Can MRI measure myelin? Systematic review, qualitative assessment, and meta-analysis of studies validating microstructural imaging with myelin histology

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

A systematic review was conducted using PRISMA guidelines and incorporating best practices from the AMSTAR 2 checklist.

“A systematic review was conducted using PRISMA guidelines (Shamseer et al., 2015) and incorporating best practices from the AMSTAR 2 checklist for clinical meta-analyses where applicable (e.g. searching across multiple datasets, performing study selection in duplicate) (Shea et al., 2017).”

• cites Shamseer et al., 2015 sourcePartly supported

  The paper describes PRISMA-P as a guideline for documenting systematic review and meta-analysis protocols, building on the PRISMA guidelines (Abstract; no page available). It therefore supports a connection to PRISMA, but it does not establish that the citing authors conducted their completed review using PRISMA guidelines. The AMSTAR 2 portion is attributed to Shea et al., not to this source. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites Shea et al., 2017 sourceSupported

May combine two claims.

DWI metrics are known to be influenced by both myelin and non-myelin factors, but are often used as microstructural markers.

“these metrics are known to be influenced by both myelin and non-myelin factors, but are often used as microstructural markers (Beaulieu, 2014, De Santis, Drakesmith, Bells, Assaf, Jones, 2014, Pierpaoli, Chiro, Basser, Trace, 1996)”

• cites Beaulieu, 2014 sourceNot checked — no full text of the source
• cites De Santis, Drakesmith, Bells, Assaf, Jones, 2014 sourceSupported
• cites Pierpaoli, Chiro, Basser, Trace, 1996 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable).

Validation is a necessary step towards making MRI-based methods biologically interpretable, and applicable to clinical and basic research.

“In light of these limitations, validation is a necessary step towards making MRI-based methods biologically interpretable, and applicable to clinical and basic research (Barros, Arribarat, Combis, Chaynes, 2019, Cohen-Adad, 2018).”

• cites Barros, Arribarat, Combis, Chaynes, 2019 sourcePartly supported

  The Discussion (Qualitative Histological Studies and MRI; page unavailable) says that correlating MRI with iron histology helps establish what an in vivo MRI signal represents. The Conclusion (page unavailable) says such correlations support reliable brain-iron estimation, while cautioning that methodological issues must be addressed before extrapolating to in vivo MRI. This supports validation for iron-sensitive brain MRI, but not the claim's broader assertion about MRI-based methods generally. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites Cohen-Adad, 2018 sourceNot checked — no full text of the source

• AssumptionRests on an assumption: It is currently unclear which MR marker is the most biologically accurate non-invasive measure for myelin, and how measures differ in sensitivity to specific aspects of myelination.

May combine two claims.

Different meta-analyses need to be run for papers reporting Spearman and Pearson correlation coefficients, which can differ in value even when applied to the same data, since Spearman correlation coefficients measure monotonicity and Pearson correlations measure linearity.

“For example, different meta-analyses need to be run for papers reporting Spearman and Pearson correlation coefficients, which can differ in value even when applied to the same data, since Spearman correlation coefficients measure monotonicity and Pearson correlations measure linearity (de Winter et al., 2016).”

• cites de Winter et al., 2016 sourcePartly supported

  Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The median sample size of validation studies is comparable to the median of the most cited fMRI studies during the period of publication included in the meta-analysis (12), but below the current median sample size (20).

“comparable to the median of the most cited fMRI studies during the period of publication included in our meta-analysis (12), but below current median sample size (20) (Szucs and Ioannidis, 2020)”

• cites Szucs and Ioannidis, 2020 sourcePartly supported

  The Abstract reports a median of 12 for highly cited experimental fMRI studies, supporting the historical comparison. But it gives medians of 23 and 24 for studies published in 2017 and 2018, not 20; §4.2 places an estimated median of 19–20 around 2012–2013. The source does not assess the citing paper's validation studies. (LLM, conf=0.91)

  Experimental — accuracy not yet measured

May combine two claims.

Different samples (or different anatomical areas within the same sample) are not affected homogeneously by preparation steps such as fixation or antibody penetration, which poses challenges in particular for between-subject validation designs.

“Moreover, it is known that different samples (or different anatomical areas within the same sample) are not affected homogeneously by preparation steps such as fixation or antibody penetration (Dawe, Bennett, Schneider, Vasireddi, Arfanakis, 2009, Seehaus, Roebroeck, Bastiani, Fonseca, 2015), which poses challenges in particular for between-subject validation designs.”

