List of Papers Browse by Subject Areas Author List
Abstract
White matter hyperintensities (WMH) on T2-weighted MRI
are associated with cerebrovascular pathology and neurodegeneration, including myelin loss. Luxol Fast Blue (LFB) histopathology visualizes myelin integrity, and optical density (OD) serves as a quantitative proxy for myelin concentration. However, variability in staining and tissue processing introduces batch effects that confound OD comparison across histology runs. To address this, we present an automated pipeline that identifies reference (non-pathologic) white-matter regions directly from whole-slide images, without manual annotation, and uses them to normalize OD heatmaps. Against expert ratings of myelin-loss severity, automatically selected references matched expert placement (ρ = 0.88), and batch-calibrated OD best separated intact tissue from regions of severe myelin loss (AUC = 0.90). In voxel-wise comparison with co-registered 7T ex vivo MRI across 151 sections from 31 hemispheres, batch-calibrated OD correlated best with MRI signal (ρ=−0.379), and the correlation persisted within WMH (ρ=−0.336), indicating capture of graded myelin loss. By mitigating staining artifacts, this pipeline provides a validated framework for quantitative cross-modal comparison and a foundation for translating histology-derived myelin maps to in vivo imaging biomarkers. Code is available at https://github.com/
zkhodakarami/myelin_quantification.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026/paper/6153_paper.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to the Code Repository
https://github.com/zkhodakarami/myelin_quantification
Link to the Dataset(s)
N/A
BibTex
@InProceedings{KhoZah_Automated_MICCAI2026,
author = { Khodakarami, Zahra AND Emrani, Sheina AND Khandelwal, Pulkit AND Athalye, Chinmayee AND Denning, Amanda AND Trotman, Winifred AND Levorse, Lisa M. AND Teunissen-Bermeo, Eric AND Radhakrishnan, Hamsanandini AND Ohm, Daniel T. AND Olm, Christopher AND Capp, Noah AND Ittyerah, Ranjit AND Prabhakaran, Karthik AND Detre, John A. AND Das, Sandhitsu R. AND Wolk, David A. AND McMillan, Corey T. AND Tisdall, M. Dylan AND Irwin, David J. AND Robinson, John L. AND Lee, Edward B. AND Yushkevich, Paul A.},
title = { { Automated Optical Density Normalization for Myelin Quantification: Cross-Modal Validation with 7T Ex Vivo MRI } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 16896},
month = {September},
page = {pending}
}
Reviews
Review #1
- Please describe the contribution of the paper
This paper introduces a new method for normalizing optical density values obtained through whole slide imaging of Luxol Blue staining. The authors utilized an inverted attention mechanism in a pretrained network that detects white matter pallor to circumscribe stable regions for reference by using the areas of lowest attention.
- 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.
The major strength of this paper is using a highly innovative and creative approach to solve an outstanding problem, especially for microscopy-MRI fusion. The study uses a proper held-out test set with good performance. Results and interpretation are honestly reported and adequately tempered given the limitations. But the results are still very impressive.
- 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.
I cannot find any major weaknesses in this work.
- 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 has provided an anonymized link to the source code, dataset, or any other dependencies.
- 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.
(6) Strong Accept — must be accepted due to excellence
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
This project fulfills a crucial gap in bridging micro and macro neuroscience, greatly improving the comparison between histochemical staining and MRI. The methods are sound and robust. Limitations are honest and interpretations are not overstated.
- Reviewer confidence
Very confident (4)
- [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.
N/A
- [Post rebuttal] Please justify your final decision from above.
N/A
Review #2
- Please describe the contribution of the paper
This paper proposes an automated pipeline to normalize OD measurements in LFB histopathology, addressing staining variability that obscures myelin quantification in white matter hyperintensity analysis. The pipeline identifies reference non-pathologic regions in whole-slide images to compute normalized Optical Density heatmaps, mitigating staining artifacts for consistent cross-modal comparison. Validation against expert ratings and co-registered 7T ex vivo MRI demonstrates strong concordance and improved correlation with MRI signal intensity
- 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.
1 The paper presents a solid validation scheme that combines expert comparison through statistical testing with cross-modal correlation measurements, providing a thorough and complementary assessment of the proposed pipeline’s reliability.
2 The research statement is valid and well-motivated, and this work offers a meaningful proof of concept for the use of automated histopathology normalization as a basis for grading myelin loss in white matter regions.
3 The proposed pipeline provides a time-saving procedure for neuropathologists by automating the discovery of reference regions, reducing the manual burden associated with assessing demyelination in whole-slide images.
