Paper Info Reviews Author Feedback Meta-Review Back to top
List of Papers Browse by Subject Areas Author List

Abstract

The peeling of an epiretinal membrane (ERM) is a complex procedure wherein a membrane, only a few micrometers thick, that develops on the retinal surface is delicately removed using microsurgical forceps. Insights regarding small gaps between the ERM and the retinal tissue are valuable for surgical decision-making, particularly in determining a suitable location to initiate the peeling. Depth-resolved imaging of the retina provided by intraoperative Optical Coherence Tomography (iOCT) enables visualization of this gap and supports decision-making. The common presentation of iOCT images during surgery in juxtaposition with the microscope view however requires surgeons to move their gaze from the surgical site, affecting proprioception and cognitive load. In this work, we introduce an alternative method utilizing auditory feedback as a sensory channel, designed to intuitively enhance the perception of ERM elevations. Our approach establishes an unsupervised innovative mapping between real-time OCT A-scans and the parameters of an acoustic model. This acoustic model conveys the physical characteristics of tissue structure through distinctive sound textures, at microtemporal resolution. Our experiments show that even subtle ERM elevations can be sonified. Expert clinician feedback confirms the impact of our method and an initial user study with 15 participants demonstrates the potential to perceive the gap between the ERM and the retinal tissue exclusively through auditory cues.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2024/paper/3700_paper.pdf

SharedIt Link: pending

SpringerLink (DOI): pending

Supplementary Material: https://papers.miccai.org/miccai-2024/supp/3700_supp.zip

Link to the Code Repository

N/A

Link to the Dataset(s)

N/A

BibTex

@InProceedings{Mat_Ocular_MICCAI2024,
        author = { Matinfar, Sasan and Dehghani, Shervin and Sommersperger, Michael and Faridpooya, Koorosh and Fairhurst, Merle and Navab, Nassir},
        title = { { Ocular Stethoscope: Auditory Support for Retinal Membrane Peeling } },
        booktitle = {proceedings of Medical Image Computing and Computer Assisted Intervention -- MICCAI 2024},
        year = {2024},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 15006},
        month = {October},
        page = {pending}
}

Reviews

Review #1

  • Please describe the contribution of the paper

    The paper presents a novel method for discerning the disparity between the ERM and retinal tissue solely via auditory cues.

  • 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 novelty of employing auditory cues in this intricate surgical task is commendable.

  • 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 concept is intriguing, relying on feedback from just one expert fails to provide sufficient insight into the effectiveness and user-friendliness of the proposed approach. More comprehensive testing with a diverse pool of experts would better inform the readers.

  • Please rate the clarity and organization of this paper

    Very 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.

  • Do you have any additional comments regarding the paper’s reproducibility?

    NA

  • Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html

    NA

  • 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

    Weak Reject — could be rejected, dependent on rebuttal (3)

  • Please justify your recommendation. What were the major factors that led you to your overall score for this paper?

    During the rebuttal period, I have reservations about the authors’ ability to conduct a thorough study involving a larger number of expert users. Therefore, my recommendation is Weak Reject.

  • Reviewer confidence

    Confident but not absolutely certain (3)

  • [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed

    Reject — should be rejected, independent of rebuttal (2)

  • [Post rebuttal] Please justify your decision

    I am not convinced with the reason provided for having only one expert in the study.



Review #2

  • Please describe the contribution of the paper

    This paper proposes to use sonification to support retinal membrane peeling, a microsurgical procedure in the eye. Information from optical coherence tomography is used and translated into sound signals. The system is evaluated in a user study with 15 inexperienced users.

  • 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 straightforward to follow. It provides thorough explanations of the medical background and related work. • Sonification for surgical navigation is definitely an intriguing yet relatively under-studied topic.

  • 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.

    • Unclear motivation and implementation: The paper lacks validation in certain statements, particularly in the related work section, where the limitations of existing methods lack substantiation from literature or experimental results. Consequently, the motivation behind behind the overall design is not quite clear. A central aspect of the method, how visual signals are translated into audio signals, seems missing.

    • Weak evaluation: Although a user study with 15 users was conducted, its informative value is limited due to the lack of a baseline condition or comparison to other methods. There are no quantitative measures to demonstrate the capabilities of the system.

  • Please rate the clarity and organization of this paper

    Very 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 provide sufficient information for reproducibility.

  • Do you have any additional comments regarding the paper’s reproducibility?

    Information about the used transfer function to map imaging information to sound is missing.

  • Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html

    • In the final paragraph of the related work, several statements are made that are neither substantiated by literature nor supported by experimental results. For example, stating that sonification methods without a physical model are more difficult to understand, lead to an increased mental load and longer learning periods, needs validation through experiments or related work. • Similarly, statements regarding the limitations of existing methods, e.g., that they cannot communicate high-level structural information with the necessary precision, lack specific citations. It is unclear which methods exactly these paragraphs refer to, and how this limitation manifests, ideally in terms of tangible metrics or clinical standards. • Consequently, certain aspects of the method lack clear motivation. For instance, the choice of a physical model in section 3.3 appears somewhat arbitrary, and the superiority of such a complex model over alternative approaches like constant adaptation of sound frequency or pitch remains unclear. • There is a disjoint between sections 3.2 and 3.3, as the actual mapping between the A-scan signal and sound output seems to be omitted or is just referred to as a “transfer function”. It seems that this function is a crucial aspect of the method. • As a reader not familiar with OCT ERM imaging, it would have been helpful for the authors to highlight regions of interest, such as areas with elevated ERM or attached membrane, in the images. I got distracted by the large gap in the membrane, which is, I think, not what I should be looking at. • The interpretation of the results from the user study is challenging due to the absence of a baseline condition for comparison. Additionally, there is no comparison to other augmentation methods, either visual or auditory, making it difficult to assess the effectiveness of the proposed approach in context. While the numbers sound good, it uncertain how users would have performed under e.g. visual guidance or a simpler sonification method. • Perhaps additional quantification of the model, such as its ability to detect subtle changes in structural patterns and temporal precision, could compensate for the limited expressiveness of the user study. • It remains unclear how the method would perform when encountering non-conforming structures, such as anatomical/pathological variations or imaging artifacts, in the B-scan. Would the method be able to differentiate these from actual ERM elevations/gaps?

  • 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

    Weak Reject — could be rejected, dependent on rebuttal (3)

  • Please justify your recommendation. What were the major factors that led you to your overall score for this paper?

    Although I liked that the paper was well written and clearly presented, the two main weaknesses lead me to recommend weak rejection. Firstly, the specific methodology lacks clear motivation or support from existing literature. Secondly, the experimental validation is insufficient to substantiate these statements and design choices.

  • Reviewer confidence

    Confident but not absolutely certain (3)

  • [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed

    N/A

  • [Post rebuttal] Please justify your decision

    N/A



Review #3

  • Please describe the contribution of the paper

    The paper builds a acoustic model to conveys the physical characteristics of tissue structure in order to help the doctor with the ERM surgery.By conducting user study, the method proved to be effective to improve decision-making and surgery outcomes.

  • 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 papar describes the structure of the Epiretinal Membranes from a new perspective (by acoustic model), which can help doctors to make qualitative judgments. By listening the Supplementary, it is easy to distinguish the “special” sound which is the gap.

  • 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 acoustic model in the paper is not assicated with the detailed characteristics (like the depth of gap) in the Epiretinal Membranes,a quantification is required.

  • 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 provide sufficient information for reproducibility.

  • Do you have any additional comments regarding the paper’s reproducibility?

    N/A

  • Please provide detailed and constructive comments for the authors. Please also refer to our Reviewer’s guide on what makes a good review. Pay specific attention to the different assessment criteria for the different paper categories (MIC, CAI, Clinical Translation of Methodology, Health Equity): https://conferences.miccai.org/2024/en/REVIEWER-GUIDELINES.html

    A visualization of the acoustic model is recommended be done to intuitively show the tissue strucure in 2D or 3D level.

  • 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

    Weak Accept — could be accepted, dependent on rebuttal (4)

  • Please justify your recommendation. What were the major factors that led you to your overall score for this paper?

    The paper is well structured. However, the work is just a start, more work need to be done to improve the acoustic model and show the structure of the Epiretinal Membranes clearly and intuitively.

  • Reviewer confidence

    Confident but not absolutely certain (3)