• cites Dawe, Bennett, Schneider, Vasireddi, Arfanakis, 2009 sourcePartly supported

  Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured
• cites Seehaus, Roebroeck, Bastiani, Fonseca, 2015 sourcePartly supported

  Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured

May combine two claims.

Higher resolution can be achieved with ex vivo imaging, which in studies of animals with smaller brains provides more anatomically comparable detail to human studies with larger voxel sizes.

“Also, higher resolution can be achieved with ex vivo imaging, which in studies of animals with smaller brains provides more anatomically comparable detail to human studies with larger voxel sizes (Lerch et al., 2012).”

• cites Lerch et al., 2012 sourceSupported

May combine two claims.

Fixation works by linking amino acids in proteins, which means that the molecular tissue structure is inherently changed during the process.

“This is because fixation works by linking amino acids in proteins, which means that the molecular tissue structure is inherently changed during the process (Thavarajah et al., 2012).”

• cites Thavarajah et al., 2012 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Using staining fraction to quantify myelin could circumvent the issue of inhomogeneous preparation effects, if the image segmentation intensity threshold is adjusted flexibly depending on local intensity variations.

“Using staining fraction to quantify myelin could circumvent this issue, if the image segmentation intensity threshold is adjusted flexibly (Mollink et al., 2019), depending on local intensity variations.”

• cites Mollink et al., 2019 sourcePartly supported

  Methods > Immunohistochemistry (page unavailable) describes using the stained area fraction to quantify PLP, a myelin marker, and setting intensity thresholds separately for each specimen to account for varying staining intensity. It does not describe adjusting the threshold to local intensity variations within a specimen; the local 100 × 100-pixel calculation is for stained area fraction, not threshold selection. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

May combine two claims.

MR relaxation relies on water protons behaving differently depending on their molecular environment, and fixation does change this environment, for example by affecting membrane permeability.

“Moreover, MR relaxation relies on water protons behaving differently depending on their molecular environment, and fixation does change this environment, for example by affecting membrane permeability (Shepherd et al., 2009).”

• cites Shepherd et al., 2009 sourceSupported

May combine two claims.

Advanced diffusion MRI models (Spherical Mean Technique, NODDI, diffusion basis spectrum imaging) are becoming increasingly popular compared to tensor-based metrics such as FA, but are vastly under-represented in the validation literature.

“advanced models, such as the Spherical Mean Technique (Kaden et al., 2016), neurite orientation dispersion and density imaging (NODDI; Zhang et al. (2012)), and diffusion basis spectrum imaging (Wang et al., 2011), are becoming increasingly popular compared to tensor-based metrics such as FA, but are vastly under-represented in the validation literature (Alexander, Dyrby, Nilsson, 2019, Caspers, A”

• cites Kaden et al., 2016 sourceSupported
• cites Zhang et al. (2012) sourceNot checked — no full text of the source
• cites Wang et al., 2011 sourceSupported

May combine two claims.

For studies using human brains, the option to combine in vivo scanning with ex vivo histology is rare, whereas in animal studies, both in vivo and ex vivo scanning are possible.

“For studies using human brains, the option to combine in vivo scanning with ex vivo histology is rare (see Treit et al., 2019), whereas in animal studies, both in vivo and ex vivo scanning are possible.”

• cites Treit et al., 2019 sourcePartly supported

  The Participants and Histology sections report that four patients had presurgical in vivo MRI followed by surgical resection and histology, so the paper provides a human example of combining the two. It does not establish that this opportunity is rare across studies using human brains. The animal-studies comparison is outside the clause to which this citation is attached. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

Advances in automatic co-registration tools could aid more widespread implementation in validation studies.

“Advances in automatic co-registration tools could aid a more widespread implementation in validation studies (Huszar et al., 2019)”

• cites Huszar et al., 2019 sourceSupported

May combine two claims.

Repetition of MRI metrics across time, hardware, and sequences is important and a key prerequisite for accuracy.

