List of Papers Browse by Subject Areas Author List
Abstract
Traditional ultrasound simulators solve the wave equation to model pressure distribution fields, achieving physical accuracy but requiring significant computational time and resources. Ray tracing approaches have been introduced to address this limitation, modeling wave propagation as rays interacting with boundaries and scatterers. However, existing models simplify ray propagation, generating echoes at interaction points without considering return paths to the sensor. This can result in undesired artifacts and necessitates careful scene tuning for plausible results. We propose UltraRay, a novel framework that models the full path of acoustic waves reflecting from tissue boundaries. We derive the equations for accurate reflection modeling across multiple interaction points and introduce a sampling strategy for an increased likelihood of a ray returning to the transducer. By incorporating a ray emission scheme for plane wave imaging and a standard signal processing pipeline for beamforming, we are able to simulate the ultrasound image formation process end-to-end. Built on a differentiable modular framework, UltraRay introduces an extendable foundation for differentiable ultrasound simulation based on full-path ray tracing. We demonstrate its advantages compared to the state-of-the-art ray tracing ultrasound simulation, shown both on a synthetic scene and a spine phantom.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2025/paper/1969_paper.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to the Code Repository
https://github.com/Felixduelmer/UltraRay
Link to the Dataset(s)
N/A
BibTex
@InProceedings{DueFel_UltraRay_MICCAI2025,
author = { Duelmer, Felix and Azampour, Mohammad Farid and Wysocki, Magdalena and Navab, Nassir},
title = { { UltraRay: Introducing Full-Path Ray Tracing in Physics-Based Ultrasound Simulation } },
booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2025},
year = {2025},
publisher = {Springer Nature Switzerland},
volume = {LNCS 15961},
month = {September},
page = {653 -- 663}
}
Reviews
Review #1
- Please describe the contribution of the paper
The paper proposes a novel framework for simulating B-mode ultrasound (US) images using full-path ray tracing. The authors introduce a transducer sampling strategy that models the return of rays to the transducer after their interactions within the medium. Experiments are conducted on a synthetic scene and a spine phantom to validate the proposed method against real ultrasound acquisitions and a baseline simulator.
- Please list the major strengths of the paper: you should highlight a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
- The problem addressed by the authors is interesting and practically relevant, with potential implications for improving ultrasound simulation techniques.
- The qualitative results appear promising.
- Please list the major weaknesses of the paper. Please provide details: for instance, if you state that a formulation, way of using data, demonstration of clinical feasibility, or application is not novel, then you must provide specific references to prior work.
- The validation is not fully convincing, as it relies solely on qualitative assessment using only two samples. The absence of quantitative evaluation limits the strength of the evidence supporting the method’s effectiveness.
- In Fig. 4(c), the simulated result shows a shadow below the spinous process, which is not observed in the real ultrasound acquisition. The authors should discuss this discrepancy and acknowledge the limitations of their method.
- Please rate the clarity and organization of this paper
Satisfactory
- Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.
The submission has provided an anonymized link to the source code, dataset, or any other dependencies.
- Optional: If you have any additional comments to share with the authors, please provide them here. Please also refer to our Reviewer’s guide on what makes a good review and pay specific attention to the different assessment criteria for the different paper categories: https://conferences.miccai.org/2025/en/REVIEWER-GUIDELINES.html
Minor Comment: On page 7, the reference to “figure 4” in the sentence “In the synthetic scene (figure 4)…” should be corrected to “Figure 3” to match the actual figure being discussed.
- Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making.
(3) Weak Reject — could be rejected, dependent on rebuttal
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
See the weaknesses.
- Reviewer confidence
Confident but not absolutely certain (3)
- [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.
N/A
- [Post rebuttal] Please justify your final decision from above.
N/A
Review #2
- Please describe the contribution of the paper
This paper proposes “UltraRay,” a physics-based ultrasound simulation framework using Monte Carlo ray tracing that models the full acoustic path from emission to reception, incorporating boundary reflections/transmissions and an end-to-end pipeline including beamforming.
- Please list the major strengths of the paper: you should highlight a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
-
Novel Formulation (in context): While ray tracing for US exists, UltraRay incorporates more advanced physics from PBR, such as full path tracing (emission to reception explicitly modeled), microfacet BSDFs for reflections, and a specific transducer sampling strategy. Combining this with an end-to-end beamforming pipeline in a potentially differentiable framework is a novel package for US simulation.
-
Improved Realism: The full-path modeling and more sophisticated reflection model aim to produce more realistic simulations compared to simpler ray tracers that often generate echoes directly at interactions without verifying return paths, as qualitatively demonstrated.
-
End-to-End Simulation: Includes standard plane wave imaging and beamforming steps, allowing for the generation of B-mode images that are more directly comparable to real acquisitions.
-
Potential for Differentiability: Built upon Mitsuba 3, the framework has the underlying capability for differentiation, which could enable future applications in optimization or inverse problems.