  • [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed

    N/A

  • [Post rebuttal] Please justify your decision

    N/A



Author Feedback

We thank the reviewers for their insightful comments and constructive feedback. We are delighted they recognized the novelty of our work and its value, as sonification is an intriguing yet under-studied topic (R3). They commended our approach for providing the ERM structure from a new perspective through the acoustic model, aiding doctors in qualitative judgments and noted the sounds representing the gaps were easily discernible by listening to the supp. material. (R4). We are grateful for their suggestions and will next address their comments by outlining the changes we’ll make based on their feedback. /Regarding the comments on the extent of the user study, particularly with only one medical expert (R1, R3) and comparison with baseline conditions (R3): Sonification is an emerging modality with multifaceted nature to assess. We agree that having surgeon’s feedback is crucial to ensure the system’s relevancy. Our system has been developed in close collaboration with clinical experts. However, since most surgeons are not yet trained or familiar with concepts of sonification, validating the psychoacoustics and intuitiveness aspects of such methods on non-expert users, while keeping an expert in the study design loop, is extremely important for user-centered solutions. To verify the concept and its face validity for community exposure, in the absence of any prior work for comparison, the current studies were essential, before taking the valuable time of multiple clinical experts and planning extensive clinical experiments. We find it beneficial to share the results of our study with 15 participants, which confirmed primary factors with a high percentage (98.5%) after a brief training session. The expert’s confirmation of the task’s suitability further underscores the importance of publishing these findings. This would aid in conducting further experiments, aligning with the scope of a CAI-based conference paper introducing a Novel MIC approach to addressing an unmet CAI need. /Regarding the question on the reproducibility (R3, R4), we will include the following text at the end of Section 3.3 to further clarify the translation of visuals into audio: “The transfer functions fm, fc, and fk map image intensity (I) within the normalized range of [0.001, 1] to model param m (kg), c (N.s/m), and k (N/m). These values are empirically determined to optimize the trade-off between high sound contrast and model stability: fm(I -> [5, 10]), fc(I -> [0.001, 0.01]), and fk(I -> [1, 5]).” /We appreciate the question about the design motivation (R3) and will paraphrase the last paragraph of Related Work accordingly: “Research shows the human brain is optimized for processing behaviorally relevant natural sounds [1]. Accurate representation of objects’ acoustics requires considering both spectral (frequency mapping, distinct from pitch) and the temporal features. Further, as data dimensionality increases, the freq. mapping becomes more challenging, requiring case-specific design. Physical modeling efficiently handles freq. mapping, incorporating the spatial characteristics of the imaging data, and optimizes temporal mapping with minimal parameters: mass, stiffness, and damping. These parameters affect the model’s resonant properties, determining sound factors such as attack and decay time and sustain level across the freq. Spectrum.” /We agree with comments about highlighting the RoI (elevated ERM), differing ERM and other pathological/anatomical structures, explicitly referencing existing methods (R3), and enhancing the acoustic model visualization for better structure demonstration (R4). We will address these by minor updates of the figures and video. We appreciate the reviewers’ recognition of our approach’s novelty and the positive feedback on the paper’s clarity and structure. We hope that our rebuttal addresses their questions satisfactorily.

  1. F. Theunissen et al. Neural processing of natural sounds. Nature Reviews Neuroscience.



Meta-Review

Meta-review #1

  • 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 paper proposes a novel concept to sonify gaps between an elevated ERM and retinal structure. Spectrograms as well as the user study showed that relevant structural patterns could easily be differentiated through auditory feedback. Clinician feedback confirmed the impact of this method for retinal membrane surgery and the presence of distinctive sound textures to identify even slight elevations of the ERM. After careful consideration of the authors’ rebuttal, two of the three reviewers lean towards rejecting the paper citing user study improvements wrt sample size of participants or recruiting more than one expert clinician. There was reviewer consensus that the methodology of using acoustic/sonification is commendable. I agree that the authors have addressed the major concerns and questions raised by the reviewers, however having 15 participants and an expert in a user study involving methods that are technically robust should not be considered a limitation. This said, I lean towards a accepting the paper.

  • 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).

    The paper proposes a novel concept to sonify gaps between an elevated ERM and retinal structure. Spectrograms as well as the user study showed that relevant structural patterns could easily be differentiated through auditory feedback. Clinician feedback confirmed the impact of this method for retinal membrane surgery and the presence of distinctive sound textures to identify even slight elevations of the ERM. After careful consideration of the authors’ rebuttal, two of the three reviewers lean towards rejecting the paper citing user study improvements wrt sample size of participants or recruiting more than one expert clinician. There was reviewer consensus that the methodology of using acoustic/sonification is commendable. I agree that the authors have addressed the major concerns and questions raised by the reviewers, however having 15 participants and an expert in a user study involving methods that are technically robust should not be considered a limitation. This said, I lean towards a accepting the paper.



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 paper presents auditory feedback for retinal procedure guidance, a novel technique that is under-explored. The reviewers’ scores are divided, with two recommending rejection and one recommending acceptance. While I acknowledge that the work is in its early stages, I do not believe the lack of experts in the user study should be the primary reason for rejection. I believe this work is of interest to the MICCAI community and lean toward accepting it.

  • 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).

    The paper presents auditory feedback for retinal procedure guidance, a novel technique that is under-explored. The reviewers’ scores are divided, with two recommending rejection and one recommending acceptance. While I acknowledge that the work is in its early stages, I do not believe the lack of experts in the user study should be the primary reason for rejection. I believe this work is of interest to the MICCAI community and lean toward accepting it.



back to top