}/A_{0,i}, but the derivation is not shown. Please provide a brief explanation of why resistance is distributed inversely with reference cross-sectional area and how the parallel-circuit rule is applied.
  5. [§3.4, Figure 7] The text calls the 19-vessel geometry an outlier in the compute-time scaling, but no quantitative justification is given. Please either report the timings for all four geometries or explain the source of the apparent outlier.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the forward solver is validated against an external solver, and the parameter-inference experiments are synthetic self-consistency checks rather than derivationally circular predictions.

full rationale

The forward model is derived from the Navier-Stokes equations and the tube law, and its numerical implementation is checked against an independent solver: 'we compared it to the openBF 1D hemodynamic solver [82] written and validated by Melis et al. [59]' and report 'a smaller than 1% relative error difference between openBF and the proposed implementation, which validates our implementation.' This gives the forward predictions external support. The parameter-inference sections use synthetic ground truth generated by the same solver (for example, 'The parameter set with which the ground truth was computed is ξ̂ = {6.8123 × 10^7}'), so those experiments demonstrate self-consistency of the optimization pipeline rather than clinical predictive validity. That is an evidentiary limitation, not a circular reduction: the optimized parameters are obtained from pressure residuals, not from the known parameter values, and recovering the known values from pressure data is a genuine inverse-problem check. The admitted use of unconverged transients ('To increase the efficiency of the optimization process, we simulated for time-steps despite our simulation not having converged yet') is an unsupported assumption about gradient alignment with the periodic steady state, but it is not a definitional or self-citational circularity. Self-citations [10]–[12] are background references and are not load-bearing for the solver derivation, validation, or inference claims. No step in the derivation chain equates its conclusion to its inputs by construction.

Assumptions & free parameters 2 free parameters · 8 assumptions · 0 invented entities

The central claim rests on the standard 1D hemodynamics model with an elastic tube law, a plug-like velocity profile, Windkessel outlet conditions, and empirical vessel property formulas. The forward solver is benchmarked externally against openBF, but the parameter-inference demonstrations use synthetic data produced by the same solver and depend on hand-chosen configuration values. No new physical entities are introduced.

free parameters (2)
  • velocity profile exponent gamma = 9 (with alpha set to 1)
    Chosen as a plug-like profile from references [44-47], not inferred per patient; affects the momentum flux and wall shear term in Equation 16.
  • Windkessel nominal configuration = Pmean=120 mmHg, Qcar=80 ml/s, Ctot=1e-8 m3/Pa, R1:R2=9:91
    Fixed by hand in Section 3.1 to set outlet RCR conditions for VMR anatomies; pressure waveforms in Figures 3-5 depend on these values.
assumptions (8)
  • domain assumption Blood is incompressible, Newtonian, and its properties do not vary over a cross section.
    Assumption 1 in Section 2.2 used to reduce the Navier-Stokes equations to the 1D model; standard in 1D hemodynamics but an idealization of blood rheology.
  • domain assumption Vessel walls displace only radially and by small amounts (epsilon << 1), with no longitudinal displacement.
    Assumptions 2-4 in Section 2.2; needed to obtain the averaged continuity and momentum equations.
  • domain assumption Pressure is given by the linear elastic tube law P = Pext + beta(sqrt(A/A0) - 1) with incompressible wall (nu = 1/2).
    Equations 11-15 combine Young-Laplace, Hooke's law, and incompressibility; this closure is standard but not valid for all vessels or diseases.
  • domain assumption Plug-like velocity profile with gamma = 9 and Coriolis coefficient alpha = 1.
    Section 2.2 fixes gamma=9 and alpha=1 based on literature [44-47]; affects the momentum flux and friction terms.
  • domain assumption Empirical wall thickness and Young's modulus formulas from Blanco et al. and Ottesen et al.
    Section 3.1 uses h0 = r0(a exp(br0)+c exp(dr0)) and E = (r0/h0)(k1 exp(k2 r0)+k3) with fixed coefficients to generate VMR configurations; these are population-level fits, not patient-specific.
  • ad hoc to paper Windkessel outlet resistances are distributed using total resistance and compliance nominal values with a 9:91 R1:R2 split.
    Section 3.1 sets Pmean=120 mmHg, Qcar=80 ml/s, Ctot=1e-8 m3/Pa and splits R1/R2 9:91 following SimVascular docs; the VMR pressure predictions depend on these choices.
  • ad hoc to paper Gradients from a non-converged, truncated simulation are informative for the converged periodic steady-state parameters.
    Section 3.5.1 explicitly runs optimization before the solver has converged; this assumption is load-bearing for the reported fast calibration and is not analyzed.
  • ad hoc to paper HMC priors s0 ~ N(-0.5, 0.1) and uniform draws for the softplus parameterization are reasonable.
    Section 3.5.2 sets prior distributions for the scaling factors; these affect the posterior and are not justified by clinical data.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Accelerated Patient-Specific Calibration via Differentiable Hemodynamics Simulations." pith.science (2026). https://pith.science/paper/XE5BVGME

