pith. machine review for the scientific record. sign in

arxiv: 2605.08711 · v1 · submitted 2026-05-09 · ⚛️ physics.med-ph · q-bio.NC· q-bio.QM

Recognition: 2 theorem links

· Lean Theorem

Automated Optical Density Normalization for Myelin Quantification: Cross-Modal Validation with 7T Ex Vivo MRI

Zahra Khodakarami , Sheina Emrani , Pulkit Khandelwal , Chinmayee Athalye , Amanda Denning , Winifred Trotman , Lisa M Levorse , Eric Teunissen-Bermeo
show 16 more authors
Hamsanandini Radhakrishnan Daniel Ohm Christophe Olm Noah Capp Ranjit Ittyerah Karthik Prabhakaran John A. Detre Sandhitsu R. Das David A. Wolk Corey T McMillan Gabor Mizsei M. Dylan Tisdall David J Irwin John L. Robinson Edward B Lee Paul A. Yushkevich
Authors on Pith no claims yet

Pith reviewed 2026-05-12 00:51 UTC · model grok-4.3

classification ⚛️ physics.med-ph q-bio.NCq-bio.QM
keywords optical density normalizationmyelin quantificationwhite matter hyperintensitiesLuxol Fast Blue7T ex vivo MRIhistopathologycross-modal validation
0
0 comments X

The pith

An automated pipeline normalizes optical density from myelin-stained slides to match 7T MRI signals more closely.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents an automated method that picks reference regions in whole-slide images to normalize optical density values from Luxol Fast Blue staining. Raw measurements vary too much from lab protocols and processing steps, which hides real differences in myelin. After normalization, the values align better with MRI intensity measurements across white matter areas, including inside hyperintense zones where myelin loss is gradual. This turns variable histology into a more usable quantitative signal for comparing myelin health.

Core claim

The authors demonstrate that automatically selected non-pathologic reference regions produce normalized optical density heatmaps with substantially stronger voxel-wise correlation to co-registered 7T ex vivo MRI signal intensity than raw optical density values, and that this stronger correlation remains inside white matter hyperintensities.

What carries the argument

Automated identification of reference regions in whole-slide images to compute normalized optical density as a myelin proxy.

If this is right

  • Enables direct numerical comparison of myelin content across different histology batches.
  • Supports voxel-level cross-modal checks between stained tissue and MRI.
  • Tracks continuous degrees of myelin loss rather than simple presence or absence.
  • Reduces staining-related noise that otherwise blocks consistent pathology studies.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same reference-selection step could reduce batch effects in other myelin or tissue stains.
  • Applying the pipeline to larger cohorts might improve statistical power in studies linking myelin loss to clinical outcomes.
  • If the reference selection generalizes, it could support myelin quantification in routine diagnostic slides without extra manual steps.

Load-bearing premise

The automatically chosen reference regions represent tissue without myelin loss and give an unbiased baseline that holds across staining protocols and disease states.

What would settle it

If normalized optical density shows no stronger correlation with MRI intensity than raw values, or if the correlation disappears inside white matter hyperintensities, on a new independent dataset.

Figures

Figures reproduced from arXiv: 2605.08711 by Amanda Denning, Chinmayee Athalye, Christophe Olm, Corey T McMillan, Daniel Ohm, David A. Wolk, David J Irwin, Edward B Lee, Eric Teunissen-Bermeo, Gabor Mizsei, Hamsanandini Radhakrishnan, John A. Detre, John L. Robinson, Karthik Prabhakaran, Lisa M Levorse, M. Dylan Tisdall, Noah Capp, Paul A. Yushkevich, Pulkit Khandelwal, Ranjit Ittyerah, Sandhitsu R. Das, Sheina Emrani, Winifred Trotman, Zahra Khodakarami.