“In the case of MRI metrics, repetition across time (e.g. Arshad, Stanley, Raz, 2017, Lévy, Guertin, Khatibi, Mezer, Martinu, Chen, Stikov, Rainville, Cohen-Adad, 2018), across hardware (e.g. Bane, Hectors, Wagner, Arlinghaus, Aryal, Cao, Chenevert, Fennessy, Huang, Hylton, Kalpathy-Cramer, Keenan, Malyarenko, Mulkern, Newitt, Russek, Stupic, Tudorica, Wilmes, Yankeelov, Yen, Boss, Taouli, 2018, Le”

• cites Arshad, Stanley, Raz, 2017 sourceSupported
• cites Bane et al., 2018 sourcePartly supported

  The Methods and Results assess T1 measurement repeatability and interplatform reproducibility across scanners, alongside accuracy and differences between protocols. This supports the hardware-reproducibility part of the claim, for which this paper is cited. The passages do not establish that repetition is a prerequisite for accuracy. (LLM, conf=0.82)

  Experimental — accuracy not yet measured
• cites Stikov et al., 2015aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Myelin composition can vary between tissue types, species, and pathologies.

“Its composition can vary between tissue types (González de San Román et al., 2018), species (Gopalakrishnan et al., 2013) and in pathologies (Wheeler et al., 2008)”

• cites González de San Román et al., 2018 sourcePartly supported

  Only the tissue-types clause is attributed to this paper; the species and pathology clauses cite other works. The Results section on lipid imaging and the Discussion section on lipid distribution report different lipid profiles in gray matter, white matter, and cortical layers. They do not establish that the composition of myelin itself varies between tissue types (page numbers unavailable). (LLM, conf=0.78)

  Experimental — accuracy not yet measured
• cites Gopalakrishnan et al., 2013 sourceNot checked — no full text of the source
• cites Wheeler et al., 2008 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Methods developed for quantification, rather than visualisation, of molecules (PIXE and lipid mass spectrometry) could provide quantification of myelin for each imaging pixel, for example by estimating myelin concentration from sulfur and phosphorus.

“One solution to address the limitations of these commonly used metrics would be to employ methods developed for quantification, rather than visualisation, of molecules. These methods include proton induced X-ray emission (PIXE, Stüber et al. (2014)) and lipid mass spectrometry (González de San Román et al., 2018), which could provide quantification of myelin for each imaging pixel, for example by”

• cites Stüber et al. (2014) sourceNot checked — no full text of the source
• cites González de San Román et al., 2018 sourcePartly supported

  Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured

May combine two claims.

Coregistration of MR and histological images has unique challenges: different spatial scales, morphometry changes due to tissue processing, and potential cracks and folds in tissue sections.

“While coregistration of MR images and histological images has unique challenges, such as the different spatial scales, changes in morphometry due to the tissue processing for histology, and potential cracks and folds in the tissue sections (Pichat et al., 2018)”

• cites Pichat et al., 2018 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

A few studies have aimed to use non-correlative approaches for validation of MR metrics, e.g., demonstrating effects from intervention on MR and histology, but fall outside scope.

“A few studies have aimed to use non-correlative approaches for validation of MR metrics, for example by demonstrating effects from a given intervention on MR and histology”

• cites Lodygensky et al., 2012 sourcePartly supported

  The “Myelin quantification in shiverer mice” section reports differences between mutant and control mice in both MR phase imaging and histological myelin staining, supporting a group-comparison approach alongside correlation. But the paper does not show an administered intervention producing changes in both measures: its iron-extraction experiment changed iron staining while phase contrast remained unchanged (Fig. 6). (LLM, conf=0.82)

  Experimental — accuracy not yet measured
• cites Sampaio-Baptista et al., 2013 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

With sufficient image quality and appropriate image segmentation, higher-resolution methods have the potential to estimate the myelin fraction of imaged tissue, as well as the packing density and total surface area of the myelin sheaths.

“With sufficient image quality and appropriate image segmentation (Stikov, Campbell, Stroh, Lavelee, Frey, Novek, Nuara, Ho, Bedell, Dougherty, Leppert, Boudreau, Narayanan, Duval, Cohen-Adad, Picard, Gasecka, Cote, Pike, 2015, Zaimi, Duval, Gasecka, Côté, Stikov, Cohen-Adad, 2016), these methods have the potential to estimate the myelin fraction of imaged tissue, as well as the packing density and”

• cites Stikov, Campbell, Stroh, Lavelee, Frey, Novek, Nuara, Ho, Bedell, Dougherty, Leppert, Boudreau, Narayanan, Duval, Cohen-Adad, Picard, Gasecka, Cote, Pike, 2015Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Zaimi, Duval, Gasecka, Côté, Stikov, Cohen-Adad, 2016 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Multiple different morphological aspects of the myelinated axon, including Nodes of Ranvier, sheath thickness, length and packing density, are all important to brain function.