-
- Please list the major weaknesses of the paper. Please provide details: for instance, if you state that a formulation, way of using data, demonstration of clinical feasibility, or application is not novel, then you must provide specific references to prior work.
-
Differentiability Not demonstrated: Although built on a differentiable framework (Mitsuba 3), the paper presents no experiments that utilize or validate this differentiability. All results are from forward simulation only. This limits its practical application potential.
-
Limited quantitative evaluation: validation is primarily qualitative, comparing generated B-mode images to real data and one baseline simulator. There’s a lack of quantitative metrics assessing simulation fidelity or comparing performance against other simulation types (e.g., wave solvers on simpler scenes). The baseline simulator is also outdated.
-
- Please rate the clarity and organization of this paper
Satisfactory
- Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.
The submission has provided an anonymized link to the source code, dataset, or any other dependencies.
- Optional: If you have any additional comments to share with the authors, please provide them here. Please also refer to our Reviewer’s guide on what makes a good review and pay specific attention to the different assessment criteria for the different paper categories: https://conferences.miccai.org/2025/en/REVIEWER-GUIDELINES.html
N/A
- Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making.
(4) Weak Accept — could be accepted, dependent on rebuttal
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
The work is novel in its specific combination of advanced ray tracing physics and end-to-end simulation for ultrasound. The qualitative results show promise for improved realism. However, the validation is limited, and critically, the paper doesn’t demonstrate the differentiability aspect which is implied by the framework choice. It’s accepted based on novelty and potential, assuming the differentiability can be leveraged in future work.
- Reviewer confidence
Confident but not absolutely certain (3)
- [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.
Accept
- [Post rebuttal] Please justify your final decision from above.
I keep my decison: weak accept. I hope the authors will release the related code upon acceptance.
Review #3
- Please describe the contribution of the paper
UltraRay is a physics-based ultrasound simulation framework built on full-path ray tracing, designed to address the high computational complexity of traditional ultrasound simulators and the artifact issues caused by simplified ray propagation in existing ray tracing methods. UltraRay models the complete path of sound waves reflected from tissue boundaries, derives the mathematical formula for multiple reflections, and introduces a sampling strategy to increase the likelihood of rays returning to the sensor. The framework combines ray emission schemes for plane wave imaging with a standard signal processing pipeline (including delay-and-sum beamforming, envelope detection, and logarithmic compression) to achieve end-to-end ultrasound image formation simulation. Experimental results show that UltraRay outperforms existing state-of-the-art ray tracing ultrasound simulation methods on synthetic scenes and spinal models, offering more accurate surface interaction simulations, and is extendable to include other interactions. Future work could focus on incorporating scattering and attenuation from existing ultrasound ray tracers, further enhancing the diversity and realism of the simulator.
- Please list the major strengths of the paper: you should highlight a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
UltraRay introduces an efficient sensor sampling strategy that ensures rays return to the sensor, thereby avoiding the generation of artifacts. By combining ray emission schemes for plane wave imaging with a standard signal processing pipeline (including delay-and-sum beamforming, envelope detection, and logarithmic compression), UltraRay is capable of end-to-end simulation of the ultrasound image formation process. Built on a differentiable modular framework, UltraRay provides a scalable foundation for differentiable ultrasound simulation based on full-path ray tracing.
- Please list the major weaknesses of the paper. Please provide details: for instance, if you state that a formulation, way of using data, demonstration of clinical feasibility, or application is not novel, then you must provide specific references to prior work.
The main limitation of the UltraRay framework lies in its balance between noise management and wide aperture angles. To improve image quality, the authors restrict the aperture angle of lens elements in the lens sampling strategy. While this helps reduce noise and enhances the physical accuracy of the images, it also limits the applicability of the simulator in certain scenarios. Additionally, the existing simulator has shortcomings in handling scattering and attenuation, which play a significant role in real ultrasound imaging.
- Please rate the clarity and organization of this paper
Good
- Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.
The submission does not mention open access to source code or data but provides a clear and detailed description of the algorithm to ensure reproducibility.
- Optional: If you have any additional comments to share with the authors, please provide them here. Please also refer to our Reviewer’s guide on what makes a good review and pay specific attention to the different assessment criteria for the different paper categories: https://conferences.miccai.org/2025/en/REVIEWER-GUIDELINES.html
N/A
- Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making.
(5) Accept — should be accepted, independent of rebuttal
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
This method is innovative in the field, and the experiments have demonstrated that it outperforms some traditional algorithms.
- Reviewer confidence
Somewhat confident (2)
- [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.
N/A
- [Post rebuttal] Please justify your final decision from above.