Figure 1
Figure 1. Figure 1: Overview of the proposed pipeline. Details in section 2. The top two rows belong to the same batch, whereas the bottom row is from another hemisphere pro￾cessed under a different setting. WSI: Whole-Slide Images, LFB-OD: Luxol Fast Blue Optical Density, Norm.: Normalized. (I) white matter regions are not masked red, and delineated areas with lines are white matter hyperintensities; used for cross-modal eva… view at source ↗
Figure 2
Figure 2. Figure 2: Validation of CLAM-based reference region selection against manual expert placement across 84 slides. Left: raw Luxol Fast Blue Optical Density(LFB-OD) values (Pearson r = 0.91). Right: percentile-transformed values (Pearson r = 0.85). trained raters independently placed reference regions on each slide; discrepant placements were reviewed jointly and resolved by consensus. The consensus ref￾erence OD showe… view at source ↗
Figure 3
Figure 3. Figure 3: Distribution of Luxol Fast Blue Optical Density by expert pathology grade (0 = intact, 3 = severe myelin loss; n = 97 ROIs) across four quantification methods. More details in section 3.2. Statistical comparisons: ∗ p < 0.05, ∗ ∗ p < 0.01, ∗ ∗ ∗ p < 0.001, ∗ ∗ ∗ ∗ p < 0.0001. WMH masks. Given the CC z-score normalization, the expected relationship is negative: high OD reflects intact myelin (near-zero z-sc… view at source ↗
read the original abstract

White matter hyperintensities (WMH) are bright regions on T2-weighted magnetic resonance imaging (MRI) scans and are associated with cerebrovascular pathology and neurodegeneration, including myelin loss. While Luxol Fast Blue histopathology provides visualization of myelin integrity, quantitative analysis requires measuring Optical Density as a proxy for myelin concentration. However, differences in laboratory protocols and tissue processing introduce staining variability that acts as systematic noise, obscuring the biological signal and preventing consistent comparison across histology runs. To address this, we developed an automated pipeline that identifies reference (non-pathologic) regions in whole-slide images to compute normalized Optical Density heatmaps. We validated this approach through two complementary evaluations: (1) comparison against expert ratings of myelin loss severity, and (2) cross-modal spatial comparison with co-registered 7T ex vivo MRI for voxel-wise evaluation within white matter regions. The pipeline's reference selection showed strong concordance with expert-identified reference regions, and normalized Optical Density demonstrated a substantially stronger correlation with MRI signal intensity than raw measurements. This correlation persisted within WMH, confirming that the pipeline captures continuous myelin pathology rather than merely the presence or absence of myelin loss contrast. By mitigating staining artifacts, this pipeline provides a robust, validated framework for quantitative cross-modal comparison, establishing a critical methodological foundation for future translation to in vivo myelin mapping and biomarker discovery.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript presents an automated pipeline for normalizing optical density (OD) in Luxol Fast Blue-stained whole-slide histology images to quantify myelin integrity within white matter hyperintensities (WMH). The method automatically identifies reference (non-pathologic) regions to generate normalized OD heatmaps that mitigate staining variability across protocols. Validation consists of (1) concordance between the automated reference selection and expert ratings of myelin loss severity and (2) voxel-wise spatial correlation of normalized OD with co-registered 7T ex vivo MRI signal intensity, including within WMH, to demonstrate capture of graded myelin pathology rather than binary contrast.

Significance. If the claims hold without circularity in reference selection, the work supplies a practical, automated framework for quantitative myelin assessment from routine histology. This would strengthen cross-modal validation against MRI and support biomarker studies in cerebrovascular and neurodegenerative disease. The use of high-resolution 7T ex vivo MRI for spatial ground-truth comparison is a clear methodological strength, as is the focus on continuous pathology inside WMH. The absence of numerical effect sizes in the abstract, however, limits immediate assessment of practical utility.