“multiple different morphological aspects of the myelinated axon, including Nodes of Ranvier, sheath thickness, length and packing density - are all important to brain function (Almeida, Lyons, 2017, Xin, Chan, 2020).”

• cites Almeida, Lyons, 2017 sourcePartly supported

  The section on neuronal circuit function says that nodes of Ranvier and the length, thickness, number, and distribution of myelin sheaths can affect conduction and circuit function. It does not specifically address packing density as a morphological feature, so that item in the list is not established by this source. (LLM, conf=0.82)

  Experimental — accuracy not yet measured
• cites Xin, Chan, 2020 sourcePartly supported

  In “Oligodendrocytes and circuit function” (page unavailable), the review links nodes of Ranvier and myelin sheath length to conduction timing in neural circuits. It also discusses changes in myelin thickness under “Experience-induced plasticity,” but the passages found do not establish the claimed functional importance of sheath thickness or packing density. (LLM, conf=0.85)

  Experimental — accuracy not yet measured

May combine two claims.

In Shiverer mice, MBP-immunohistochemistry is well suited to demonstrate genetic intervention but could lead to biased myelin quantification in a validation study.

“In studies using Shiverer mice, which have a genetic mutation for the MBP-gene (Molineaux et al., 1986), MBP-immunohistochemistry is well suited to demonstrate the genetic intervention, but could lead to a biased myelin quantification when used on Shiverer mice in a validation study.”

• cites Molineaux et al., 1986 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Not all asymmetric funnel plots are due to issues that impact meta-analytic inference.

“However, not all asymmetric funnel plots are due to issues that impact meta-analytic inference (Sterne et al., 2011).”

• cites Sterne et al., 2011 sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

Publication bias and questionable research practices have long been established as a key factor hindering pooling of evidence across studies in biomedical research.

“Publication bias and questionable research practices have long been established as a key factor hindering pooling of evidence across studies in biomedical research (Ahmed et al., 2012).”

• cites Ahmed et al., 2012 sourceSupported

May combine two claims.

Jespersen et al. (2010) found a correlation between FA and myelin stain of r = 0.78, and between FA and cell density of r = 0.72, questioning specificity of FA-myelin correlation.

“For example, Jespersen et al. (2010) finds a correlation between FA and myelin stain of r = 0.78, and a correlation between FA and cell density of r = 0.72, thus questioning the specificity of the FA-myelin correlation in the study.”

• cites Jespersen et al. (2010) sourcePartly supported

  The Results and discussion section reports correlations between FA and AMG myelin staining, and says FA is also highly correlated with cell soma density. It also describes reasons FA may not specifically reflect myelin. The passages available from our copy do not give the claimed coefficients of r = 0.78 and r = 0.72, so those exact values cannot be verified. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

May combine two claims.

Conditions such as MS, Cuprizone-fed and Shiverer mice are often used in validation studies because of known effect on myelin, but they also affect iron metabolism and axons.

“Conditions such as MS, or animal models such as Cuprizone-fed and Shiverer mice, are often used in validation studies, because of their known effect on myelin, but they also affect iron metabolism, due to iron’s role in myelin maintenance (e.g. Hametner, Wimmer, Haider, Pfeifenbring, Brück, Lassmann, 2013, Pandur, Pap, Varga, Jánosa, Komoly, Fórizs, Sipos, 2019, Sergeant, Vesvres, Devès, Guillou,”

• cites Hametner et al., 2013 sourcePartly supported

  The Discussion reports altered iron metabolism in MS lesions and normal-appearing white matter, including iron loss from myelin. The Results also describe iron accumulation in axons at MS lesion edges. The paper does not establish that cuprizone-fed or Shiverer mice have these effects or that these conditions are often used in validation studies. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites Pandur et al., 2019 sourcePartly supported

  The Discussion describes cuprizone-fed mice as a demyelination model and reports changes in iron metabolism after cuprizone treatment. The Introduction also describes axonal injuries in cuprizone-affected brain regions and disturbances of iron homeostasis in MS. Our copy does not address Shiverer mice or establish that these conditions are often used in validation studies. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites Stassart et al., 2018 sourcePartly supported

  The section on axonal pathology caused by oligodendroglial defects (page unavailable) says shiverer mice have severely abnormal CNS myelin and altered axonal calibre and cytoskeleton. It also says degenerative axonal changes are lacking in these mice. Searches for cuprizone, iron, and validation studies found no passage supporting those parts of the claim. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

May combine two claims.

