Abstract

Electrosurgical coagulation is widely used in laparoscopic surgery; however, intraoperative assessment of tissue response remains indirect and largely reliant on visual cues, increasing the risk of insufficient coagulation or excessive thermal damage. In this work, we investigate vibroacoustic (VA) sensing as a non-invasive modality for monitoring electrosurgical coagulation. VA sensing captures structure-borne mechanical vibrations generated at the tool-tissue interaction point without requiring modifications to standard surgical instruments or workflows. Using a bipolar electrosurgical setup, VA signals were acquired under two complementary scenarios: prolonged coagulation of ex vivo porcine liver tissue to analyze temporal signal evolution, and coagulation of ex vivo porcine liver, muscle, and fat to investigate tissue-specific responses. The signals were analyzed using time- and time-frequency-domain signal processing methods. Results show that VA signals evolve systematically throughout coagulation, revealing distinct phases up to carbonization. In addition, tissue-specific analyses demonstrate reproducible differences in temporal dynamics and spectral content, enabling clear separation of tissue types in a non-linear embedding space. These findings indicate that VA sensing captures reproducible state-dependent and tissue-specific characteristics during electrosurgical energy delivery, highlighting its potential to address the clinical challenge of perceiving and controlling tissue response during electrosurgery.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026/paper/2429_paper.pdf

SharedIt Link: Not yet available

SpringerLink (DOI): Not yet available

Supplementary Material: https://papers.miccai.org/miccai-2026/supp/2429_supp.zip

Link to the Code Repository

https://gitlab.gwdg.de/cds/va-electrosurgery

Link to the Dataset(s)

https://gitlab.gwdg.de/cds/va-electrosurgery

BibTex

@InProceedings{EsmNaz_When_MICCAI2026,
        author = { Esmaeili, Nazila AND Urrutia, Robin AND Espejo, Diego AND Boese, Axel AND Horvath, Philipp AND Hansen, Christian AND Hagenah, Jannis AND Illanes, Alfredo},
        title = { { When Tissue Speaks: Vibroacoustic Monitoring of Tissue Response During Electrosurgery } },
        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 presents the first use of vibroacoustic (VA) sensing for non-invasively monitoring electrosurgical coagulation of tissue. To accomplish this, the authors propose the combination of standard surgical instruments with a small VA sensing unit using existing standardized connectors and thus not negatively affecting the surgical workflow. In an ex vivo setting, prolonged coagulation as well as coagulation of different tissue types showed distinct coagulation phases along the coagulation procedure and differences in temporal dynamics and spectral content promising to allow separation of different tissue types.

  • 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.

    Clinical relevance: Proposing and analyzing a method for tissue response monitoring, the paper addresses a real and important limitation of conventional electrosurgery which is especially relevant in laparoscopic and robotic settings where perception is limited. This is a clinically well motivated sensing problem that can matter in CAI.

    Hardware setup and workflow: The hardware setup and sensing workflow provides an elegant solution how to integrate tissue response measurements into electrosurgical applications without disturbing the conventional surgical workflow. Mounting the vibroacoustic sensor proximally on standard instruments allows direct tissue response sensing without requiring custom tool tips or increasing the invasiveness of the intervention.

  • 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.

    Phase transition claim: From the experiments performed for scenario I, the claim that the observed phase-dependent changes appear consistent with thermally induced tissue transformations is hard to verify from the performed experiments as no real indicators (e. g. tissue temperature) of the transition from one tissue transformation phase into the other is provided. This leaves this part of the paper in the speculative, a more thorough investigation including real-world reference measures would be required.

    Tissue differentiation claim: While the performed analysis provide some indication that tissue differentiation can be possible with this method, a proper analysis should include a hold out test set to establish differentiation performance

    Methodological reproducibility: The methodological description lacks detail to allow for straightforward reproduction of the method. In particular, the description of the implementation and parametrization of the individual processing steps remains rather vague.

    Statistical rigor: The authors claim statistical significance for their results shown in Fig. 3e but do not provide statistical tests, confidence intervals or similar to support this claim.

  • 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 authors claimed to release the source code and/or dataset upon acceptance of the submission.