major comments (3)
  1. [Abstract] Abstract: The central claim that 'normalized Optical Density demonstrated a substantially stronger correlation with MRI signal intensity than raw measurements' is presented without any quantitative metrics (Pearson r, R², sample size, confidence intervals, or p-values). This omission prevents evaluation of the magnitude and reliability of the reported improvement, which is load-bearing for the validation narrative.
  2. [Methods] Methods (reference region selection): The pipeline's automatic identification of reference regions is reported to show 'strong concordance with expert-identified reference regions,' yet no details are given on the algorithm's feature set or explicit tests confirming independence from subtle myelin loss or co-varying staining artifacts. If selection criteria partially track myelin content or MRI-correlated artifacts, the observed correlation gain could arise from circularity rather than unbiased normalization, undermining the 'continuous pathology' claim inside WMH.
  3. [Results] Results (WMH-specific correlation): The persistence of improved MRI correlation within WMH is used to argue that the pipeline captures graded myelin pathology. However, the manuscript provides no description of WMH segmentation criteria on histology, no quantitative correlation values stratified by WMH vs normal-appearing white matter, and no evaluation across different staining batches or disease states, leaving generalizability untested.
minor comments (2)
  1. [Abstract] Abstract: The phrase 'expert concordance' should be accompanied by a specific statistic (e.g., Cohen's kappa or percentage agreement) to allow readers to judge the strength of agreement.
  2. [Discussion] The manuscript would benefit from a dedicated limitations paragraph addressing potential dependence of reference selection on tissue processing variations.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their insightful and constructive comments, which have helped clarify several aspects of our work. We have revised the manuscript to include quantitative metrics in the abstract, expanded methodological details on reference region selection, and added stratified analyses and limitations regarding generalizability. Our point-by-point responses follow.

read point-by-point responses

  1. Referee: [Abstract] The central claim that 'normalized Optical Density demonstrated a substantially stronger correlation with MRI signal intensity than raw measurements' is presented without any quantitative metrics (Pearson r, R², sample size, confidence intervals, or p-values). This omission prevents evaluation of the magnitude and reliability of the reported improvement, which is load-bearing for the validation narrative.

    Authors: We agree that the abstract should report quantitative metrics to allow immediate assessment of effect size and reliability. We have revised the abstract to incorporate the Pearson r values, sample sizes, confidence intervals, and p-values from the voxel-wise analyses presented in the Results section. revision: yes

  2. Referee: [Methods] The pipeline's automatic identification of reference regions is reported to show 'strong concordance with expert-identified reference regions,' yet no details are given on the algorithm's feature set or explicit tests confirming independence from subtle myelin loss or co-varying staining artifacts. If selection criteria partially track myelin content or MRI-correlated artifacts, the observed correlation gain could arise from circularity rather than unbiased normalization, undermining the 'continuous pathology' claim inside WMH.

    Authors: We have expanded the Methods section to fully describe the algorithm's feature set, which is based on anatomical location priors for reference white matter and global intensity histogram statistics, without using local myelin density or any MRI information. Reference selection is performed exclusively on histology images, eliminating direct circularity with the MRI-based validation. The reported strong concordance with expert ratings (performed independently of MRI) provides supporting evidence that selection does not track subtle myelin loss in a manner that would artifactually inflate the cross-modal correlation. revision: yes

  3. Referee: [Results] The persistence of improved MRI correlation within WMH is used to argue that the pipeline captures graded myelin pathology. However, the manuscript provides no description of WMH segmentation criteria on histology, no quantitative correlation values stratified by WMH vs normal-appearing white matter, and no evaluation across different staining batches or disease states, leaving generalizability untested.

    Authors: We have added a description of the WMH segmentation criteria on histology (neuropathologist-guided manual delineation based on LFB staining intensity, texture, and morphology) to the Methods section. We have also included quantitative stratified correlation results (normalized vs. raw OD within WMH and within normal-appearing white matter) in the Results to demonstrate the improvement in each compartment. Our dataset uses a single staining protocol and cohort; we have added an explicit limitations statement in the Discussion acknowledging this and outlining the need for future multi-batch and multi-disease validation. revision: partial

Circularity Check

0 steps flagged

No circularity: empirical validation against independent expert ratings and co-registered MRI

full rationale

The paper describes an automated reference-region pipeline for OD normalization whose output is validated by two external benchmarks: concordance with expert myelin-loss ratings and voxel-wise correlation against co-registered 7T MRI. No equations, fitted parameters, or self-citations are shown that would make the reported correlation gain equivalent to the normalization step by construction. The correlation result is presented as an outcome of the pipeline, not as an input used to tune it. The derivation chain is therefore self-contained against the stated external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The work rests on standard domain assumptions in histopathology and image analysis without introducing new physical entities or many explicit free parameters.

axioms (2)
  • domain assumption Optical density in Luxol Fast Blue staining serves as a monotonic proxy for myelin concentration.
    Invoked in the abstract as the basis for quantitative myelin measurement.
  • domain assumption Non-pathologic reference regions can be reliably identified and used to correct for global staining intensity differences.
    Central premise of the normalization pipeline stated in the abstract.