N/A
Author Feedback
We thank the reviewers for their constructive feedback and appreciate the recognition of our approach as innovative (R1) and novel (R2, R3), its potential for (differentiable) ultrasound simulation (R1–R3), improved realism (R2, R3), and lower computational cost compared to wave-based simulators (R1). Our main contribution is UltraRay, a scalable foundation (R1) for ultrasound simulation that uses full‑path ray tracing to reduce unrealistic artefacts (R1, R3), featuring an end-to-end signal processing pipeline that generates B-mode images directly from simulated RF pre-beamformed channel data via a standard beamforming process. We present the core concept, derive the underlying mathematical framework, and validate our approach, focusing on two representative examples that highlight the method’s capabilities. UltraRay delivers the qualitatively superior results noted by R1–R3, but we did not include numerical scores (R2, R3) because common metrics such as SSIM or PSNR presuppose pixel-level registration between simulation and measurement. Currently, free-hand acquisitions and unmodelled scatter violate this assumption, therefore, these metrics would primarily reflect mis-registration noise rather than physical fidelity. We, therefore, present here qualitative side-by-side images and compare with the latest open-source ray-tracing simulator, which operates under similar assumptions and supports a largely equivalent scene representation. Wave-based solvers operate on voxel grids or point scatterers and would first need a separate conversion and tuning stage, making a one-to-one scene comparison challenging (R3). Extending the evaluation to include such methods is an important direction for future research. The shadow underneath the spinous process (R2) results from the narrow receive-directivity function we employ: echoes are only recorded when the reflection point falls inside the reception cone defined by each element’s opening angle, a setting chosen to limit noise. If a strong reflector (e.g., bone) obstructs that path, the rays cannot reach the transducer. Real transducers operate with a broader element directivity function, so energy returning to “off-axis” elements is still captured, making partially occluded regions visible. We already noted this limitation in the discussion section (R1), but will further emphasise its impact on visibility and echo formation. In response to the comment about differentiability not being demonstrated (R3), we note that UltraRay is built on Mitsuba 3, whose kernels are automatically differentiable via autograd as documented. This makes gradient-based extensions straightforward in principle, though such experiments are beyond the scope of this conceptual and methodological introduction of our framework. We will emphasize this point in the paper. While UltraRay currently focuses on reflections and transmissions at tissue boundaries, the principles guiding the current implementation—ray tracing and transducer sampling—naturally support the incorporation of scattering within heterogeneous tissue (R1). This is a logical next step, and we will present it as such in the final version. Minor comments will be clarified in the paper, and the code will be publicly available upon acceptance (R1). Importantly, UltraRay is intended as an open-source project for the community, designed to be extensible and reproducible. By releasing the full simulator, we aim to provide a foundation for further research, allowing others to experiment with the framework, evaluate aspects such as noise behaviour, and contribute additional features such as scattering, attenuation, or inverse reconstruction workflows. We again thank the reviewers for realizing the practical relevance (R3) and future potential (R1, R2) and hope that our simulation framework, UltraRay, can be offered to the MICCAI community for future advances in ultrasound simulation and tasks that require differentiable simulation.
Meta-Review
Meta-review #1
- Your recommendation
Invite for Rebuttal
- If your recommendation is “Provisional Reject”, then summarize the factors that went into this decision. In case you deviate from the reviewers’ recommendations, explain in detail the reasons why. You do not need to provide a justification for a recommendation of “Provisional Accept” or “Invite for Rebuttal”.
The submission has received varied feedback from the three reviewers, resulting in a broad range of scores. The authors are therefore encouraged to submit a rebuttal to clarify and support their work. Particular attention should be given to the concerns raised in R1 and R2’s reviews, as well as addressing the missing implementation details highlighted by R3. Given the limited space available for the rebuttal, the authors should focus on the most essential issues identified by the reviewers to deliver a clear and concise response.
- After you have reviewed the rebuttal and updated reviews, please provide your recommendation based on all reviews and the authors’ rebuttal.
Accept
- Please justify your recommendation. You may optionally write justifications for ‘accepts’, but are expected to write a justification for ‘rejects’
N/A
Meta-review #2
- After you have reviewed the rebuttal and updated reviews, please provide your recommendation based on all reviews and the authors’ rebuttal.
Accept
- Please justify your recommendation. You may optionally write justifications for ‘accepts’, but are expected to write a justification for ‘rejects’
The main issues to be addressed focused on reproducibility and limited validation. Given that these authors have promised open-source availability of their work, these two issues are addressed.
Meta-review #3
- After you have reviewed the rebuttal and updated reviews, please provide your recommendation based on all reviews and the authors’ rebuttal.
Reject
- Please justify your recommendation. You may optionally write justifications for ‘accepts’, but are expected to write a justification for ‘rejects’
The reviewers appreciate the practical impact, technical novelty, and compelling property, such as the presumably improved realistic simulation brought by the physics-inspired reflection model. However, the main drawback of this work is that there is no quantitative evaluation at all. I agree with R2 that qualitative results are not sufficient to demonstrate the effectiveness of the proposed method.