@misc{pith2026241214572,
  author       = {Pith},
  title        = {Pith review of: Accelerated Patient-Specific Calibration via Differentiable Hemodynamics Simulations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XE5BVGME}},
  note         = {Machine review of arXiv:2412.14572}
}
read the original abstract

One of the goals of personalized medicine is to tailor diagnostics to individual patients. Diagnostics are performed in practice by measuring quantities, called biomarkers, that indicate the existence and progress of a disease. In common cardiovascular diseases, such as hypertension, biomarkers that are closely related to the clinical representation of a patient can be predicted using computational models. Personalizing computational models translates to considering patient-specific flow conditions, for example, the compliance of blood vessels that cannot be a priori known and quantities such as the patient geometry that can be measured using imaging. Therefore, a patient is identified by a set of measurable and nonmeasurable parameters needed to well-define a computational model; else, the computational model is not personalized, meaning it is prone to large prediction errors. Therefore, to personalize a computational model, sufficient information needs to be extracted from the data. The current methods by which this is done are either inefficient, due to relying on slow-converging optimization methods, or hard to interpret, due to using `black box` deep-learning algorithms. We propose a personalized diagnostic procedure based on a differentiable 0D-1D Navier-Stokes reduced order model solver and fast parameter inference methods that take advantage of gradients through the solver. By providing a faster method for performing parameter inference and sensitivity analysis through differentiability while maintaining the interpretability of well-understood mathematical models and numerical methods, the best of both worlds is combined. The performance of the proposed solver is validated against a well-established process on different geometries, and different parameter inference processes are successfully performed.

Figures

Figures reproduced from arXiv: 2412.14572 by the authors.

Figure 1
Figure 1. We provide comparisons between the pressure obtained from the openBF and the proposed solver for arteries from different anatomies. The plots show that for each model there is barely any difference in the output of the two vascular simulations. Furthermore, there is a smaller than 1% relative error difference between openBF and the proposed implementation, which validates our implementation; see [PITH_FULL_IMAGE:fi… view at source ↗
Figure 2
Figure 2. We present a comparison between the execution wall-clock time of openBF and jaxFlowSim in seconds. 3.3 Results for Different Anatomies In this section, we present the pressure waves provided by our implementation on the four different anatomies dis￾cussed before. The results of the first model of the VMR library, the aorta model, can be seen in [PITH_FULL_IMAGE:figures/full_fig_p016_2.png] view at source ↗
Figure 3
Figure 3. We present the jaxFlowSim results for different arteries of the Aorta anatomy. 16 [PITH_FULL_IMAGE:figures/full_fig_p016_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: We present the jaxFlowSim results for different arteries of the Abdominal Aorta anatomy. Legend Vessel Name 1 aorta 2 celiac trunk 3 celiac branch 4 superior mesentric artery 5 left renal artery 6 right renal artery 7 left common iliac artery 8 left internal iliac arte…
Figure 5
Figure 5. Figure 5: We present the jaxFlowSim results for different arteries of the Cerebral anatomy. Legend Vessel Name 1 right vertebral artery 2 right posterior meningeal branch of vertebral artery 3 left posterior meningeal branch of vertebral artery 4 left vertebral artery 5 right an…
Figure 6
Figure 6. Figure 6: We present the jaxFlowSim results for different arteries of the ADAN56 anatomy. This sketch is adjusted from Boil et. el. [89]. for large anatomies. The compilation and computation time are presented against the number of segmented vessels in [PITH_FULL_IMAGE:figures/…
Figure 8
Figure 8. Figure 8: A sketch of the geometry considered for the parameter inference. 3.5.1 Deterministic Inference We assume the relation Pξ = H(ξ), where H is the PDE operator mapping the parameter set ξ to the solution for the pressure Pξ. We also introduce the pressure values produced …
Figure 9
Figure 9. Figure 9: Optimization of the R1,1 parameter with 0, 100, 500, and1000 iterations from top left to bottom right. The Adafactor optimization already shows improvement after 100 iterations. Hence, the ground truth and the learned pressure wave overlap. corresponding losses and wal…
Figure 10
Figure 10. Figure 10: Optimization of the R1,1, R2,1, R1,2, R2,2 parameter with 0, 100, 500, and1000 iterations from top left to bottom right. The Adafactor optimization reaches a high accuracy after 1000 iterations. in 6. As can be seen from the plots, the learned waveform first overshoot…
Figure 11
Figure 11. Figure 11: Optimization of the R1,1, R2,1, R1,2, R2,2 parameter with 10 warm-up and 10, 100, 500, 1000 sampling steps from top left to bottom right. The Numpyro optimization reaches a high accuracy after 500 iterations. ing losses and wall-clock times are recorded in 7. As can b…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

