List of Papers Browse by Subject Areas Author List
Abstract
Cortical visual prostheses aim to restore sight by electrically stimulating early visual cortex (V1). As high-density and flexible neural interfaces emerge, electrode placement within folded cortex becomes a critical surgical planning problem. Existing strategies emphasize visual-field coverage and anatomical heuristics, but do not directly optimize predicted perceptual outcomes under safety constraints. We present a percept-aware framework that formulates cortical electrode placement as constrained optimization in anatomical space. Electrode coordinates are treated as learnable parameters and optimized end-to-end using a differentiable model of prosthetic vision. The objective minimizes task-level perceptual error while incorporating vascular avoidance and gray matter feasibility. Using realistic folded cortical geometry (FreeSurfer fsaverage), we evaluate the method on simulated reading and natural-image tasks. Percept-aware optimization improves reconstruction fidelity relative to visual-field tiling and improves or matches coverage-based placement depending on task structure. Vascular constraints eliminate safety-margin violations with only modest effects on perceptual performance, and the framework supports co-optimization of multi-electrode thread configurations under fixed insertion budgets. These results show how differentiable percept models can support anatomically grounded, safety-aware planning for cortical visual prostheses.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026/paper/3855_paper.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to the Code Repository
N/A
Link to the Dataset(s)
N/A
BibTex
@InProceedings{PogGal_PerceptAware_MICCAI2026,
author = { Pogoncheff, Galen AND Wang, Alvin AND Granley, Jacob AND Beyeler, Michael},
title = { { Percept-Aware Surgical Planning for Visual Cortical Prostheses with Vascular Avoidance } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 16893},
month = {September},
page = {pending}
}
Reviews
Review #1
- Please describe the contribution of the paper
This paper introduces a percept-aware optimization framework for surgical planning of cortical visual prostheses, focusing on electrode placement under anatomical constraints. The authors propose a systematic methodology and compare it with existing approaches, demonstrating partial improvements across different settings and datasets.
- 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 topic is timely and relevant, and well aligned with the scope of miccai. The paper addresses a clearly defined problem and proposes a structured methodological solution. The approach is novel and sufficiently developed for a conference contribution, and the method is generally clear and reproducible. The manuscript is also well written and easy to follow overall, with a clear presentation of the main ideas.
- 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 authors give a reasonable explanation for not using learned perceptual metrics, because phosphene images are substantially different from normal images. However, a similar concern may also apply to the use of resnet 50 for task-level evaluation, since this model is also trained on natural images. The paper should explain more clearly why this evaluation is still appropriate in this context.
-
The claims about performance improvements are not fully consistent with the reported results. The manuscript states that percept-aware optimization “consistently reduced reconstruction error relative to both visual field tiling and visual field coverage. ” However, for cifar 10, the results show that performance is only comparable to coverage-based optimization, and in some cases does not outperform it. This inconsistency between the general claim and the reported results should be clarified or the wording should be made more precise.
-
The claim that perceptual quality “remained largely unchanged” after introducing safety constraints is not sufficiently supported by statistical analysis. While small percentage decreases in SSIM are reported, it isn’t clear whether these differences are statistically significant. Providing appropriate statistical tests would strengthen this claim.
-
Some aspects of the figures could also be clarified. In Fig. 2 (left), it is not clear why coverage-based results are not included in the visual comparison. In addition, the interpretation of the metrics (MSE and SSIM) is not immediately clear from the figure, as the direction of improvement (e. g. , lower is better for MSE, higher is better for SSIM) is not explicitly indicated. This makes the comparison somewhat difficult to interpret.
-
In Section 4.3, the description of “threaded arrays” is unclear, as the concept is not sufficiently explained earlier in the manuscript. A brief definition or reference to the device design would help the reader understand what is being optimized. In addition, methodological aspects such as the optimization setup (e. g. , the choice of N_insert = 128) appear in the results section without prior explanation, which makes the structure difficult to follow. These design choices should be introduced and justified in the methods section, while the results section should focus on reporting outcomes.
-
There are also inconsistencies between the Results and Discussion sections. The Discussion states that percept-aware placement “consistently improved reconstruction fidelity relative to coverage-based strategies,” whereas the Results indicate that performance relative to coverage-based optimization depends on the dataset (e. g. , comparable performance on CIFAR-10 rather than consistent improvement). In addition, tiling-based baselines are not mentioned in the Discussion, despite being part of the main comparison in the Results. The Discussion should more accurately reflect the reported findings.
-
- 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.
- Based on your review and your understanding of the MICCAI Scientific Code of Ethics, do you believe this submission may involve a potential ethics concern or violation?
N/A
- 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/2026/en/REVIEWER-GUIDELINES.html
The separation between Results and Discussion could be improved. In Sections 4.2 and 4.3, interpretative statements and methodological explanations are mixed with the presentation of results, which makes these sections difficult to follow. At the same time, the Discussion section remains relatively high-level and does not fully engage with the detailed findings. A clearer distinction between reporting results and interpreting them would improve the overall structure and readability.
- 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 — marginally below the acceptance threshold, but would not mind if accepted, dependent on rebuttal
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
While the paper presents a relevant problem and a promising methodological contribution that is close to the acceptance threshold, several issues regarding evaluation justification, consistency of claims, and lack of statistical support reduce confidence in the results. Imho, these concerns appear addressable but may require revisions beyond what is feasible within the current review process.
