List of Papers By topics Author List
| Paper Info | Reviews | Meta-review | Author Feedback | Post-Rebuttal Meta-reviews |
Authors
Ashley Bruce, Michael Beyeler
Abstract
Visual neuroprostheses are the only FDA-approved technology for the treatment of retinal degenerative blindness. Although recent work has demonstrated a systematic relationship between electrode location and the shape of the elicited visual percept, this knowledge has yet to be incorporated into retinal prosthesis design, where electrodes are typically arranged on either a rectangular or hexagonal grid. Here we optimize the intraocular placement of epiretinal electrodes using dictionary learning. Importantly, the optimization process is informed by a previously established and psychophysically validated model of simulated prosthetic vision. We systematically evaluate three different electrode placement strategies across a wide range of possible phosphene shapes and recommend electrode arrangements that maximize visual subfield coverage. In the near future, our work may guide the prototyping of next-generation neuroprostheses.
Link to paper
DOI: https://link.springer.com/chapter/10.1007/978-3-031-16449-1_57
SharedIt: https://rdcu.be/cVRXv
Link to the code repository
N/A
Link to the dataset(s)
N/A
Reviews
Review #1
- Please describe the contribution of the paper
This paper presents a greedy optimization approach to selection of optimal electrode sites for a retinal prosthesis design. The approach uses a phosphene shape model to optimize the arrangement of electrodes to maximize predicted coverage of the visual field, as predicted by the model.
- Please list the main strengths of the paper; you should write about 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 paper is well written and organized. To my knowledge, this is a novel approach for optimizing retinal prosthesis design, which is an important clinical area. This approach could be broadly applicable to the design of other neuromodulators. A validation study shows potential performances advantages over state of the art design.
- Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.
The application does not involve the use of medical images like most MICCAI submissions, so it may have less broad interest across the conference. But I think the work still fits well under the computer aided intervention umbrella and is a nice example of model-guided design.
As written, the validation study lacks data showing the optimized utility function will translate to optimized image quality.
- Please rate the clarity and organization of this paper
Excellent
- Please comment on the reproducibility of the paper. Note, that authors have filled out a reproducibility checklist upon submission. Please be aware that authors are not required to meet all criteria on the checklist - for instance, providing code and data is a plus, but not a requirement for acceptance
Methods are clear and reproducible.
- Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review: https://conferences.miccai.org/2022/en/REVIEWER-GUIDELINES.html
I think the work could benefit from some examples of image “reconstructions,” i.e., it would be interesting to see examples of how the optimized electrode positions might be leveraged to create better quality image perception.
- Rate the paper on a scale of 1-8, 8 being the strongest (8-5: accept; 4-1: reject). Spreading the score helps create a distribution for decision-making
6
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
It’s a well-written, interesting study with moderate weakness as described above.
- Number of papers in your stack
5
- What is the ranking of this paper in your review stack?
2
- Reviewer confidence
Very confident
- [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed
Not Answered
- [Post rebuttal] Please justify your decision
Not Answered
Review #2
- Please describe the contribution of the paper
This paper is in a very specialized field of electrode placement for epiretinal prothesis. The paper defines the problem and proposes a disctionary learning based optimization for arrangement of such electrode for getting the desired outcome.
- Please list the main strengths of the paper; you should write about 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.
This paper’s strenght is in modeling of a clincal problem and proposing a solution in form of adapting existing methodological approaches and evaluating the results in terms of simulated visual subfield coverege.
- Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.
This paper is way to specialized for being presented at MICCAI conference. The mathematical and algorithmic approaches are known but the modeling of the problem at hand and its implementation in this filed is new, however this paper will not be appreciated in MICCAI community.
- Please rate the clarity and organization of this paper
Very Good
- Please comment on the reproducibility of the paper. Note, that authors have filled out a reproducibility checklist upon submission. Please be aware that authors are not required to meet all criteria on the checklist - for instance, providing code and data is a plus, but not a requirement for acceptance
The work here is reproducible in terms of simulation and computation.
- Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review: https://conferences.miccai.org/2022/en/REVIEWER-GUIDELINES.html
If accepted and published at MICCAI, I think that the authors will not receive the feedback and citations they would get in case they submit and publish this work in a more relevant community. As citations in this paper clearly shows, no prior work has been published in MICCAI, IPMI, IPCAI, Medical Image Analysis or TMI, which are the main conferences and journals relevant to MICCAI. I think that this work will have much more impact if published in a more specialized community.
- Rate the paper on a scale of 1-8, 8 being the strongest (8-5: accept; 4-1: reject). Spreading the score helps create a distribution for decision-making
4
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
While the paper is well written, the methods are reasonable and the numarical results are valuable, I strongly beleive that MICCAI is not the right community for publishing this work at.
- Number of papers in your stack
5
- What is the ranking of this paper in your review stack?
3
- Reviewer confidence
Somewhat Confident
- [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed
Not Answered
- [Post rebuttal] Please justify your decision
Not Answered
Review #4
- Please describe the contribution of the paper
This paper describes a novel approach to determine the distribution and number of retinal implant electrodes to maximize / optimize the resulting visual perception.
- Please list the main strengths of the paper; you should write about 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.
Nice paper that describes, by using a formulation of dictionary optimization, how to achieve an optimized electrophysiological simulation with, e.g. the Argus II implant. The methodological foundations are quite well exposed and is complemented by an analysis of the algorithm’s performance. This approach suggests that a rectilinear array of electrodes not necessarily be the optimal one when compared to the results of this approach.
- Please list the main weaknesses of the paper. Please provide details, for instance, if you think a method is not novel, explain why and provide a reference to prior work.
While the approach per se is novel, it is not clear why the authors reproduced (?) a figure of Ref. 2, or why the did never mention that Fig. 2 was calculated with the Python library in use for this study. It seems that the author only refer to one single other work. If so, this should deserve mentioning. If not, the section on related work is incomplete. While this work is exploring a rather exotic field of neurostimulation it is quite surprising that the introductory part is extremely short. It would be highly recommendable to introduce the readers to the field with its challenges and problems. Concerning the Phosphene model, some more information on this would be readily welcome. In the section H and W are used but never defined. This has to be fixed, as later on these two values pop up again; comprehension of eq. 3 suffersb dramatically. In addition, F as defined to be [0,HW] is not clear. Furhtermore, eq. 3 is not motivated or referenced clearly. In view of these shortcomings it is hard to judge the real novelty of the approach. Concerning Fig 2 an improved visualization of the values for the precentual coverage is suggested.
- Please rate the clarity and organization of this paper
Good
- Please comment on the reproducibility of the paper. Note, that authors have filled out a reproducibility checklist upon submission. Please be aware that authors are not required to meet all criteria on the checklist - for instance, providing code and data is a plus, but not a requirement for acceptance
na
- Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review: https://conferences.miccai.org/2022/en/REVIEWER-GUIDELINES.html
please see above.
- Rate the paper on a scale of 1-8, 8 being the strongest (8-5: accept; 4-1: reject). Spreading the score helps create a distribution for decision-making
6
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
The work itself seems to be good and well done, but it suffers from inconsistencies, missing reference to other basic work, negligencies and cold be improved in the visual presentation of results.
- Number of papers in your stack
6
- What is the ranking of this paper in your review stack?
2
- Reviewer confidence
Confident but not absolutely certain
- [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed
Not Answered
- [Post rebuttal] Please justify your decision
Not Answered
Primary Meta-Review
- Please provide your assessment of this work, taking into account all reviews. Summarize the key strengths and weaknesses of the paper and justify your recommendation. 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.
This manuscript describes planning of the placement of electrodes for retinal implants to maximise/optimise visual perception.