92 extracted references · 77 canonical work pages

  1. [1]

    Heart disease and stroke statis tics–2015 update

    AHA Statistical Update. Heart disease and stroke statis tics–2015 update. Circulation, 131:e29–e322, 2015

    2015

  2. [2]

    Eurostat Database, pages 1–2

    Edviges Coelho. Eurostat Database, pages 1–2. Springer International Publishing, 2020

    2020

  3. [3]

    Towards precision medicine

    Euan A Ashley. Towards precision medicine. Nature Reviews Genetics, 17(9):507–522, 2016

    2016

  4. [4]

    Toward precision medic ine: building a knowledge network for biomedical research and a new taxonomy of disease

    National Research Council et al. Toward precision medic ine: building a knowledge network for biomedical research and a new taxonomy of disease. National Academic Press, 2011

    2011

  5. [5]

    Role of imaging in the era of precision medicine

    Angela Giardino, Supriya Gupta, Emmi Olson, Karla Sepul veda, Leon Lenchik, Jana Ivanidze, Rebecca Rakow- Penner, Midhir J Patel, Rathan M Subramaniam, and Dhakshina moorthy Ganeshan. Role of imaging in the era of precision medicine. Academic radiology, 24(5):639–649, 2017

    2017

  6. [6]

    Integrating multiple clinical tests to in crease predictive power

    Harry B Burke. Integrating multiple clinical tests to in crease predictive power. Tumor Marker Protocols, pages 3–10, 1998

    1998

  7. [7]

    4d flow mri

    Michael Markl, Alex Frydrychowicz, Sebastian Kozerke, Mike Hope, and Oliver Wieben. 4d flow mri. Journal of Magnetic Resonance Imaging , 36(5):1015–1036, 2012

    2012

  8. [8]

    Beyond the placental bed: placent al and systemic determinants of the uterine artery doppler waveform

    TR Everett and CC Lees. Beyond the placental bed: placent al and systemic determinants of the uterine artery doppler waveform. Placenta, 33(11):893–901, 2012

    2012

Show all 92 references
  1. [9]

    Adverse cerebral events detected after subarachn oid hemorrhage using brain oxygen and microdialysis probes

    Rupert Kett-White, Peter J Hutchinson, Pippa G Al-Rawi, Arun K Gupta, John D Pickard, and Peter J Kirk- patrick. Adverse cerebral events detected after subarachn oid hemorrhage using brain oxygen and microdialysis probes. Neurosurgery, 50(6):1213–1222, 2002

    2002

  2. [10]

    Fuse: Fast unified simulation and estimation for pdes

    Levi E Lingsch, Dana Grund, Siddhartha Mishra, and Geor gios Kissas. Fuse: Fast unified simulation and estimation for pdes. arXiv preprint arXiv:2405.14558 , 2024

    2024 arXiv

  3. [11]

    T owards Digital Twins for Cardiovascular Flows: A Hybrid Machine Learning and Computa- tional Fluid Dynamics Approach

    Georgios Kissas. T owards Digital Twins for Cardiovascular Flows: A Hybrid Machine Learning and Computa- tional Fluid Dynamics Approach . PhD thesis, University of Pennsylvania, 2023. 24

    2023

  4. [12]

    Feasibility of vascular parameter estimation for assessing hypertensive pregnancy disorders

    Georgios Kissas, Eileen Hwuang, Elizabeth W Thompson, Nadav Schwartz, John A Detre, Walter R Witschey, and Paris Perdikaris. Feasibility of vascular parameter estimation for assessing hypertensive pregnancy disorders. Journal of Biomechanical Engineering , 144(12):121011, 2022

    2022

  5. [13]

    Bayesian win dkessel calibration using optimized 0d surrogate models

    Jakob Richter, Jonas Nitzler, Luca Pegolotti, Karthik Menon, Jonas Biehler, Wolfgang A Wall, Daniele E Schi- avazzi, Alison L Marsden, and Martin R Pfaller. Bayesian win dkessel calibration using optimized 0d surrogate models. arXiv preprint arXiv:2404.14187 , 2024

    2024 arXiv

  6. [14]