The validation literature uses a wide range of histological markers, with 9 studies using more than one marker.

“The validation literature uses a wide range of histological markers (with 9 studies using more than one marker, Fig. 3F).”

May combine two claims.

Validation studies use varying methodologies and have varying outcomes, making it difficult to assess to what extent justifications for employing a specific microstructural MRI measure are valid.

“However, validation studies use varying methodologies and have varying outcomes, making it difficult to assess to what extent such justifications are valid.”

• AssumptionRests on an assumption: Neuroimaging studies conducted in vivo often use individual validation studies as evidence to justify employing a specific microstructural MRI measure to study myelination.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Abstract screening led to exclusion of 253 studies, while full-text screening further excluded 62 studies, with a total of 71 remaining studies fitting our inclusion criteria.

“Abstract screening led to exclusion of 253 studies, while full-text screening further excluded 62 studies, with a total of 71 remaining studies fitting our inclusion criteria (Table 1).”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Of these articles, 294 were from Pubmed, 46 from Scopus, 45 from our expert library.

“Of these articles, 294 were from Pubmed, 46 from Scopus, 45 from our expert library.”

May combine two claims.

Formal comparisons between MRI-based myelin markers are limited by methodological variability, inconsistent reporting and potential for publication bias.

“Formal comparisons between MRI-based myelin markers are limited by methodological variability, inconsistent reporting and potential for publication bias.”

May combine two claims.

Many MR markers are sensitive to myelin, including FA, RD, MP, MTR, Susceptibility, R1, but not MD.

“We find many MR markers are sensitive to myelin, including FA, RD, MP, MTR, Susceptibility, R1, but not MD.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Our search yielded 386 unique articles.

“Our search yielded 386 unique articles (Fig. 2).”

Probably not a substantial claim: may not pass S3 (load-bearing).

For R1, MP and MTR, a meta-analysis on within-subject correlations could be performed.

“For the three metrics R1, MP and MTR, a meta-analysis on within-subject correlations could be performed.”

May combine two claims.

The remainder of the MWF studies (n = 5) use different methodologies and outcome measures from each other, making it difficult to pool them together.

“The remainder of the studies (n = 5) use different methodologies (within- and between-subject design) and outcome measures (Spearman and Pearson) from each other, making it difficult to pool them together.”

Classified as probably not a substantial claim.

A notable absence from the meta-analyses is Myelin Water Fraction (MWF).

“A notable absence from our meta-analyses is Myelin Water Fraction (MWF).”

May combine two claims.

Except for mean diffusivity, there is meta-analytic evidence for correlations between histological markers and all markers investigated, with no clear marker having a stronger effect size than others.

“Except for mean diffusivity, we find meta-analytic evidence for correlations between histological markers and all markers investigated, with no clear marker having a stronger effect size than others.”

May combine two claims.

Of the MWF studies, 7 use a mixed within- and between-subject design, 1 does not report the type of static measure used, and 1 uses sample sizes too small for inclusion, which is why they did not fit inclusion criteria for the meta-analysis.

“While 14 studies in total examine MWF, a high number of these studies use a mixed within- and between- subject design (n = 7), do not report the type of static measure used (n = 1), or use sample sizes too small for inclusion (n = 1), which is why they did not fit our inclusion criteria for the meta-analysis.”

May combine two claims.

In the majority of studies assessed, the fixation process ranged from a few days to a few weeks, which means the MR signal was likely very different between studies.

“In the majority of studies we assessed, the fixation process ranged from a few days to a few weeks, which means the MR signal was likely very different between studies.”

May combine two claims.

The studies in the review validate MR metrics with either in vivo or ex vivo MR scanning, each of which has advantages and disadvantages.

“we find that the studies in our review validate MR metrics with either in vivo or ex vivo MR scanning, each of which has advantages and disadvantages.”

May combine two claims.

Of the 32 studies selected for meta-analysis, 24 reported Pearson only, 7 reported Spearman only, and 1 reported both; 22 used a between-subject design and 10 used a within-subject design.