All reviewers found that the paper was clearly written, justifying the application of methods to a very novel application for the MICCAI community. It also validated this approach, demonstrating that the results improve visual subfield coverage.
However, one major concern on the applicability of this manuscript to the MICCAI community as it does not use medical images and is related to a very specific implant technology. It might be worthwhile to discuss or relate the proposed method and application to other neuromodulators and planning techniques and discuss the similarities and differences.
Reviewers also highlighted that the introduction and works cited were very limited in scope and perhaps this section could be expanded upon.
Finally, a qualitative visualisation of the results through some type of reconstruction or graphical representation would improve clarity of the paper.
- What is the ranking of this paper in your stack? Use a number between 1 (best paper in your stack) and n (worst paper in your stack of n papers). If this paper is among the bottom 30% of your stack, feel free to use NR (not ranked).
5
Author Feedback
We thank the reviewers for their detailed comments.
We agree that our work does not operate on medical images (MIC) [Reviewer 1, Reviewer 2, Meta-Reviewer]. Instead we suggest a computer-aided interventional approach (CAI) for retinal prostheses, which is why we are submitting to the CAI portion of the conference (see selected “conference category”). We believe this work is relevant to the MICCAI community, because it uses computer simulations to inform the design of retinal prostheses with the goal of optimizing visual outcomes (i.e., visual subfield coverage), and therefore falls under the CAI umbrella (as pointed out by Reviewer 1). Similar work published at MICCAI in recent years includes examples from the cochlear implant [1-2] and retinal implant [3] communities.
That being said, we agree that the manuscript should better explain how the proposed method could be used for other neuromodulators and planning techniques [Meta-Reviewer]. Our method should apply wherever there is a topological mapping from stimulus space to perception (e.g., visual, auditory, tactile stimulation). This means that our approach could be extended to other neuromodulation technologies that include (but are not limited to) other electronic prostheses and optogenetic technologies.
We agree that the Related Work section could be extended [Reviewer 4, Meta-Reviewer]. Although there is limited related work due to the novelty of our approach and the application, we should have highlighted other papers with similar methods (e.g., [1, 4]) and given more background information [5] to address clarification questions about the phosphene model [Reviewer 4].
Finally, we agree that an additional figure would be helpful, where we show a reconstruction of the artificial vision using the optimized design compared against baseline [Reviewer 1, Meta-Reviewer]. This figure could be integrated with Figure 3 that is currently showing the visual subfield coverage. Another row/column of panels might show through simulation what an Argus II patient might see when looking at an image vs. what a patient with an optimized implant might see. This would also bode well with MICCAI’s focus on image data.
The manuscript is currently ~3/4 pages short of 8 pages and we have plenty of room for additional references. If given the opportunity to submit a camera-ready version, we are confident that we can address all major concerns raised in the reviews and meta-review.
References [1] Bratu, E., Dwyer, R., Noble, J. (2020). A Graph-Based Method for Optimal Active Electrode Selection in Cochlear Implants. MICCAI 2020, doi:10.1007/978-3-030-59716-0_4 [2] Noble, J.H., Dawant, B.M. (2015). Automatic Graph-Based Localization of Cochlear Implant Electrodes in CT. MICCAI 2015, doi:10.1007/978-3-319-24571-3_19 [3] Beyeler, M., Boynton, G.M., Fine, I., Rokem, A. (2019). Model-Based Recommendations for Optimal Surgical Placement of Epiretinal Implants. MICCAI 2019, doi:10.1007/978-3-030-32254-0_44 [4] N. P. Shah et al. (2019). Optimization of Electrical Stimulation for a High-Fidelity Artificial Retina. 9th International IEEE/EMBS Conference on Neural Engineering (NER) 2019, doi:10.1109/NER.2019.8716987. [5] Luo, Y., Zhong, J., Clemo, M., Cruz L. (2016). Long-term Repeatability and Reproducibility of Phosphene Characteristics in Chronically Implanted Argus II Retinal Prosthesis Subjects. American Journal of Ophthalmology, doi:10.1016/j.ajo.2016.07.021
Post-rebuttal Meta-Reviews
Meta-review # 1 (Primary)
- Please provide your assessment of the paper taking all information into account, including rebuttal. Highlight the key strengths and weaknesses of the paper, clarify how you reconciled contrasting review comments and scores, indicate if concerns were successfully addressed in the rebuttal, and provide a clear justification of your decision. If you disagree with some of the (meta)reviewer statements, you can indicate so in your meta-review. Please make sure that the authors, program chairs, and the public can understand the reason for your decision.