    The fr ontier of simulation-based inference

    Kyle Cranmer, Johann Brehmer, and Gilles Louppe. The fr ontier of simulation-based inference. Proceedings of the National Academy of Sciences , 117(48):30055–30062, 2020

    2020

  7. [15]

    Simulation-based inference for cardiova scular models

    Antoine Wehenkel, Jens Behrmann, Andrew C Miller, Guil lermo Sapiro, Ozan Sener, Marco Cuturi, and Jörn- Henrik Jacobsen. Simulation-based inference for cardiova scular models. arXiv preprint arXiv:2307.13918 , 2023

    2023 arXiv

  8. [16]

    Addressing misspecification in simulatio n-based inference through data-driven calibration

    Antoine Wehenkel, Juan L Gamella, Ozan Sener, Jens Behr mann, Guillermo Sapiro, Marco Cuturi, and Jörn- Henrik Jacobsen. Addressing misspecification in simulatio n-based inference through data-driven calibration. arXiv preprint arXiv:2405.08719 , 2024

    2024 arXiv

  9. [17]

    Inverse problems in blood flow modeling: A review

    David Nolte and Cristobal Bertoglio. Inverse problems in blood flow modeling: A review. International journal for numerical methods in biomedical engineering , 38(8):e3613, 2022

    2022

  10. [18]

    Infinite number of solutions to the hemo- dynamic inverse problem

    Christopher M Quick, William L Y oung, and Abraham Noord ergraaf. Infinite number of solutions to the hemo- dynamic inverse problem. American Journal of Physiology-Heart and Circulatory Phys iology, 280(4):H1472– H1479, 2001

    2001

  11. [19]

    Patient-specific mode ling of cardiovascular mechanics

    Charles A Taylor and CA Figueroa. Patient-specific mode ling of cardiovascular mechanics. Annual review of biomedical engineering, 11:109–134, 2009

    2009

  12. [20]

    Parameter estimation in a cardiovascul ar computational model using numerical optimization: Patient simulation, searching for a digital twin, 2022

    Giulia Tuccio. Parameter estimation in a cardiovascul ar computational model using numerical optimization: Patient simulation, searching for a digital twin, 2022

    2022

  13. [21]

    Optimization in cardiovascular mode ling

    Alison L Marsden. Optimization in cardiovascular mode ling. Annual review of fluid mechanics , 46:519–546, 2014

    2014

  14. [22]

    Optimization framework for patient-specific car diac modeling

    Joshua Mineroff, Andrew D McCulloch, David Krummen, Ba skar Ganapathysubramanian, and Adarsh Krish- namurthy. Optimization framework for patient-specific car diac modeling. Cardiovascular engineering and technology, 10:553–567, 2019

    2019

  15. [23]

    Patient-specific modelling and param- eter optimisation to simulate dilated cardiomyopathy in ch ildren

    Selim Bozkurt, Waleed Paracha, Kaushiki Bakaya, and Si lvia Schievano. Patient-specific modelling and param- eter optimisation to simulate dilated cardiomyopathy in ch ildren. Cardiovascular Engineering and T echnology, 13(5):712–724, 2022

    2022

  16. [24]

    Multiscale models of the vascular system

    Luca Formaggia, Alfio Quarteroni, and Alessandro V enez iani. Multiscale models of the vascular system. Car- diovascular Mathematics: Modeling and simulation of the ci rculatory system, pages 395–446, 2009

    2009

  17. [25]

    Geometric multiscale modeling of the cardiovas- cular system, between theory and practice

    Alfio Quarteroni, Alessandro V eneziani, and Christian V ergara. Geometric multiscale modeling of the cardiovas- cular system, between theory and practice. Computer Methods in Applied Mechanics and Engineering , 302:193– 252, 2016

    2016

  18. [26]

    P14 blood flow analysis of the aortic arch using computational fluid dynamics in a cou pled 3d-0d framework, 2020

    Scott Black, Konstantinos Ritos, Craig Maclean, Robbi e Brodie, and Asimina Kazakidi. P14 blood flow analysis of the aortic arch using computational fluid dynamics in a cou pled 3d-0d framework, 2020

    2020

  19. [27]

    Investigating the effects of a penetrating vessel occlusion with a multi- scale microvasculature model of the human cerebral cortex

    Wahbi K El-Bouri and Stephen J Payne. Investigating the effects of a penetrating vessel occlusion with a multi- scale microvasculature model of the human cerebral cortex. NeuroImage, 172:94–106, 2018

    2018

  20. [28]

    Numerical simulation of blood flow and p ressure drop in the pulmonary arterial and venous circulation

    M Umar Qureshi, Gareth DA V aughan, Christopher Sainsbu ry, Martin Johnson, Charles S Peskin, Mette S Olufsen, and NA Hill. Numerical simulation of blood flow and p ressure drop in the pulmonary arterial and venous circulation. Biomechanics and modeling in mechanobiology , 13(5...