“A total of 32 studies were selected for inclusion in a quantitative meta-analysis based on this selection process (24 reporting Pearson only, 7 reporting Spearman only, and 1 reporting both; 22 using between-subject design, 10 using within-subject design).”

May combine two claims.

There is no evidence of abnormal p-value distribution, with a left-skewed p-curve and 80% of positive results providing p-values of 0.01 or lower.

“We find no evidence of abnormal p-value distribution, with a left-skewed p-curve and 80% of positive results providing p-values of 0.01 or lower (Supplementary Fig. S1).”

May combine two claims.

Meta-analyses of 7 commonly used MR markers (FA, RD, MD, susceptibility, R1, MTR and MP) show that there is evidence all these markers correlate with myelin, with the exception of MD.

“Our meta-analyses of 7 commonly used MR markers (FA, RD, MD, susceptibility, R1, MTR and MP) show that there is evidence all these markers correlate with myelin, with the exception of MD.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Most validation studies aim to quantify overall myelin content in a tissue, with the rationale that this is most closely related to the signals measured with MRI, and that for many applications in basic science and clinical translation, total myelin content measures can already be hugely beneficial.

“Most validation studies aim to quantify overall myelin content in a tissue, with the rationale that this is most closely related to the signals measured with MRI, and that for many applications in basic science and clinical translation, total myelin content measures can already be hugely beneficial.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Many studies in our review focus on accuracy.

“Many of the studies in our review focus on accuracy, i.e. the measured parameter being in agreement with the true underlying biological parameter of interest.”

May combine two claims.

Most assessed validation papers try to achieve spatial correspondence without coregistration; only 21 out of 70 studies use coregistration.

“Most of the assessed validation papers try to achieve spatial correspondence without coregistration (only 21 out of 70 studies use coregistration), but rather by manually outlining ROIs in the same anatomical location, identifying these ROIs based on landmarks.”

May combine two claims.

In the review, processing of the tissue for histology varied considerably across studies, with both paraffin embedding and cryosectioning being used with about equal frequency.

“In our review, processing of the tissue for histology varied considerably across studies, with both paraffin embedding and cryosectioning being used with about equal frequency.”

May combine two claims.

Our meta-analysis finds evidence for correlations between myelin histology and FA, RD, susceptibility, R1, MP and MTR, but not MD.

“Our meta-analysis finds evidence for correlations between myelin histology and a range of MR markers: FA, RD, susceptibility, R1, MP and MTR, but not MD.”

• AssumptionRests on an assumption: Myelin histology provides a valid reference standard for myelin content in the validation studies included in the meta-analysis.

May combine two claims.

Within studies selected for systematic review, only a few (n = 7) performed iron histology, while half considered axons (n = 35).

“Within the studies selected for our systematic review, only a few (n = 7) performed iron histology, while half of the studies considered axons (n = 35).”

Classified as probably not a substantial claim.

Heterogeneous and inconsistent reporting makes it difficult to assess quality of studies and factors driving differences in results.

“Second, we find that heterogeneous and inconsistent reporting makes it difficult to assess the quality of the studies, and the factors driving differences in results.”

Classified as probably not a substantial claim. May combine two claims.

Some microstructural features of the brain, such as iron and axonal density, are related to myelin and have similar effects as myelin on MR signals.

“Some microstructural features of the brain, such as iron and axonal density, are related to myelin and have similar effects as myelin on MR signals (Möller et al., 2019).”

• cites Möller et al., 2019 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Abstract screening was independently run by two investigators, and discordant decisions on article inclusions were resolved by discussion and consensus-finding.

“As most studies were excluded at this stage, abstract screening was independently run by two investigators (I.L. and A.L.) and discordant decisions on article inclusions were resolved by discussion and consensus-finding between the investigators.”

May combine two claims.

The validation literature has considered different field strengths, species, and MRI and histology approaches to quantify myelin, making common results across papers highly generalizable.

“Taken together, these factors highlight that the validation literature has considered different field strengths, species, and MRI and histology approaches to quantify myelin, making common results across papers highly generalizable.”

Classified as probably not a substantial claim. May combine two claims.

Stage and type of tissue fixation both affect MR parameters.

“As stage and type of tissue fixation both affect MR parameters (Dusek et al., 2019), we extracted information on the state of the tissue during MR imaging.”

• cites Dusek et al., 2019 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Funnel plots are a common tool to estimate publication bias in meta-analyses.