The only real point of contention for this manuscript is if it was within the scope of CAI, being primarily focused on optimising the design of an implantable device. I think overall the technical concept in the paper could be applied to other types of surgical planning (not just device design) and would be of interest to members of the community. The authors indicated they will spend some time revising the manuscript to make these similarities more explicit and more clearly identify how the paper fits into the conference themes.
- After you have reviewed the rebuttal, please provide your final rating based on all reviews and the authors’ rebuttal.
Accept
- What is the rank of this paper among all your rebuttal papers? Use a number between 1/n (best paper in your stack) and n/n (worst paper in your stack of n papers). If this paper is among the bottom 30% of your stack, feel free to use NR (not ranked).
5
Meta-review #2
- Please provide your assessment of the paper taking all information into account, including rebuttal. Highlight the key strengths and weaknesses of the paper, clarify how you reconciled contrasting review comments and scores, indicate if concerns were successfully addressed in the rebuttal, and provide a clear justification of your decision. If you disagree with some of the (meta)reviewer statements, you can indicate so in your meta-review. Please make sure that the authors, program chairs, and the public can understand the reason for your decision.
It seems the major issues of whether this paper is appropriate for the MICCAI conference not with the actual paper which is well written and interesting and solves a clinical problem. I believe that the authors as they mention in their rebuttal could relate the work to other targeting tasks such as neuromodulation, etc. I do believe that the paper could be of interest to the MICCAI audience and would lean towards accepting.
- After you have reviewed the rebuttal, please provide your final rating based on all reviews and the authors’ rebuttal.
Accept
- What is the rank of this paper among all your rebuttal papers? Use a number between 1/n (best paper in your stack) and n/n (worst paper in your stack of n papers). If this paper is among the bottom 30% of your stack, feel free to use NR (not ranked).
NR
Meta-review #3
- Please provide your assessment of the paper taking all information into account, including rebuttal. Highlight the key strengths and weaknesses of the paper, clarify how you reconciled contrasting review comments and scores, indicate if concerns were successfully addressed in the rebuttal, and provide a clear justification of your decision. If you disagree with some of the (meta)reviewer statements, you can indicate so in your meta-review. Please make sure that the authors, program chairs, and the public can understand the reason for your decision.
The main strenghts of this manuscript include an interesting application that convincingly demonstrates improved subfield coverage and is clearly presented. The primary weakness of the paper is the question whether or not this paper is a good fit for presentation at MICCAI, with two related but somewhat different considerations: Would the work receive helpful feedback through presentation at the conference?, and: Would the audience be interested in this? In regards to the first question, my perception is that the authors are willing to take this bet, potentially at the cost of not receiving the attention the work could get elsewhere. Regarding the second question, personally I feel that this is not the right community for the work (and would perhaps argue that it also was not the best fit for the work referenced in the rebuttal), but it appears that reviewers and ACs of this and other work have diverging opinions. To me, this is a borderline decision and because it’s borderline not because of validity but because of “perceived fit” I will vote for acceptance.
- After you have reviewed the rebuttal, please provide your final rating based on all reviews and the authors’ rebuttal.
Accept
- What is the rank of this paper among all your rebuttal papers? Use a number between 1/n (best paper in your stack) and n/n (worst paper in your stack of n papers). If this paper is among the bottom 30% of your stack, feel free to use NR (not ranked).
9