    2014

  21. [29]

    V as- cular graph model to simulate the cerebral blood flow in reali stic vascular networks

    Johannes Reichold, Marco Stampanoni, Anna Lena Keller , Alfred Buck, Patrick Jenny, and Bruno Weber. V as- cular graph model to simulate the cerebral blood flow in reali stic vascular networks. Journal of Cerebral Blood Flow & Metabolism , 29(8):1429–1443, 2009

    2009

  22. [30]

    JAX: composable transfor- mations of Python+NumPy programs, 2018

    James Bradbury, Roy Frostig, Peter Hawkins, Matthew Ja mes Johnson, Chris Leary, Dougal Maclaurin, George Necula, Adam Paszke, Jake V anderPlas, Skye Wanderman-Milne, and Qiao Zhang. JAX: composable transfor- mations of Python+NumPy programs, 2018

    2018

  23. [31]

    Machine learning for cardio- vascular biomechanics modeling: challenges and beyond

    Amirhossein Arzani, Jian-Xun Wang, Michael S Sacks, an d Shawn C Shadden. Machine learning for cardio- vascular biomechanics modeling: challenges and beyond. Annals of Biomedical Engineering , 50(6):615–627, 2022. 25

    2022

  24. [32]

    Fundamentals of Aerodynamics (SI units)

    John Anderson. Fundamentals of Aerodynamics (SI units) . McGraw hill, 2011

    2011

  25. [33]

    An elastic tube theory of pulse transmission and oscillator y flow in mammalian arteries

    John R Womersley. An elastic tube theory of pulse transmission and oscillator y flow in mammalian arteries . Number 614. Wright Air Development Center, 1957

    1957

  26. [34]

    Mathematical analysis o f the quasilinear effects in a hyperbolic model blood flow through compliant axi-symmetric vessels

    Suncica Canic and Eun Heui Kim. Mathematical analysis o f the quasilinear effects in a hyperbolic model blood flow through compliant axi-symmetric vessels. Mathematical Methods in the Applied Sciences , 26:1161 – 1186, 09 2003

    2003

  27. [35]

    Dimension Reduced Modeling of Blood Flow in Large Arteries: An Introduc- tion for Master Students and First Y ear Doctoral Students

    Tobias Köppl and Rainer Helmig. Dimension Reduced Modeling of Blood Flow in Large Arteries: An Introduc- tion for Master Students and First Y ear Doctoral Students. Springer Nature, 2023

    2023

  28. [36]

    A theory o f fluid flow in compliant tubes

    ACL Barnard, W A Hunt, WP Timlake, and E V arley. A theory o f fluid flow in compliant tubes. Biophysical journal, 6(6):717–724, 1966

    1966

  29. [37]

    Reduced modelling of blood flow in the cerebral circulation: coupling 1-d, 0-d and cerebral auto-regulation models

    Jordi Alastruey, SM Moore, KH Parker, T David, J Peiró, a nd SJ Sherwin. Reduced modelling of blood flow in the cerebral circulation: coupling 1-d, 0-d and cerebral auto-regulation models. International journal for numerical methods in fluids , 56(8):1061–1067, 2008

    2008

  30. [38]

    Analysis of blood flow in one dim ensional elastic artery using navier-stokes conserva- tion laws

    Cristóbal Rodero Gómez. Analysis of blood flow in one dim ensional elastic artery using navier-stokes conserva- tion laws. arXiv preprint arXiv:1710.06258 , 2017

    2017 arXiv

  31. [39]

    Traite de Mecanique celeste, volume 1

    Pierre Simon comte de Laplace. Traite de Mecanique celeste, volume 1. Crapelet, 1799

  32. [40]

    An essay on the cohesion of fluids

    Y oung Thomas. An essay on the cohesion of fluids. Philos. Trans. R. Soc. London , 95:65–87, 1805

  33. [41]

    R. Hooke. Lectures de Potentia Restitutiva, Or of Spring Explaining t he Power of Springing Bodies . Cutlerian lecture. John Martyn, 1678

  34. [42]

    One-dimensional modelling of a vascular network in space-time variables

    SJ Sherwin, V Franke, J Peiró, and K20389821200 Parker. One-dimensional modelling of a vascular network in space-time variables. Journal of engineering mathematics , 47:217–250, 2003

    2003

  35. [43]

    Computational modelling of 1d blood flow with variable mechanical properties and its application to the simulation of wave propagation in the human arterial system

    Spencer J Sherwin, Luca Formaggia, Joaquim Peiro, and V Franke. Computational modelling of 1d blood flow with variable mechanical properties and its application to the simulation of wave propagation in the human arterial system. International journal for numerical methods in flu...