4 The figures and tables are of high quality and well suited to the evaluation presented, clearly supporting the reported findings and aiding the reader in interpreting the results.
- 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.
1 The paper does not provide sufficient justification for the choice of the segmentation network used for white matter segmentation. The selection criteria and any comparison with alternative segmentation approaches are absent, making it difficult to assess whether this choice is optimal for the task.
2 The core model, namely CLAM, which is used for identifying the optical density reference regions for the normalization strategy, is not adequately explained in the methodology section. Given that this is a central component of the pipeline, a more thorough technical description of how CLAM operates within this context is necessary for reproducibility and clarity.
3 The paper would greatly benefit from a diagram illustrating the full pipeline, accompanied by visual examples of the resulting images at each phase of the process. Such a figure would considerably improve the reader’s understanding of how the different components interact and how the normalization progressively affects the output at each stage
- 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 has provided an anonymized link to the source code, dataset, or any other dependencies.
- 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.
(5) Accept — should be accepted, independent of rebuttal
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
1 The paper addresses a valid and clinically important research question.
2 The proposed pipeline represents a novel contribution to quantitative histopathology analysis for myelin loss assessment.
3 The results are of high quality and supported by a strong validation strategy combining expert assessment and cross-modal MRI correlation.
- Reviewer confidence
Very confident (4)
- [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.
N/A
- [Post rebuttal] Please justify your final decision from above.
N/A
Review #3
- Please describe the contribution of the paper
The main contribution of this paper is the development of a framework that maps histopathological stain-absorption measurements to imaging on a per- tissue basis (with a specific interest in white matter).
- 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 has several strengths which are as follows. 1.The first strength is the framework around mapping histology to imaging. This is a hard problem to solve and has been one of the major frameworks several fields that utilize imaging for studying pathology has been searching for. So, this is extremely important research. 2.The authors are using an information- rich dataset. 3.The methods used are methodologically sound, including the model parameters used in the CLAM model and the normalization strategy. 4.The statistical methods used are sound. The inclusion of both Spearman and Pearson is correct (even though, just Spearman would have been enough). However, including the Pearson results is also sound.
- 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.
There were no major weaknesses, this was such a well thought out experiment and design. The only thing I would suggest is to increase the font in figure 2.It’s hard to read the metrics in the statistics box. Use figure 3 as a reference.
- 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 mention open access to source code or data, but provides a clear and detailed description of the algorithm to ensure reproducibility.
- 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.
(6) Strong Accept — must be accepted due to excellence
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
The paper as a whole. I believe this is such a great application of deep learning for clinical imaging, which, I know, can have a direct effect on the ability of researchers to better understand biology from imaging. The paper is well organized and very detailed, which is necessary for reproducibility.
- Reviewer confidence
Very confident (4)
- [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.
N/A
- [Post rebuttal] Please justify your final decision from above.
N/A
Author Feedback
We sincerely appreciate the reviewers’ enthusiastic and positive evaluations of our work. We are especially grateful that the reviewers recognized the novelty and methodological strength of our contribution, highlighting the “highly innovative and creative approach”, the “novel contribution to quantitative histopathology analysis”, and the importance of “mapping histology to imaging” as a long-standing challenge in the field. Below, we provide further explanations and outline the updates we will make to the camera-ready paper to ensure more clarity.
(R2-W1) Segmentation network. nnU-Net v2, trained on 256 WSIs (Dice 0.896), with annotations excluding cut edges and periventricular tracts that would distort myelin OD. nnU-Net is the established default for biomedical segmentation, outperforming specialized solutions across 23 international challenges [10] and has been specifically validated for WSI tissue segmentation [Helgesen et al., COMPAYL 2025]. The inverted-attention framework itself is agnostic to the segmentation backbone. We will add this justification to the camera-ready.
(R2-W2,3) (R4) CLAM, inverted-attention mechanism, and figure enhancement. We will expand Section 2.2 in the camera-ready to provide a more thorough description of CLAM and the inverted-attention mechanism. Our submission already includes a pipeline figure (Fig. 1), which we will refine in the camera-ready to further improve readability (R2). We will also enhance the readability of Figure 2 (R4).
Meta-Review
Meta-review #1
- Your recommendation
Provisional Accept
- 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.
All reviewers unanimously endorse this paper for its novelty, rigor, and clinical impact. It provides an elegant solution to a long-standing problem in multimodal microscopy-MRI fusion. I recommend Provisional Accept primarily due to its technical novelty and clinical impact.