pith-pipeline@v0.9.0 · 5681 in / 1380 out tokens · 60742 ms · 2026-05-12T00:51:40.647353+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

26 extracted references · 26 canonical work pages

  1. [1]

    Slic superpixels compared to state-of-the-art superpixel methods

    Achanta, R., Shaji, A., Smith, K., Lucchi, A., Fua, P., Süsstrunk, S.: SLIC Superpixels Compared to State-of-the-Art Superpixel Methods34(11), 2274–2282. https://doi.org/10.1109/TPAMI.2012.120, http://ieeexplore.ieee.org/ document/6205760/

    work page doi:10.1109/tpami.2012.120 2012

  2. [2]

    https://doi.org/10

    Athalye, C., Bahena, A., Khandelwal, Pulkit, e.a.: Operationalizing postmortem pathology-MRI association studies in Alzheimer’s disease and related disor- ders with MRI-guided histology sampling13(1), 120. https://doi.org/10. 1186/s40478-025-02030-y, https://actaneurocomms.biomedcentral.com/articles/ 10.1186/s40478-025-02030-y

    work page doi:10.1186/s40478-025-02030-y

  3. [3]

    https://doi.org/10.1038/s41598-017-17204-5, https://www.nature.com/articles/s41598-017-17204-5

    Bankhead, P., Loughrey, M.B., Fernández, J.A., Dombrowski, Y., McArt, D.G., Dunne, P.D., McQuaid, S., Gray, R.T., Murray, L.J., Coleman, H.G., James, J.A., Salto-Tellez, M., Hamilton, P.W.: QuPath: Open source software for digital pathol- ogy image analysis7(1), 16878. https://doi.org/10.1038/s41598-017-17204-5, https://www.nature.com/articles/s41598-017-17204-5

    work page doi:10.1038/s41598-017-17204-5

  4. [4]

    https://doi.org/10

    Bartzokis, G.: Age-related myelin breakdown: A developmental model of cognitive decline and Alzheimer’s disease25(1), 5–18. https://doi.org/10. 1016/j.neurobiolaging.2003.03.001, https://linkinghub.elsevier.com/retrieve/pii/ S0197458003001647

    work page 2003

  5. [5]

    https://doi.org/10.1016/j.jalz.2007

    Bartzokis, G., Lu, P.H., Mintz, J.: Human brain myelination and amyloid beta de- position in Alzheimer’s disease3(2), 122–125. https://doi.org/10.1016/j.jalz.2007. 01.019, https://alz-journals.onlinelibrary.wiley.com/doi/10.1016/j.jalz.2007.01.019

    work page doi:10.1016/j.jalz.2007 2007

  6. [6]

    https://doi.org/10.1212/WNL.0b013e31824e8e7f, https:// www.neurology.org/doi/10.1212/WNL.0b013e31824e8e7f

    Deramecourt, V., Slade, J., Oakley, A., Perry, R., Ince, P., Maurage, C.A., Kalaria, R.: Staging and natural history of cerebrovascular pathology in demen- tia78(14), 1043–1050. https://doi.org/10.1212/WNL.0b013e31824e8e7f, https:// www.neurology.org/doi/10.1212/WNL.0b013e31824e8e7f

    work page doi:10.1212/wnl.0b013e31824e8e7f

  7. [7]

    https://doi.org/10.1016/j.jpi.2025.100423, https://linkinghub

    Dunn, C., Brettle, D., Hodgson, C., Hughes, R., Treanor, D.: An international study of stain variability in histopathology using qualitative and quantitative anal- ysis17, 100423. https://doi.org/10.1016/j.jpi.2025.100423, https://linkinghub. elsevier.com/retrieve/pii/S2153353925000057

    work page doi:10.1016/j.jpi.2025.100423 2025

  8. [8]

    https://doi.org/ 10.1136/jnnp.2009.204685, https://jnnp.bmj.com/lookup/doi/10.1136/jnnp.2009