- 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
The authors present a differentiable surgical planning framework for visual prosthetics that optimizes electrode placement by balancing anatomical safety with predicted perceptual outcomes.
The framework employs a two-stage differentiable forward model to map electrode coordinates to visual percepts:
Spatial Mapping: It utilizes retinotopic mapping and electrode trajectories to determine visual field coordinates.
Phosphene Simulation: It estimates phosphene generation at those coordinates using image-sampled amplitude signals and isotropic Gaussian spreads.
The optimization process simultaneously accounts for surgical constraints, specifically grey matter coverage and vasculature avoidance. Validated in a simulated framework using the Freesurfer fsaverage atlas, this method outperformed traditional placement strategies across multiple array configurations and simulated phosphene spreads.
- 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.
Although dense, the article is well-organized and introduces a novel framework for optimizing visual implant electrode placement. The proposed approach maximizes perceptual estimation while simultaneously integrating surgical constraints into a single, scalable model. Notably, the study demonstrates that the inclusion of anatomical safety requirements does not significantly diminish predicted perceptual performance.
- 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.
As the author stated in the discussion, the major weak point is related to the experiments, limited to the fsaverage and not presenting any subject-specific validation. Nonetheless, I found the article quite interesting.
The are some minor issues I’ve identified along the articles:
- In the introduction, the following phrase is repeated “Prior work has optimized stimulation for fixed implants or arranged electrodes to maximize visual field coverage [7, 17], but perceptual objectives and surgical constraints have not been jointly optimized within a unified planning framework.” in two following paragraphs;
- Formula (1) -> what is c_j?
- In section 3.4, the formula for L_vasc is not correctly expressed (there is an if inside the Summatory operator)
- 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.
- Based on your review and your understanding of the MICCAI Scientific Code of Ethics, do you believe this submission may involve a potential ethics concern or violation?
N/A
- 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/2026/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?
I found the article very interesting, with a strong bibliography and providing an interesting advance, even if the validation is still quite limited. The methodology is clear, reproducible and leaves space for improvements in the future, both in methodology and validation.
- 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
Review #3
- Please describe the contribution of the paper
It proposes a surgical planning framework for cortical visual prostheses that jointly optimizes electrode placement for perceptual quality and anatomical safety — specifically by incorporating grey matter feasibility and vascular avoidance as penalty terms within a differentiable optimization framework.
- 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 main strength of this paper is the novel framework that jointly optimizes electrode placement for both perceptual quality and anatomical safety constraints. Previous work only focused on visual field coverage, so incorporating perceptual objectives together with grey matter and vascular avoidance constraints in one framework is a meaningful contribution. The results also show that adding vascular safety constraints can eliminate margin violations without hurting perceptual performance, which is a clinically relevant finding
- 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.
First, the proposed method does not significantly outperform VF Coverage on CIFAR-10 across any reported metric. The authors provide a post-hoc explanation, but this limitation was not anticipated in the experimental design. Furthermore, Fig. 2 only provides qualitative comparison against VF Tiling, leaving the more competitive VF Coverage baseline without visual comparison.
Second, the framework is evaluated entirely on simulation without any real patient data. We acknowledge this is common practice in the field and consistent with MICCAI norms, but the lack of any validation against real implant data limits the ability to assess clinical relevance.
Third, it is unclear how the optimized electrode coordinates would translate into practical surgical guidance. The paper does not discuss how surgeons would use this framework intraoperatively, which is an important gap for a work framed as a surgical planning tool.
- 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.
- Based on your review and your understanding of the MICCAI Scientific Code of Ethics, do you believe this submission may involve a potential ethics concern or violation?
N/A
- 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/2026/en/REVIEWER-GUIDELINES.html
A minor: On page 2, the paragraph starting with “Computational modeling has shown that…..” has the sentence that is completely replicated in the paragraph above. This is the sentence I refer to “Prior work has either optimized stimulation for fixed implants or arranged electrodes to maximize visual field coverage [7, 17]. However, perceptual objectives and surgical constraints have not been jointly optimized within a unified planning framework.”
- 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 — marginally above the acceptance threshold, but would not mind if rejected, dependent on rebuttal
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
This paper presents a meaningful contribution by proposing a unified framework that jointly optimizes perceptual quality and anatomical safety for cortical visual prosthesis surgical planning. However, the evaluation has some weaknesses that limit confidence in the results: the method does not significantly outperform the strongest baseline on natural images, and the qualitative comparisons do not include this baseline. The lack of real patient validation is understandable given the nature of the work, but the clinical applicability of the framework could be better justified. These issues do not invalidate the contribution but suggest the paper would benefit from a stronger rebuttal addressing these points
- 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
Author Feedback
N/A
Meta-Review
Meta-review #1
- Your recommendation
Provisional Accept
- Please justify your decision. In case you deviate from the reviewers’ recommendations, explain in detail the reasons why. In case of an invitation for rebuttal, clarify which points are important to address in the rebuttal.
The paper was evaluated by reviewers with expertise in the field who identified limitations that lead to mixed enthusiasm for this work. Questions were raised about the design of the validation study and clinical applicability. However, the reviewers recognize the proposed methodology is of interest and addresses an important clinical problem. The strengths outweigh the weaknesses. I recommend acceptance.