  • 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

    Detailed comments on the paper can be found in the following: 1.It remains unclear how the frequency-dependent noise threshold was determined. 2.The visual inspection “to identify frequency bands showing consistent dynamical changes” is currently not described in sufficient detail for reproduction. 3.How were the homomorphic envelope, the low-pass filter and the moving average filter parameterized to determine the indicators? 4.How were the peaks in the signal detected? Which method has been used for this? 5.Why are there gaps between the pseudo-frequency regions? 200-250 Hz between LF and MF? 3000-5000 Hz between MF und HF? 6.For the sentence “Fig. 3(b–d) show the instantaneous band-energy contribution indicator (%) of the identified LF, MF, and HF bands. “, “LF, MF, and HF” should be reordered such that it reflects the same order as Fig. 3(b-d). 7.How were the phase transition points determined exactly? 8.While the dominant MF activity is visible in the example given in Figure 3a-d, the aggregated statistical result do not seem to show such a clear peaking in Ph2.

  • 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.

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

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

    The paper addresses a clinically relevant problem and proposes an elegant, surgical workflow-compatible sensing approach with promising early results shown on ex-vivo tissue. However, the current evidence remains preliminary. The main claims on tissue-state phases and tissue differentiation are not sufficiently supported by objective reference measurements, held-out validation, or rigorous statistics. Together with missing implementation details, this limits confidence in the robustness and reproducibility of the findings.

  • Reviewer confidence

    Confident but not absolutely certain (3)

  • [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.

    Reject

  • [Post rebuttal] Please justify your final decision from above.

    I appreciate the authors’ clarification with respect to the transition point determination, the UMAP/kNN analysis and the exploratory character of the performed study. I still believe that this is a very interesting and clinically valuable research direction. However, the rebuttal does not fully address my main concerns about the current level of validation. In particular, I would still recommend extending the analysis to include objective reference measurements of tissue state such as temperature or other thermal/tissue-damage indicators, a more comprehensive validation of tissue differentiation beyond exploratory visualization, and a complete description of the methodological details. In its current form, I therefore maintain my original assessment, but believe that a revised and resubmitted version could become a valuable contribution to the field.



Review #2

  • Please describe the contribution of the paper

    This paper introduces vibroacoustic (VA) sensing as a way to monitor electrosurgical bipolar coagulation. The tissue heating during the procedure generates vibrations throught the tool that get picked up by a sensor that is clipped on, allowing the monitoring of tissue responses and upgrading already existing tools in a cost effective way.

  • 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.

    S1 : Electrosurgical tissue state monitoring is an open clinical problem. The VA sensing solution is not invasive, tool agnostic and addresses a gap not covered by existing systems (impedance or acoustic based methods)

    S2 : Scenario 1 (temporal evolution) and Scenario 2 (tissue type differentiation) address two clinically relevant and separate questions, providing a more complete picture.

    S3 : Signal analyisis choice: four indicators (homomorphic envelope, peak density, stationary CWT, normalized stationary CWT) capture complementary elements of the signal.

    S4 : No instrument modification is a strong practical contribution.

    S5 : Data and code will be released on acceptance, bolstering reproducibility.

    S6 : Experiment performed by an experienced surgeon with a standardized protocol.

  • 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.

    W1 : Boxplots for phase wise band energy contributions (Fig. 3e, 3f) reported but have no p-values, confidence intervals, nor statistical tests. Paired Wilcoxon tests or similar would be expected with N=15 in Scenario 1.The lack of quantitative significance data is a big methodological gap.

    W2 : Phase transitions (~2.3 s, ~7.5 s) are identified from a representative single trial, then used to segment all 15 trials, then boxplots are computed. The method does not verify if transitions occur at consistent times across trials or if data driven phase detection phase detection would yield the same structure.

    W3 : All tissue is ex vivo porcine. In vivo involves perfusion, blood oozing, varying tissue hydration etc… Extension to in vivo scenarios is entirely unexplored.

    W4 : N=15 for S1 and N=10 per tissue for S2 : with small sample sizes, overfitting becomes an issue when using UMAP (https://arxiv.org/abs/1802.03426, Section 6)

  • 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 authors claimed to release the source code and/or dataset upon acceptance of the submission.

