List of Papers By topics Author List
| Paper Info | Reviews | Meta-review | Author Feedback | Post-Rebuttal Meta-reviews |
Authors
Tiancheng Li, Yang Song, Peter Walker, Kai Pan, Victor A. van de Graaf, Liang Zhao, Shoudong Huang
Abstract
Total hip arthroplasty (THA) is an orthopaedic surgery to replace the diseased ball and socket of the hip joint with artificial implants. Achieving appropriate leg length and offset in THA is critical to avoid instability, leg length discrepancies, persistent pain, or early implant failure. This paper provides the first electromagnetic (EM) sensor based approach for accurately measuring the change in leg length and offset intraoperatively. The proposed approach does not require the direct line-of-sight, avoids the need for accurately returning the leg back to the neutral reference position, and has an efficient closed-form solution from least squares optimisation. Validations using simulations, phantom experiments, and cadaver tests demonstrate that the proposed method can provide more accurate results than the conventional mechanical method by manual gauge, the optical tracking based approach, and the direct use of one EM reading, thus showing significant potential clinical value. mechanical method by manual gauge, the optical tracking based approach, and the direct use of one EM reading, thus showing significant potential clinical value.
Link to paper
DOI: https://doi.org/10.1007/978-3-031-43996-4_35
SharedIt: https://rdcu.be/dnwPe
Link to the code repository
N/A
Link to the dataset(s)
N/A
Reviews
Review #2
- Please describe the contribution of the paper
The paper describes a novel way to estimate changes in leg length and femoral offset during total hip arthroscopy. Instead of using a single measurement the paper uses the fact that motion is constrained to a rotation around the femoral head to estimate the leg length and femoral offset from multiple dynamic measurements and the surgeon rotates the femur around the new implant.
- 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.
Utilising knowledge about the motion constraints to allow the use of a larger sample of data is a nice idea and from my understanding is novel. The authors demonstrate through their phantom and cadaver experiment that the proposed method outperforms existing methods. The method is well described and the clinical application is common, so the method has excellent potential to contribute to better patient outcomes.
- 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 need for the closed form solution isn’t well backed up by the results. It is certainly quicker but from figure 4 the least squares solution at 100 samples takes less than a second, so there’s not really going to be any clinical impact from using the closed form solution. Even at 1000 samples it’s only around 20 seconds to find the least squares solution, which is very short in a typical clinical workflow.
- The comparison with other methods (Table 2) is not clear. How the other methods are performed is not explained in detail nor adequately referenced in the methods section, giving me very little confidence that it is a fair comparison.
- The paper claims that the EM tracked system outperforms an optical tracked system, however this isn’t supported by the results. As stated above I am unclear on exactly how you performed the optical tracking method, but it does not appear to be by using the optical tracker as a straight swap with the EM system and then using the same rotational sampling approach. It would be of more interest if you used the optical tracker with the same method at the EM tracker. As it is it is perhaps more relevant to compare the Optical tracking with the one EM reading, in which case optical tracking outperforms EM.
- 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
I don’t think the results could be reproduced easily. The supplementary video makes things clearer but from the paper there is insufficient information on the data collection to be confident that an attempt to repeat the experiment would yield similar results. The key point is on page 6, para 2 “the surgeon slightly rotated the femur”. This is an insufficient description and could not be replicated. The authors haven’t included any data or software implementation.
- 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/2023/en/REVIEWER-GUIDELINES.html
In general the paper presents a nice idea to solve a clinical problem. It is well written and demonstrates that the method is better than the state of the art. However there are some significant weakness as described in section 6. Below are some other minor comments.
- It’s very hard to asses the clinical relelevance of some of the errors. For example, from page 2 “ … can result in additional measurement errors since only 4° of abduction/adduction could cause 5–7 mm error … “. I don’t know if 5 - 7 mm is good or bad. Some discussion of what surgeons currently aim for would be good.
- Section 3.1 simulation: This section doesn’t really contribute anything. EM and optical tracking systems introduce system specific errors that may not be well modeled by zero mean Gaussian errors. Yet your least squares and closed form solutions are built on the assumption that errors are zero mean Gaussian. So the simulation doesn’t show anything of interest.
- From the text it’s not clear how the surgeon finds and records the neutral reference position. More information would be helpful.
- Reference 19 is used to support a claim on clinical outcome and hip durability, but appears to be about radiography of a specific cohort. I don’t think it has anything to say about longer term outcomes.
- 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?
The paper doesn’t demonstrate that the closed form solution has any benefit over an iterative approach. The comparison with the claimed state of the art is weak.
- 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
5
- [Post rebuttal] Please justify your decision
I’ve revised my ranking slightly, on the basis that the authors include some statistical tests (from the rebuttal) and address some of the other comments.
Review #3
- Please describe the contribution of the paper
The authors present a method for the measurement of limb length in THA using electromagnetic tracking using a closed-form solution.