    2003

  36. [44]

    Numerical simulation of arterial blood flow

    Peter Hunter. Numerical simulation of arterial blood flow. PhD thesis, ResearchSpace@ Auckland, 1972

    1972

  37. [45]

    Generation of an anatomically based geometric coronary model

    Nicolas P Smith, AJ Pullan, and PJ Hunter. Generation of an anatomically based geometric coronary model. Annals of biomedical engineering , 28:14–25, 2000

    2000

  38. [46]

    An anatomically based model of transient coronary blood flow in the heart

    NP Smith, AJ Pullan, and Peter J Hunter. An anatomically based model of transient coronary blood flow in the heart. SIAM Journal on Applied mathematics , 62(3):990–1018, 2002

    2002

  39. [47]

    Cardiovascular Mathematics: Modeling and simulation of the circulatory system , volume 1

    Luca Formaggia, Alfio Quarteroni, and Allesandro V enez iani. Cardiovascular Mathematics: Modeling and simulation of the circulatory system , volume 1. Springer Science & Business Media, 2010

    2010

  40. [48]

    Numerical schemes f or unsteady fluid flow through collapsible tubes

    D Elad, D Katz, E Kimmel, and S Einav. Numerical schemes f or unsteady fluid flow through collapsible tubes. Journal of biomedical engineering , 13(1):10–18, 1991

    1991

  41. [49]

    Numerical mode lling of the pressure wave propagation in the arterial flow

    Giuseppe Pontrelli and Enrico Rossoni. Numerical mode lling of the pressure wave propagation in the arterial flow. International journal for numerical methods in fluids , 43(6-7):651–671, 2003

    2003

  42. [50]

    V alidation of a one-dimensional model of the systemic arterial tree

    Philippe Reymond, Fabrice Merenda, Fabienne Perren, D aniel Rufenacht, and Nikos Stergiopulos. V alidation of a one-dimensional model of the systemic arterial tree. American Journal of Physiology-Heart and Circulatory Physiology, 297(1):H208–H222, 2009

    2009

  43. [51]

    Puls atile pressure and flow through distensible vessels

    Victor L Streeter, W Ford Keitzer, and David F Bohr. Puls atile pressure and flow through distensible vessels. Circulation Research, 13(1):3–20, 1963

    1963

  44. [52]

    The non-linearities of arterial blood flow

    WE Bodley. The non-linearities of arterial blood flow. Physics in Medicine & Biology , 16(4):663, 1971

    1971

  45. [53]

    Forward and backward running waves in the arteries: analysis using the method of characteristics

    Kim H Parker and CJH Jones. Forward and backward running waves in the arteries: analysis using the method of characteristics. Journal of Biomechanical Engineering , 1990

    1990

  46. [54]

    Wave propagation in a model of the a rterial circulation

    JJ Wang and KH Parker. Wave propagation in a model of the a rterial circulation. Journal of biomechanics , 37(4):457–470, 2004

    2004

  47. [55]

    Time-domain representation of ventricular-arterial coupling as a windkessel and wave s ystem

    Jiun-Jr Wang, Aoife B O’Brien, Nigel G Shrive, Kim H Park er, and John V Tyberg. Time-domain representation of ventricular-arterial coupling as a windkessel and wave s ystem. American Journal of Physiology-Heart and Circulatory Physiology, 284(4):H1358–H1368, 2003

    2003

  48. [56]

    Shock-capturing methods for free-surface shallow flows

    Eleuterio Francisco Toro et al. Shock-capturing methods for free-surface shallow flows . Wiley and Sons Ltd., 2001. 26

    2001

  49. [57]

    Efficient implementati on of essentially non-oscillatory shock-capturing schemes

    Chi-Wang Shu and Stanley Osher. Efficient implementati on of essentially non-oscillatory shock-capturing schemes. Journal of computational physics , 77(2):439–471, 1988

    1988

  50. [58]

    Uniformly high order accurate essentially non-oscillatory schemes, III

    Ami Harten, Bjorn Engquist, Stanley Osher, and Sukumar R Chakravarthy. Uniformly high order accurate essentially non-oscillatory schemes, III . Springer, 1997

    1997

  51. [59]

    Gaussian process emulators for 1D vascular models

    Alessandro Melis. Gaussian process emulators for 1D vascular models . PhD thesis, University of Sheffield, 2017

    2017

  52. [60]