  • 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

    Q1 : The phase transitions (~2.3 s, ~7.5 s) were identified visually from a single representative trial before being applied to segment all 15 trials for quantitative analysis. What is the inter trial variability of these transition times? A formal change point detection analysis would bolster reproducibility claims.

    Q2 : The UMAP embedding in Fig. 4e uses n=3, seems far below typical dataset sizes. Could you report a leave on out cross validation or similar on the four indicator vectors?

  • 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 is a solid proof of concept demonstrating that vibroacoustic sensing can capture tissue-state information during electrosurgical coagulation. The application is novel and sound yet some methodological concerns are significant warranting a revision.

  • 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.

    Clarified two main concerns that I had.

    UMAP/kNN confirmed as descriptive visualization and not a predictive classifier.

    Phase transitions clarified as per-trial manual annotation and not transferred from one representative trial, needs to be stated explicitely in the final paper.

    Statistical rigor and R1 reproducibility concerns remain unsatisfied.

    Sensing concept is genuinely novel. Clip-on + tool-agnostic hardware is an elegant and cost-effective contribution.



Review #3

  • Please describe the contribution of the paper

    The main contribution of this paper is showing that vibrations made when using electrosurgery can give us useful information about the tissue. The researchers found that these vibrations change in predictable ways as the tissue is heated and even show differences between different types of tissue (like liver, muscle, and fat). This could help surgeons better control the process and avoid damaging tissue.

  • 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.
    • This paper is the first to use vibrations (called “vibroacoustics”) to learn more about tissue during electrosurgery. It’s a new way to get information, going beyond just looking and measuring electricity.
    • The sensor is designed to easily attach to existing surgical tools without needing any changes to how surgeons work or putting anything inside the patient’s body
    • The experiments were designed very well to test different aspects of the technique (how vibrations change over time and how they differ between tissues). They followed a clear plan and were done consistently.
  • 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 should clarify whether the UMAP + kNN analysis is intended as descriptive visualization or as a predictive classifier. If the former, the classifier framing and claims of ‘complete separability’ and ‘robust differentiation’ should be softened. If the latter, a proper cross-validation (e.g., leave-one-out given n=30) should be reported with accuracy metrics.”

    • Another potential limitation is the extremely small sample size (only 30 total samples across all tissue types). With so few data points, it’s difficult to draw firm conclusions about the technique’s reliability and generalizability. A larger sample size would provide more confidence in the 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 authors claimed to release the source code and/or dataset upon acceptance of the submission.

  • 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.

    (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?

    The paper introduces vibroacoustic sensing as a genuinely novel monitoring modality for electrosurgery, an application not previously explored. The proposed clip-on sensor is tool-agnostic, non-invasive, and translationally attractive, and the two-scenario experimental design is clean and well-structured. The results are visually compelling. However, enthusiasm is tempered by several concerns. The authors should clarify whether the UMAP + kNN analysis is intended as descriptive visualization or as a predictive classifier: if the former, the classifier framing and claims of “complete separability” and “robust differentiation” should be softened; if the latter, a proper cross-validation (e.g., leave-one-out given n=30) should be reported with accuracy metrics. Another limitation is the extremely small sample size (only 30 samples across all tissue types), which makes it difficult to draw firm conclusions about reliability and generalizability. Overall, the paper introduces a promising new sensing direction and provides preliminary yet suggestive evidence, which I believe will benefit the community, provided that the authors address the concerns mentioned above in the revision.

  • Reviewer confidence

    Not confident (1)

  • [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.

    The rebuttal adequately addresses my main concerns on the UMAP framing and sample size. I maintain my Weak Accept.



