List of Papers By topics Author List
| Paper Info | Reviews | Meta-review | Author Feedback | Post-Rebuttal Meta-reviews |
Authors
Haibo Yang, Shengjie Zhang, Xiaoyang Han, Botao Zhao, Yan Ren, Yaru Sheng, Xiao-Yong Zhang
Abstract
Small lesions in magnetic resonance imaging (MRI) images are crucial for clinical diagnosis of many kinds of diseases. However, the MRI quality can be easily degraded by various noise, which can greatly affect the accuracy of diagnosis of small lesion. Although some methods for denoising MR images have been proposed, task-specific denoising methods for improving the diagnosis confidence of small lesions are lacking. In this work, we propose a voxel-wise hybrid residual MLP-CNN model to denoise three-dimensional (3D) MR images with small lesions. We combine basic deep learning architecture, MLP and CNN, to obtain an appropriate inherent bias for the image denoising and integrate each output layers in MLP and CNN by adding residual connections to leverage long-range information. We evaluate the proposed method on 720 T2-FLAIR brain images with small lesions at different noise levels. The results show the superiority of our method in both quantitative and visual evaluations on testing dataset compared to state-of-the-art methods. Moreover, two experienced radiologists agreed that at moderate and high noise levels, our method outperforms other methods in terms of recovery of small lesions and overall image denoising quality. The implementation of our method is available at https://github.com/laowangbobo/Residual_MLP_CNN_Mixer.
Link to paper
DOI: https://link.springer.com/chapter/10.1007/978-3-031-16437-8_28
SharedIt: https://rdcu.be/cVRte
Link to the code repository
https://github.com/laowangbobo/Residual_MLP_CNN_Mixer
Link to the dataset(s)
Reviews
Review #1
- Please describe the contribution of the paper
- Authors propose a voxel-wise hybrid residual MLP-CNN model to denoise 3D MR images with small lesions.
- The proposed method shows a good performance (SSIM and PSNR) compared to some state-of-the-art methods. -The diagnosis confidence for small lesions is confirmed by radiologists.
- 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.
-
Novelty: Authors present a deep learning method based on a voxel-wise hybrid residual MLP-CNN model
-
Radiologist confirmation: two experienced radiologists confirmed that the proposed method outperforms other methods in terms of recovery of small lesions and overall image denoising quality.
-
- 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.
- Information about computational time is not presented.
- Limited comparison to state-of-the-art: Only PSNR and SSIM mesures are studied. But, it’s enough for a conference paper. A computational time comparison is missing.
- 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 code is not available at the moment, but enough details to reproduce it are given. The dataset used by authors is publicly available. The used dataset is described and cited.
- 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
Some suggestions:
- It would be interesting that authors gave some information about the computational time required by the proposed method.
- Summarize the research limitations and future research directions.
- Extend the Conclusion with details on the method performance when compared to other tested techniques (in terms of PSNR improvement and SSIM).
- 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 topic of the paper is relevant and interesting to the MICCAI community.
- It presents an innovative idea to noise reduction of 3D MR images: deep learning technique based on a voxel-wise hybrid residual MLP-CNN model
- The results are clearly presented and the conclusions are supported by the results.
- Radiologist confirmation: the diagnosis confidence of small lesion is confirmed by radiologists.
- Number of papers in your stack
6
- What is the ranking of this paper in your review stack?
3
- 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
N/A
- [Post rebuttal] Please justify your decision
N/A
Review #2
- Please describe the contribution of the paper
The authors proposed a method for image denoising of 3D (multislice) MRI using an MLP-CNN architecture. The method processed each image patch with MLPs followed by an encoder-decoder CNN with residual connections. Results show that the proposed method outperforms other methods in terms of recovery of small lesions.
- 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 has a very good structure.
- The method improve image quality while restoring the details of small lesions in the image.
- 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 authoer only evaluate the proposed method on data with simulated noise.
- Albation study of the proposed MLP-CNN architecture is missing.
- Relatively week baseline. Recent proposed Transformer-based denoisers are not compared, e.g., SwinIR.
- 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
The authors used a public dataset with URL provided.
- 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
- Ablation studies need to be performed to evaluate the contributions from the MLP and the CNN.
- How does the patch size affect the performance of the model?
- Why only the central slice and its neighboring slices were extracted? Do the authors also did this for the test data?
- 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
5
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
The proposed method outperforms baseline in terms of both quantitative metrics and recovery of small lesions. However, the authors only experiment with simulated noise and ablation studies are lacking.
- Number of papers in your stack
5
- 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
N/A
- [Post rebuttal] Please justify your decision
N/A
Review #3
- Please describe the contribution of the paper
Authors propose a MLP-CNN structure for 3D MRI denoising. Compared with some other denoising methods, the proposed method shows superior results by presenting clear small lesions.
- 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 organized and has presented enough figures and tables to support authors’ ideas.
A reader study is conducted to evaluate super-resolution results and show the possibility of the using the proposed method in real-world applications.
- 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 only suggestion is to consider conducting ablation study to show the contribution of residual MLPs and residual convolutional subnetwork parts for final denoising results.
- 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
It is hard to be reproduced as there is no detailed description about the proposed method and no code publicly available.
- 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
This paper proposes a neural network with multiple residual MLPs and a residual convolutional subnetwork for MRI denoising. Compared with some other denoising methods, the proposed method shows superior results by presenting clear small lesions. The paper is well organized and has presented enough figures and tables to support authors’ ideas.
The only suggestion is to consider conducting ablation study to show the contribution of residual MLPs and residual convolutional subnetwork parts for final denoising 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
6
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
Novelty, experimental design, result presentation.
- Number of papers in your stack
5
- What is the ranking of this paper in your review stack?
1
- Reviewer confidence
Very confident
- [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
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 work is interesting and the novelty of the work is sufficient. Radiologists’ scoring introduced is appreciated and the results are promising. Ablation study can be introduced to better demonstrate the effectiveness of the proposed method.
- 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).
3
Author Feedback
We thank the reviewers very much for their positive recommendation. The following is our response to the remaining comments.
- The work is interesting and the novelty of the work is sufficient. Radiologists’ scoring introduced is appreciated and the results are promising. Ablation study can be introduced to better demonstrate the effectiveness of the proposed method (reviewer 2) RE: We will include the ablation study when the paper is formally submitted.
- Why only the central slice and its neighboring slices were extracted? Do the authors also did this for the test data? (reviewer 2) RE: The reason is that small lesions are easily found in the central slice and its neighboring slices of the brain. So, we extracted these slices and performed the same operation for the test data.
- Recent proposed Transformer-based denoisers are not compared, e.g., SwinIR. (reviewer 2) RE: This work focuses on comparisons with state-of-the-art MRI denoising methods, including RED-WGAN. Although SwinIR performs well in restoring natural images, its performance in MRI image denoising is still unknown. We will discuss this method in the final submitted version.
- It is hard to be reproduced as there is no detailed description about the proposed method. (reviewer 3) RE: We have provided sufficient details in the section of ‘Methods’.