“Funnel plots are a common tool to estimate publication bias in meta-analyses (Egger et al., 1997).”

• cites Egger et al., 1997 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Tissue treatment for histology (cryosectioning or paraffin embedding) can affect tissue staining success and intensity.

“Last but not least, we report how tissue was treated for histology (cryosectioning or paraffin embedding), which can affect tissue staining success and intensity (Werner et al., 2000).”

• cites Werner et al., 2000 sourceNot checked — no full text of the source

Classified as probably not a substantial claim.

Studies using within-subject and between-subject designs need to be considered in different analyses.

“Moreover, studies using within-subject and between-subject designs need to be considered in different analyses.”

Classified as probably not a substantial claim. May combine two claims.

In within-subject designs, variability is driven by anatomical and tissue differences in myelin content; whereas in between-subject designs, variability is driven by external factors (such as pathology, age or genotype).

“In these two types of designs, variability is driven by different factors: in within-subject designs, variability is driven by anatomical and tissue differences in myelin content; whereas in between-subject designs, variability is driven by external factors (such as pathology, age or genotype).”

Classified as probably not a substantial claim. May combine two claims.

Temperature has an effect on MR relaxation times and diffusion.

“As temperature has an effect on MR relaxation times (Birkl et al., 2016) and diffusion (Dhital et al., 2016), tissue temperature during scanning was extracted.”

• cites Birkl et al., 2016 sourceNot checked — no full text of the source
• cites Dhital et al., 2016 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

If the temperature varied across samples or across ROIs within the same sample, this could induce unwanted variability in the MRI metrics.

“If the temperature varied across samples or across ROIs within the same sample, this could induce unwanted variability in the MRI metrics.”

May combine two claims.

Tissue fixation is a necessary step for stopping tissue decomposition in ex vivo scans, but it also has a major influence on MR signals.

“Tissue fixation is a necessary step for stopping tissue decomposition in ex vivo scans, but it also has a major influence on MR signals.”

May combine two claims.

Some markers are only validated with one type of statistical design, which limits their potential applications.

“Some markers are only validated with one type of statistical design, which limits their potential applications.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

In vivo scanning has the advantage that it better resembles the parameters of in vivo human studies, while ex vivo scanning has the advantage that the tissue is scanned and processed for histology in a similar state.

“In vivo scanning has the advantage that it better resembles the parameters of in vivo human studies, while ex vivo scanning has the advantage that the tissue is scanned and processed for histology in a similar state.”

May combine two claims.

Staining intensity is often assumed to scale linearly with myelin content, but this is not always true, for example in cases where staining saturation can take place.

“Staining intensity is often assumed to scale linearly with myelin content. However, this is not always true, for example in cases where staining saturation can take place (e.g. in silver staining of white matter; Pistorio et al., 2006).”

• cites Pistorio et al., 2006 sourceNot checked — no full text of the source

May combine two claims.

Both staining fraction and average staining intensity come with limitations.

“However, both metrics come with limitations.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The tissue temperature could unintentionally change throughout the scanning session due to gradient heating, making it important to monitor scanning temperature in ex vivo scans.

“Importantly, the tissue temperature could unintentionally change throughout the scanning session due to gradient heating, making it important to monitor scanning temperature in ex vivo scans.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Classical stainings and immunohistochemistry allow imaging large fields of view, which is often needed in a validation study to compensate for the comparatively low spatial resolution of MRI relative to microscopy.

“Their advantage is that they allow imaging large fields of view, which is often needed in a validation study to compensate for the comparatively low spatial resolution of MRI relative to microscopy.”

Probably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.

Validity depends on accuracy and accuracy depends on precision.

“First, validity depends on accuracy and accuracy depends on precision.”

May combine two claims.

Higher-resolution methods provide more detailed information compared to stainings or immunohistochemistry, and could provide further validation for microstructural MR markers.

“While they are not as common in the validation literature, and often not feasible over large areas of the brain, these methods provide more detailed information compared to stainings or immunohistochemistry, and could provide further validation for microstructural MR markers (Sternberger, Itoyama, Kies, Webster DeF., 1978, Vincze, Mázló, Seress, Komoly, Ábrahám, 2008).”

• cites Sternberger, Itoyama, Kies, Webster DeF., 1978 sourceNot checked — no full text of the source
• cites Vincze, Mázló, Seress, Komoly, Ábrahám, 2008 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

LFB is likely attracted by basic amino acids of myelin proteins and by phospholipids.