Author Feedback

We thank the reviewers and meta-reviewer for the constructive feedback. All reviewers recognized the clinical relevance, novelty, and translational potential of the proposed vibroacoustic (VA) sensing approach for electrosurgery. Regarding the concern about statistical support for the reported phase differences, Scenario I aims to show that VA signals evolve systematically during electrosurgical coagulation, revealing reproducible signal changes up to tissue carbonization. Fig. 3(a–d) illustrates the temporal evolution of the VA signal, while Fig. 3(e,f) was intended to generalize these observations across the 15 independently analyzed VA data from a signal perspective, rather than to provide statistical validation of tissue-state transitions. Across VA signals, consistent redistribution of energy between LF/MF/HF bands and reduced total spectral energy in the late phase were observed. We apologize if some wording suggested stronger statistical claims. The interpretation of these behaviors as thermally induced tissue transformations was intended as hypothesis-driven, consistent with established thermal tissue literature and the manuscript conclusion that the biological interpretation is based on indirect evidence and requires validation. Regarding the clarification on how phase transitions were determined and their consistency across trials, we apologize if the methodological description was not sufficiently clear. In the manuscript, signals were described as segmented according to transition points identified during the qualitative analysis. However, the manual annotation procedure was not described explicitly enough. The transition points shown in the representative example were not transferred to all signals. Instead, phase transitions were independently identified for each VA signal through manual annotation based on changes observed in the CWT spectrum and band-energy indicators. The goal was to characterize systematic changes in VA signal behavior during coagulation. The temporal duration of the phases naturally varied across signals. The reproducible aspect was therefore not fixed transition timing, but the consistent evolution of VA signal behavior during coagulation. Regarding the descriptive or predictive role of the UMAP/kNN analysis, Scenario II was intended as a descriptive exploratory analysis rather than a predictive classification framework. UMAP was used as an unsupervised visualization of the extracted VA indicators without tissue labels, while the kNN boundaries were included only to illustrate local label consistency. Therefore, the observed grouping reflects structure inherently present in the extracted signal features rather than supervised optimization toward tissue classes. Regarding tempering claims about tissue differentiation given the small dataset, we agree that the current ex vivo dataset is limited and does not support strong claims regarding generalizability. Scenario II should therefore be interpreted as exploratory evidence that different tissues exhibit distinguishable VA behaviors in this controlled setting. Regarding reproducibility, the methodology combines standard signal-processing methods with the proposed VA sensing framework. Therefore, the manuscript focused primarily on the sensing concept and signal behaviors rather than low-level implementation details. For clarification, the pseudo-frequency bands were identified from recurrent spectral activity patterns consistently observed across signals, and all processing steps were applied using fixed parameters. The planned release of code and data will allow full reproducibility and inspection of the processing pipeline. Overall, we believe the work fits well within the MICCAI scope and evaluation criteria. The paper addresses a clinically relevant CAI problem, introduces a novel sensing concept with translational potential, and presents systematic exploratory experiments supporting the feasibility of the proposed approach.




Meta-Review

Meta-review #1

  • Your recommendation

    Invite for Rebuttal

  • 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.

    Given the mixed but generally borderline-positive reviews, I recommend inviting a rebuttal. Several key concerns: (1) provide stronger statistical support for the reported phase differences; (2) clarify how phase transitions were determined and whether they are consistent across trials; (3) clarify whether UMAP/kNN is descriptive or predictive, and provide proper cross-validation if predictive claims are made; (4) temper claims about tissue differentiation given the small ex vivo sample size; and (5) improve methodological detail for reproducibility.

  • 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.

    Reviewers recognized the novelty, clinical relevance, and translational appeal of the proposed sensing concept. Following the rebuttal, two reviewers were satisfied with the clarification that the UMAP/kNN analysis is descriptive rather than predictive, and that phase transitions were annotated independently per trial rather than transferred from a single representative example. However, one reviewer still has concerns regarding the validation and lack of objective reference measurements of tissue state, such as temperature or tissue-damage indicators. Overall, the rebuttal appropriately tempered some claims and clarified the exploratory nature of the work, but concerns remain regarding statistical rigor and biological validation. Given the novelty and potential impact of the sensing approach, I recommend acceptance.



Meta-review #2

  • 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.

    The rebuttal has not addressed the validation questions raised by the first reviewer, which are important in understanding the value of the proposed work.



Meta-review #3

  • 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.

    The reviewers’ concerns have been successfully addressed in the rebuttal, making the work a solid contribution especially for what concerns clinical applications in CAI.



back to top