- 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 approach is simple and has potential to be translated into clinical practice. Furthermore, THA is a commonly performed surgery, therefore, the potential clinical impact is high. The method also eliminates the need for femoral repositioning which can reduce soft tissue stress in THA. The presented results look promising.
- 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 technical implementation is simple and the problem is formulated as a variant of a Pivot calibration problem which has been extensively used in prior work. The authors claim that the novelty of their work is the usage of EM tracking in THA which has been used in previous work.
Therefore, I consider the novelty of the work as moderate. - 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 description of the method is enough to reproduce the results.
- 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/2023/en/REVIEWER-GUIDELINES.html
The approach that the authors describe in the paper is one possible approach to measure the limb length in THA, another approach which is used clinically is to measure the distance from the minor trochanter to the pin of the femoral stem implant which is defined based on the preoperative plan. Therefore, the authors should elaborate on this in the paper.
“The optical tracking system is often used in THA for measurement as it has shown higher accuracy and reliability during interventions involving dynamic motion” –The authors should clarify if they are talking about research projects as cited in the remainder of the paragraph or clinical use. Therefore, this sentence is confusing.
“Our work is the first to use EM sensors as the primary measurement tools in THA surgery” – EM tracking has been used in previous work for computer-assisted cup positioning in THA [1].
“The component alignment requires the surgeon to return the femur to the neutral reference position and measure the change in leg length and offset.” – The authors should better explain the neutral reference position and how it is reached in clinical practice.
The authors should specify what a “slight movement” means and how they define it.
What is the accuracy of the optical and EM tracking systems used within the presented study?
[1] Jolles, Brigitte M MD*; Genoud, Patrick PhD†; Hoffmeyer, Pierre MD†. Clinical Orthopaedics and Related Research (1976-2007): September 2004 - Volume 426 - Issue - pp 174-179 doi: 10.1097/01.blo.0000141903.08075.83
- 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 clinical relevance is given, the novelty of the work is moderate.
- Reviewer confidence
Very confident
- [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed
5
- [Post rebuttal] Please justify your decision
The authors have addressed the comments raised in my previous review sufficiently especially by explaining the difference to vanilla pivot calibration. For the camera ready version (if accepted) I recommend to include a statement regarding this point. Furthermore, they should reformulate the claim about the usage of EM tracking according to their explanations in the rebuttal.
Review #4
- Please describe the contribution of the paper
This paper presents an approach to measure the length offset and leg length during Total Hip Arthroplasty (THA) using an electromagnetic sensor (EM) without requiring the need to return to neutral leg position or to have a direct line-of-sight. Only small movements of the femur are necessary to collect sufficient data to derive a set of equations that can be linearized and solved efficiently and quickly with a closed-form solution without any significant loss of accuracy.
- 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 presents a novel effective way to measure intraoperatively the length discrepancies occurring after the replacement of the femoral head, that does not disrupt the surgery and introduces minimal interventions (slight movements). The theoretical framework is sound and the paper is very well written. Experiments are also convincing including simulations, phantoms and cadavers.
- 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.
I don’t see any major weakness in the paper worth mentioning, although results could benefit from a more rigorous statistical approach (see comments).
- 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
Despite code or data will not be shared, authors did report sufficient information to support the reproducibility of the paper.
- 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/2023/en/REVIEWER-GUIDELINES.html
The only criticism I may have for the experiments, is that statistical significance is not established when comparing results (although the differences often look quite large) so this weakens the results.
- 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
7
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
The paper is clear, well written and brings novelty in length discrepancy measurement in THA with a convincing mathematical formulation and clinical experiments.
- Reviewer confidence
Very confident
- [Post rebuttal] After reading the author’s rebuttal, state your overall opinion of the paper if it has been changed
6
- [Post rebuttal] Please justify your decision
I agree with the metareviewer that imaging or image analysis is not present which may decrease its appeal to the CAI community, so I slightly revised my opinion. However, from a surgical assistance viewpoint I still think the paper has a strong impact and I found the rebuttal sufficiently convincing. From my experience, any reduction of effort for surgeons (do not impose line of sight, small movement to perform, etc.) is positive. I also found that close-form are always welcomed and reinforce the mathematical soundness and can always be used as good initialization for non-linear optimization usually better suited for noisy data. Statistical significance (my major concern) has been commented by authors who provided p-values. This should be added in a revised manuscript along with the name of chosen statistical test (whose relative conditions should be checked with care).
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.
The manuscript received mixed reviews with one reviewer recommending acceptance and two others recommending rejection.
The strengths of the paper include cadaver and phantom evaluations, a strong clinical motivation, and an interesting approach of using prior knowledge to constrain the problem setting.
However, the reviewers have also identified considerable shortcomings that I share. The chief concerns are as follows:
- The overall innovation is unclear; pivot calibration is a routine tool and EM has been used before in THA. A clear description of the innovative aspects is necessary.
- The potential benefits of the approach are unclear. While somewhat faster than prior methods, these techniques already run in sub-second time, so what is the possible benefit?