    Numerical solution s for unsteady gravity-driven flows in collapsible tubes: evolution and roll-wave instability of a steady stat e

    BS Brook, SAEG Falle, and TJ Pedley. Numerical solution s for unsteady gravity-driven flows in collapsible tubes: evolution and roll-wave instability of a steady stat e. Journal of Fluid Mechanics , 396:223–256, 1999

    1999

  53. [61]

    A model for time-dependent flow in (giraffe jugular) veins: uniform tube properties

    BS Brook and TJ Pedley. A model for time-dependent flow in (giraffe jugular) veins: uniform tube properties. Journal of biomechanics, 35(1):95–107, 2002

    2002

  54. [62]

    On the coupling of 3d and 1d navier–stokes equations for flow problems in compliant vess els

    Luca Formaggia, Jean-Frédéric Gerbeau, Fabio Nobile, and Alfio Quarteroni. On the coupling of 3d and 1d navier–stokes equations for flow problems in compliant vess els. Computer methods in applied mechanics and engineering, 191(6-7):561–582, 2001

    2001

  55. [63]

    A one-dimensional finite element method for simulat ion-based medical planning for cardiovascular disease

    Jing Wan, Brooke Steele, Sean A Spicer, Sven Strohband, Gonzalo R Feijo´ o, Thomas JR Hughes, and Charles A Taylor. A one-dimensional finite element method for simulat ion-based medical planning for cardiovascular disease. Computer Methods in Biomechanics & Biomedical Engineeri...

    2002

  56. [64]

    A finite-element model of blood flow in arteries including taper, branches, and obstructions

    G Porenta, DF Y oung, and TR Rogge. A finite-element model of blood flow in arteries including taper, branches, and obstructions. Journal of Biomechanical Engineering , 1986

    1986

  57. [65]

    A finite-element sim ulation of pulsatile flow in flexible obstructed tubes

    Elkana Rooz, DF Y oung, and TR Rogge. A finite-element sim ulation of pulsatile flow in flexible obstructed tubes. Journal of Biomechanical Engineering , 1982

    1982

  58. [66]

    Experimental validation of a time-domain-based wave propagation model of blood flow i n viscoelastic vessels

    David Bessems, Christina G Giannopapa, Marcel CM Rutte n, and Frans N van de V osse. Experimental validation of a time-domain-based wave propagation model of blood flow i n viscoelastic vessels. Journal of biomechanics, 41(2):284–291, 2008

    2008

  59. [67]

    W ave propagation in fluids: models and numerical techniques

    Vincent Guinot. W ave propagation in fluids: models and numerical techniques . John Wiley & Sons, 2012

    2012

  60. [68]

    Towards the ultimate conservative diffe rence scheme

    Bram V an Leer. Towards the ultimate conservative diffe rence scheme. iv. a new approach to numerical convec- tion. Journal of computational physics , 23(3):276–299, 1977

    1977

  61. [69]

    Towards the ultimate conservative diffe rence scheme

    Bram V an Leer. Towards the ultimate conservative diffe rence scheme. v. a second-order sequel to godunov’s method. Journal of computational Physics , 32(1):101–136, 1979

    1979

  62. [70]

    The piecewise para bolic method (ppm) for gas-dynamical simulations

    Phillip Colella and Paul R Woodward. The piecewise para bolic method (ppm) for gas-dynamical simulations. Journal of computational physics , 54(1):174–201, 1984

    1984

  63. [71]

    Numerical methods for conservation laws , volume 214

    Randall J LeV eque and Randall J Leveque. Numerical methods for conservation laws , volume 214. Springer, 1992

    1992

  64. [72]

    Über die F ortpflanzung ebener Luftwellen von endlicher Schw ingungsweite, volume 8

    Bernhard Riemann. Über die F ortpflanzung ebener Luftwellen von endlicher Schw ingungsweite, volume 8. V er- lag der Dieterichschen Buchhandlung, 1860

  65. [73]

    The method of characteristics with appli cations to conservation laws

    Scott A Sarra. The method of characteristics with appli cations to conservation laws. Journal of Online mathe- matics and its Applications , 3:1–16, 2003

    2003

  66. [74]

    Lumped parameter out flow models for 1-d blood flow simulations: Effect on pulse waves and parameter estimation

    KH Parker, J Peiró, and SJ Sherwin. Lumped parameter out flow models for 1-d blood flow simulations: Effect on pulse waves and parameter estimation. Communications in Computational Physics , 4(2):317–336, 2008

    2008

  67. [75]

    The W ashington manual of surgery

    Mary E Klingensmith et al. The W ashington manual of surgery. Lippincott Williams & Wilkins, 2008

