- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes a unified framework for bidirectional CT-to-MRI and MRI-to-CT translation using shared latent dynamics. The framework employs modality-specific VAEs and a shared velocity field to model both translation directions within a single parameterization. To alleviate the performance degradation associated with parameter sharing, the authors introduce Anchor BiFlow, which constructs a source- and direction-conditioned latent anchor to guide velocity prediction. An image-space refiner is further employed to correct residual reconstruction errors. Experiments on SynthRAD2023 and PCM2025 demonstrate competitive quantitative performance and favorable computational efficiency, with only 4.15M parameters and approximately 6.26 ms inference latency.
Critical strengths
-
Modeling both CT–MRI translation directions through shared latent dynamics is well motivated and practically relevant. Evaluation on two datasets, both translation directions, and multiple MRI contrasts supports the potential utility of the framework.
-
The complete model achieves competitive translation quality with 4.15M parameters and a reported average inference latency of 6.26 ms (Appendix C.2). This represents a favorable balance between translation quality and computational cost relative to the iterative diffusion and flow baselines at the evaluated inference settings.
Critical weaknesses
-
Equation (14) directly supervises the anchor toward the target latent. With α=1, the integration endpoint is this anchor plus an accumulated residual. Without direct anchor decoding or a capacity-matched single-pass residual baseline, the incremental benefit of iterative dynamics remains unclear.
-
Table 2 supports recovery of the sharing gap before refinement, but the final advantage over Uni Flow remains unresolved. The MRI→CT PSNR gain is 0.09 dB before refinement, while adding refinement yields a further 0.43 dB. A Uni Flow baseline with the same refiner is needed to assess the advantage under comparable refinement.
-
Tables 2–3 do not define the reported ± values or specify the repetition protocol. Clarifying the independent runs, random seeds, and uncertainty would help assess the stability of the small pre-refinement gains.
Questions for authors
-
How does Anchor BiFlow compare with direct anchor decoding and a capacity-matched single-pass residual predictor under comparable supervision? These comparisons would clarify the incremental benefit of iterative dynamics.
-
Does the advantage over Uni Flow persist when both use the same VAE representations and image refiners under comparable training settings? Please report parameter counts, training budgets, and supervision settings.
-
How many independent training runs were conducted for Tables 2–3, and what do the reported ± values represent? Are the small pre-refinement gains consistent across training seeds?
3 comments
Other: We would like to thank the reviewer once again for their careful review; we will incorporate the necessary revisions and additions to address the points raised in the updated version of the manuscript.
Furthermore, we have added comparative experiments with SynDiff [1] and SelfRDB [2], as shown in the table below:
| Method | Training | Params (M) | FLOPs (G) | SynthRAD2023 CT→MRI | SynthRAD2023 MRI→CT | PCM2025 T1 CT→MRI | PCM2025 T1 MRI→CT | PCM2025 T2 CT→MRI | PCM2025 T2 MRI→CT |
|---|---|---|---|---|---|---|---|---|---|
| SelfRDB [2] | Separate | 39.73 × 2 | 141.83 × (10–20) | 18.85 / 0.7740 | 19.43 / 0.8023 | 15.92 / 0.3709 | 10.67 / 0.3251 | 13.55 / 0.3148 | 10.54 / 0.3306 |
| SynDiff [1] | Joint | 79.46 | 141.66 × 4 | 16.65 / 0.6790 | 17.68 / 0.7566 | 15.15 / 0.3431 | 10.87 / 0.3185 | 13.05 / 0.3082 | 10.05 / 0.2978 |
| Proposed | Joint | 4.15 | 14.75 + 1.61 × 5 | 19.47 / 0.7794 | 19.69 / 0.8068 | 19.43 / 0.5092 | 14.94 / 0.4023 | 17.60 / 0.4907 | 15.16 / 0.4105 |
[1] M. Özbey et al., “Unsupervised Medical Image Translation With Adversarial Diffusion Models,” IEEE Transactions on Medical Imaging, 2023.
[2] F. Arslan et al., “Self-consistent recursive diffusion bridge for medical image translation,” Medical Image Analysis, 2025.
Dear Reviewer,
I hope this message finds you well. As the discussion period is nearing its end with less than eight hours remaining, I wanted to ensure we have addressed all your concerns satisfactorily. If there are any additional points or feedback you'd like us to consider, please let us know. Your insights are invaluable to us, and we're eager to address any remaining issues to improve our work.
Thank you for your time and effort in reviewing our paper.
We thank the reviewer for their careful review. Our point-by-point responses are provided below:
Reply to Q1: For SynthRAD2023, we introduced two ablation variants to compare against our full model, Anchor BiFlow (5-step). The Anchor-only variant decodes directly from the predicted anchors. Anchor BiFlow (1-step) retains the same anchor mapper and velocity network, performing a single application of the predicted residual onto the anchors before decoding. Both the one-step and five-step models are trained using a joint bidirectional approach. Including the two VAEs, the shared anchor mapper and velocity network, and the refinement modules for both directions, both models have a total parameter count of 4.15M. During the generation of endpoints, the one-step and five-step models invoke the shared velocity network once and five times, respectively. As shown in the table below, both before and after refinement, the complete model achieves the highest average SSIM in both directions.
Reply to Q2: On SynthRAD2023, Uni Flow + Refiner and Anchor BiFlow + Refiner use the same frozen VAE and refiner architecture. Both use losses for flow matching, latent endpoints, image reconstruction, and target VAE projection consistency. Anchor BiFlow also uses an anchor loss. Uni Flow trains one model per direction for 40 epochs each, while Anchor BiFlow trains one joint model for 40 epochs. Both use five training seeds and 16 CT–MRI pairs per batch. Anchor BiFlow achieves higher mean SSIM in both directions.
Reply to Q3: For each configuration in Tables 2 and 3, we report mean ± SD over five independent runs using the same random data split and training seeds 1234, 111, 222, 333, and 444. The table below reports two-tailed paired tests comparing five-step Anchor BiFlow with each variant before refinement.
When the two directions require different velocity predictions and the source state lacks sufficient information to distinguish between them, a shared model struggles to fit both mappings effectively. Anchors conditioned on the translation direction mitigate this issue, allowing the bidirectional model to achieve average performance comparable to that of Uni Flow prior to refinement.