    Gouw, A.A., Seewann, A., Van Der Flier, W.M., Barkhof, F., Rozemuller, A.M., Scheltens, P., Geurts, J.J.G.: Heterogeneity of small vessel disease: A systematic review of MRI and histopathology correlations82(2), 126–135. https://doi.org/ 10.1136/jnnp.2009.204685, https://jnnp.bmj.com/lookup/doi/10.1136/jnnp.2009. 204685

    work page doi:10.1136/jnnp.2009.204685 2009

  9. [9]

    https://doi.org/10.1007/s00401-009-0635-8, http://link.springer.com/10.1007/s00401-009-0635-8 10 Khodakarami et al

    Ihara, M., Polvikoski, T.M., Hall, R., Slade, J.Y., Perry, R.H., Oakley, A.E., En- glund, E., O’Brien, J.T., Ince, P.G., Kalaria, R.N.: Quantification of myelin loss in frontal lobe white matter in vascular dementia, Alzheimer’s disease, and dementia with Lewy bodies119(5), 579–589. https://doi.org/10.1007/s00401-009-0635-8, http://link.springer.com/10.10...

    work page doi:10.1007/s00401-009-0635-8

  10. [10]

    Nature Methods 18(2), 203–211 (2021)

    Isensee, F., Jaeger, P.F., Kohl, S.A.A., Petersen, J., Maier-Hein, K.H.: nnU-Net: A self-configuring method for deep learning-based biomedical image segmenta- tion18(2), 203–211. https://doi.org/10.1038/s41592-020-01008-z, https://www. nature.com/articles/s41592-020-01008-z

    work page doi:10.1038/s41592-020-01008-z

  11. [11]

    https://doi.org/10.3389/fmed.2021.746307, https: //www.frontiersin.org/articles/10.3389/fmed.2021.746307/full

    Kang, H., Luo, D., Feng, W., et al.: StainNet: A Fast and Robust Stain Nor- malization Network8, 746307. https://doi.org/10.3389/fmed.2021.746307, https: //www.frontiersin.org/articles/10.3389/fmed.2021.746307/full

    work page doi:10.3389/fmed.2021.746307 2021

  12. [12]

    https://doi.org/10.48550/ ARXIV.2506.19106, https://arxiv.org/abs/2506.19106

    Khan, U., Härkönen, J., Friman, M., et al.: Staining normalization in histopathol- ogy: Method benchmarking using multicenter dataset. https://doi.org/10.48550/ ARXIV.2506.19106, https://arxiv.org/abs/2506.19106

    work page arXiv

  13. [13]

    https://doi.org/10.1162/imag_a_00171, https://direct.mit.edu/imag/article/doi/10.1162/imag_a_00171/120741/ Automated-deep-learning-segmentation-of-high

    Khandelwal, P., Duong, M.T., Sadaghiani, S., et al.: Automated deep learning segmentation of high-resolution 7 Tesla postmortem MRI for quantitative analysis of structure-pathology correlations in neurodegener- ative diseases2, imag–2–00171. https://doi.org/10.1162/imag_a_00171, https://direct.mit.edu/imag/article/doi/10.1162/imag_a_00171/120741/ Automate...

    work page doi:10.1162/imag_a_00171

  14. [14]

    https://doi

    Khodanovich, M.Y., Sorokina, I.V., Glazacheva, Valentina Yu, e.a.: Histological validation of fast macromolecular proton fraction mapping as a quantitative myelin imaging method in the cuprizone demyelination model7(1), 46686. https://doi. org/10.1038/srep46686, https://www.nature.com/articles/srep46686

    work page doi:10.1038/srep46686

  15. [15]

    In: Hod- der Arnold Publications, vol

    Kiernan, J.: Histological and histochemical methods: Theory and practice. In: Hod- der Arnold Publications, vol. 6th

    work page

  16. [16]

    https://doi.org/10.1097/ 00005072-195312040-00008, https://academic.oup.com/jnen/article-lookup/doi/ 10.1097/00005072-195312040-00008

    Klüver, H., Barrera, E.: A Method for the Combined Staining of Cells and Fibers in the Nervous System12(4), 400–403. https://doi.org/10.1097/ 00005072-195312040-00008, https://academic.oup.com/jnen/article-lookup/doi/ 10.1097/00005072-195312040-00008

    work page doi:10.1097/00005072-195312040-00008

  17. [17]

    https://doi.org/10

    Lin, S., Zhou, H., Watson, M., et al.: Impact of stain variation and color normal- ization for prognostic predictions in pathology15(1), 2369. https://doi.org/10. 1038/s41598-024-83267-w, https://www.nature.com/articles/s41598-024-83267-w

    work page

  18. [18]