    2008

  68. [76]

    An introduction to numerical analysis

    Kendall Atkinson. An introduction to numerical analysis . John wiley & sons, 1991

    1991

  69. [77]

    On a class of frozen regularized gauss-newt on methods for nonlinear inverse problems

    Qinian Jin. On a class of frozen regularized gauss-newt on methods for nonlinear inverse problems. Mathematics of Computation, 79(272):2191–2211, 2010

    2010

  70. [78]

    Maximum efficiency for a family of newton-like methods with frozen derivatives and some appli cations

    Sergio Amat, Sonia Busquier, Àngela Grau, and Miquel Gr au-Sánchez. Maximum efficiency for a family of newton-like methods with frozen derivatives and some appli cations. Applied Mathematics and Computation , 219(15):7954–7963, 2013

    2013

  71. [79]

    On two high-order families of frozen newton-type methods

    Sergio Amat, I Argyros, Sonia Busquier, and MA Hernánde z-V erón. On two high-order families of frozen newton-type methods. Numerical Linear Algebra with Applications , 25(1):e2126, 2018

    2018

  72. [80]

    One-dimensional models for blood flow in arteries

    Luca Formaggia, Daniele Lamponi, and Alfio Maria Quarte roni. One-dimensional models for blood flow in arteries. Journal of Engineering Mathematics , 47:251–276, 2003

    2003

  73. [81]

    V ascular model repository, 2023

    SimV ascular. V ascular model repository, 2023. 27

    2023

  74. [82]

    A. Melis. openbf, 2023

    2023

  75. [83]

    An anatomically detailed arterial network model for one-dime nsional computational hemodynamics

    Pablo J Blanco, Sansuke M Watanabe, Marco Aurélio RF Pas sos, Pedro A Lemos, and Raúl A Feijóo. An anatomically detailed arterial network model for one-dime nsional computational hemodynamics. IEEE Trans- actions on biomedical engineering , 62(2):736–753, 2014

    2014

  76. [84]

    Applied mathematical models in human physiology

    Johnny T Ottesen, Mette S Olufsen, and Jesper K Larsen. Applied mathematical models in human physiology . SIAM, 2004

    2004

  77. [85]

    Simvascular, 2023

    SimV ascular. Simvascular, 2023

    2023

  78. [86]

    Julia: A fresh approach to numerical com- puting

    Jeff Bezanson, Alan Edelman, Stefan Karpinski, and Vir al B Shah. Julia: A fresh approach to numerical com- puting. SIAM Review, 59(1):65–98, 2017

    2017

  79. [87]

    Aortic input impedance in normal man: relationship to pressure wave forms

    JOSEPH P Murgo, NICO Westerhof, JOHN P Giolma, and STEPH EN A Altobelli. Aortic input impedance in normal man: relationship to pressure wave forms. Circulation, 62(1):105–116, 1980

    1980

  80. [88]

    Blood flow distribution in an anatomically detailed arterial netw ork model: criteria and algorithms

    Pablo J Blanco, Sansuke M Watanabe, Enzo A Dari, Marco Au rélio RF Passos, and Raúl A Feijóo. Blood flow distribution in an anatomically detailed arterial netw ork model: criteria and algorithms. Biomechanics and modeling in mechanobiology , 13:1303–1330, 2014

    2014

  81. [89]

    Blanco, Lucas O

    Etienne Boileau, Perumal Nithiarasu, Pablo J. Blanco, Lucas O. Müller, Fredrik Eikeland Fossan, Leif Rune Hellevik, Wouter P . Donders, Wouter Huberts, Marie Willeme t, and Jordi Alastruey. A benchmark study of numerical schemes for one-dimensional arterial blood flow m odellin...

    2015

  82. [90]

    Adafactor: Adaptive l earning rates with sublinear memory cost

    Noam Shazeer and Mitchell Stern. Adafactor: Adaptive l earning rates with sublinear memory cost. In Interna- tional Conference on Machine Learning , pages 4596–4604. PMLR, 2018

    2018

  83. [91]

    The deepmind jax ecosystem, 2020

    I Babuschkin DeepMind, K Baumli, A Bell, S Bhupatiraju, J Bruce, P Buchlovsky, D Budden, T Cai, A Clark, I Danihelka, et al. The deepmind jax ecosystem, 2020

    2020

  84. [92]

    Just After eXecution

    Du Phan, Neeraj Pradhan, and Martin Jankowiak. Composa ble effects for flexible and accelerated probabilistic programming in numpyro. arXiv preprint arXiv:1912.11554 , 2019. A Implementation Since differentiability was a priority in the implementati on of this model, the JAX li...

    1912 arXiv

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.