- The above concern should be contextualized with other concerns pertaining to the validity of the experimental design (see R2), which is further backed by R3 who is concerned about lack of significance analysis.
Overall, compelling answers to these concerns are required. In addition, I would mention that this work, while CAI, does not consider imaging or image analysis, which tends to be central to MICCAI papers.
Author Feedback
We thank Reviewer #2 (R2), R3, R4, and Meta-Reviewer (M). They appreciate that our work has “cadaver and phantom evaluations, a strong clinical motivation” (M), “the use of a larger sample of data is a nice idea” (R2), “theoretical framework is sound” (R4), “potential clinical impact is high” (R3).
Novelty of our work w.r.t. pivot calibration (M, R3): (i) Pivot calibration is commonly used for estimating tip location of a pointer tool or finding a rotation centre. It requires a relatively large rotational motion to accurately estimate the centre. (ii) Our problem focuses on estimating limb change after femoral head replacement rather than estimating rotation centre (Problem Statement, Page 4). We agree that part of our formulation is similar to pivot calibration (thanks R3 and M), but one key novelty of our method is, ACCURATE ROTATION CENTRE ESTIMATE IS UNNECESSARY FOR OBTAINING ACCURATE LIMB CHANGE ESTIMATES. As shown in framework (Fig. 1) we only require “slight movement” of leg for sampling. In this case, rotation centre estimate is inaccurate but our approach can still get accurate limb length estimates (Tab. 2 and Tab. 3). The reason is, with “slight movement”, although the uncertainty of [b, c] in Eq. (4) is large due to the near singularity of information matrix, uncertainty of solution in Eq. (5) is still small. Thus our problem is different from pivot calibration. Take an extreme case for illustration, if data are all sampled at the exact neutral position (without any leg movement) and assume zero sensor noise, then the pivot calibration problem is not solvable because of singularity, but our problem still has optimal solution which is the ground truth (limb change is 0). (iii) The “slight movement” (R2, R3) is an operational recommendation to surgeons – rotating around the neutral position within a few degrees.
Novelty of using EM (M, R3): (i) As stated in our abstract and introduction, we provide the first EM sensor based closed-form approach to estimate limb length change intraoperatively using a set of sampled poses from EM readings, which can greatly improve accuracy than using one reading. (ii) EM has been used in THA before, but EM was usually used as an auxiliary tracking tool in conjunction with other medical equipment. [1] mentioned by R3 used only one EM reading to position cup orientation and must be coordinated with CT.
Benefit of our closed-form solution (M, R2): Besides higher efficiency (R2), as compared with iterative nonlinear least squares solution, our closed-form solution (i) can obtain the optimal solution in one step for any input, avoiding potential local minima, no need of accurate initial value, (ii) is much easier to implement, (iii) is able to support EM system with higher acquisition frequency.
Validity of experimental design (M, R2, R4): (i) In Tab. 2, we are not claiming “EM tracked system outperforms an optical tracked system” (R2). Instead, we want to show our EM based closed-form solution outperforms existing methods including existing optical tracking based approach [13, 16] using only one optical reading, and manual gauge [1, 3, 11, 17]. To further demonstrate the benefit of our closed-form solution, we also compare with a straightforward idea of using EM sensor (one EM reading only). (ii) For statistical significance (M, R4), compared with our method, the P values of optical tracking, one EM reading, and manual gauge in Tab.2 are 3.35e-4, 1.27e-4, and 2.38e-5, respectively.
Contribution of simulation (R2). We consider the system-specific errors by adding zero-mean Gaussian noises to the sampling poses from EM READINGS (first sentence in Sec. 3.1). Our simulation aims to show the robustness of our approach to additionally added sensor noises (Tab. 1).
Neutral reference position (R2, R3): In standard THA, surgeons find the neutral position by experience. Our approach does not require experience-based repositioning to the neutral, thus improving accuracy
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.
After rebuttal, all reviewers are in favor of acceptance. However, all reviewers also emphasize the need for clarifications and answers provided in the rebuttal to be adequately reflected in the final version, and I emphatically second this request.
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.
The paper presents a novel method to estimate changes in limb length and femoral offset during total hip arthroplasty using an electromagnetic sensor. The methodology, demonstrated through phantom and cadaver experiments, has potential for clinical applicability due to its non-intrusive nature and efficiency. Certain aspects of the paper were critiqued, such as the lack of a clear rationale for the closed-form solution, the relative novelty of the work, the performance comparison with other methods, and the necessity for a more robust statistical analysis. Reviewer #1’s concern about the comparison with other methods was not completely addressed. Further, the authors’ claim that the closed-form solution avoids potential local minima needs to be empirically substantiated.
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.
Based on the rebuttal all three reviewers are in favour of accepting the paper (2 weak, 1 accept). I believe that the authors have done a good job addressing the initial concerns and despite no imaging in this study, there is a strong clinical motivation and testing on phantom and cadaver that would make the paper interesting to the MICCAI community.