    Macenko, M., Niethammer, M., Marron, J. S. and, e.a.: A method for normalizing histology slides for quantitative analysis. In: 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro. pp. 1107–1110. IEEE. https://doi. org/10.1109/ISBI.2009.5193250, http://ieeexplore.ieee.org/document/5193250/

    work page doi:10.1109/isbi.2009.5193250 2009

  19. [19]

    https://doi.org/10.1186/s40478-022-01425-5, https: //actaneurocomms.biomedcentral.com/articles/10.1186/s40478-022-01425-5

    McKenzie, A.T., Marx, G.A., Koenigsberg, Daniel, e.a.: Interpretable deep learning of myelin histopathology in age-related cognitive impair- ment10(1), 131. https://doi.org/10.1186/s40478-022-01425-5, https: //actaneurocomms.biomedcentral.com/articles/10.1186/s40478-022-01425-5

    work page doi:10.1186/s40478-022-01425-5

  20. [20]

    https://doi.org/10.1109/38.946629, http: //ieeexplore.ieee.org/document/946629/

    Reinhard, E., Adhikhmin, M., Gooch, B., Shirley, P.: Color transfer between images21(4), 34–41 (July-Aug/2001). https://doi.org/10.1109/38.946629, http: //ieeexplore.ieee.org/document/946629/

    work page doi:10.1109/38.946629 2001

  21. [21]

    Ruifrok, A., Johnston, D.: Quantification of histochemical staining by color decon- volution23

    work page

  22. [22]

    N4ITK: Improved N3 Bias Correction,

    Tustison, N.J., Avants, B.B., Cook, Philip A, e.a.: N4ITK: Improved N3 Bias Correction29(6), 1310–1320. https://doi.org/10.1109/TMI.2010.2046908, http: //ieeexplore.ieee.org/document/5445030/

    work page doi:10.1109/tmi.2010.2046908 2010

  23. [23]

    https://doi.org/10.1109/TMI.2016.2529665, http://ieeexplore.ieee.org/document/ 7460968/ Optical Density Normalization for Myelin 11

    Vahadane, A., Peng, T., Sethi, Amit, e.a.: Structure-Preserving Color Normal- ization and Sparse Stain Separation for Histological Images35(8), 1962–1971. https://doi.org/10.1109/TMI.2016.2529665, http://ieeexplore.ieee.org/document/ 7460968/ Optical Density Normalization for Myelin 11

    work page doi:10.1109/tmi.2016.2529665 1962

  24. [24]

    https://doi.org/10.1161/JAHA.114.001140, https://www

    Wardlaw, J.M., Valdés Hernández, M.C., Muñoz-Maniega, S.: What are White Matter Hyperintensities Made of?: Relevance to Vascular Cognitive Impair- ment4(6), e001140. https://doi.org/10.1161/JAHA.114.001140, https://www. ahajournals.org/doi/10.1161/JAHA.114.001140

    work page doi:10.1161/jaha.114.001140

  25. [25]

    https://doi.org/10.3174/ajnr

    Warntjes, J., Persson, A., Berge, J., Zech, W.: Myelin Detection Using Rapid Quantitative MR Imaging Correlated to Macroscopically Registered Luxol Fast Blue–Stained Brain Specimens38(6), 1096–1102. https://doi.org/10.3174/ajnr. A5168, http://www.ajnr.org/lookup/doi/10.3174/ajnr.A5168

    work page doi:10.3174/ajnr

  26. [26]

    https://doi.org/10.1016/j.jalz.2016.06

    Yushkevich, P.A., Pluta, J., Wang, Hongzhi, e.a.: IC-P-174: Fast Automatic Seg- mentation of Hippocampal Subfields and Medial Temporal Lobe Subregions In 3 Tesla and 7 Tesla T2-Weighted MRI12. https://doi.org/10.1016/j.jalz.2016.06. 205, https://alz-journals.onlinelibrary.wiley.com/doi/10.1016/j.jalz.2016.06.205

    work page doi:10.1016/j.jalz.2016.06 2016