“For example, LFB is likely attracted by the basic amino acids of myelins proteins and by phospholipids”

• cites de Almeida, Pearse, 1958 sourceNot checked — no full text of the source
• cites Kiernan, 2007 sourceNot checked — no full text of the source
• cites Klüver, Barrera, 1953 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Myelin dry mass consists of about 70% lipids and 30% proteins.

“Myelin is a highly complex substrate, with the dry mass consisting of about 70% lipids and 30% proteins”

• cites Gopalakrishnan, Awasthi, Belkaid, Jr, Liazoghli, Colman, Dhaunchak, 2013 sourceNot checked — no full text of the source
• cites O’Brien, Sampson, 1965 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Gold atoms in Gold stains are chemically reduced by myelin lipids in formaldehyde fixed tissue, leading to black appearance in myelin fibres.

“whereas Gold atoms present in Gold stains are chemically reduced by myelin lipids in formaldehyde fixed tissue, leading to the black appearance of gold staining in myelin fibres”

• cites Schmued, Bowyer, Cozart, Heard, Binienda, Paule, 2008 sourceNot checked — no full text of the source
• cites Schmued, Slikker, 1999 sourceNot checked — no full text of the source

May combine two claims.

The myelin structure in the stained tissue is therefore inherently affected by how it was processed for histology and may not be comparable across studies that used different processing strategies.

“The myelin structure in the stained tissue is therefore inherently affected by how it was processed for histology and may not be comparable across studies that used different processing strategies (Recommendation 2.1).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Automatic atlas-based ROI definitions are recommended practice in the neuroimaging field.

“and automatic atlas-based ROI definitions are recommended practice in the neuroimaging field (Nichols et al., 2017).”

• cites Nichols et al., 2017 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Manual ROI placement may vary depending on the experimenter performing it.

“First, placing ROIs may vary depending on the experimenter performing it”

May combine two claims.

While tissue fixation or freezing do not have a strong effect on lipids, extraction by solvents, such as used in paraffin embedding, extracts most lipids and only retains those covalently bound to protein.

“While tissue fixation or freezing do not have a strong effect on lipids (Carriel et al., 2017a), extraction by solvents, such as used in paraffin embedding, extracts most lipids and only retains those covalently bound to protein (Carriel, Campos, Aneiros-Fernandez, Kiernan, 2017, Kiernan, 2007).”

• cites Carriel et al., 2017aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Carriel, Campos, Aneiros-Fernandez, Kiernan, 2017 sourceNot checked — no full text of the source
• cites Kiernan, 2007 sourceNot checked — no full text of the source

May combine two claims.

Correlation does not equal causation and factors that affect both MRI and histology may artificially inflate or deflate correlation coefficients.

“Likewise, correlation does not equal causation and factors that affect both MRI and histology (such as post-mortem times) may artificially inflate or deflate correlation coefficients.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Coregistration has the advantage that ROI definition can be performed more reproducibly.

“it has the advantage that ROI definition can be performed more reproducibly.”

Classified as probably not a substantial claim.

Precision (low variability in repeated measurements, high reliability) is crucial to validity of a metric.

“However, precision, i.e. low variability in repeated measurements, and thus high reliability, is also crucial to validity of a metric.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Coregistration may greatly improve spatial correspondence between MR and histological images.

“Second, coregistration may greatly improve the spatial correspondence between MR and histological images.”

May combine two claims.

Correlative approaches are key to assessing accuracy in histological validation studies because for most MR metrics there is no mathematical model capable of converting metric units to myelin histology readouts.

“Second, correlative approaches are key to assessing accuracy in histological validation studies, because for most MR metrics, there is no mathematical model capable of converting the units of the metric to readouts from myelin histology.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

Different visualization strategies may be useful for different scenarios.

“and therefore different visualization strategies may be useful for different scenarios.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

It remains poorly understood to what extent staining fraction values are comparable across different staining methods.

“However, it remains poorly understood to what extent staining fraction values are comparable across different staining methods.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Traditional histology methods gain in field of view size but lose in resolution.

“Another disadvantage of traditional histology methods is that what they gain in field of view size they lose in resolution.”

Probably not a substantial claim: may not pass S2 (checkable).

Tackling these issues will be crucial to improving strategies to measure myelin in humans.

“Tackling these issues will be crucial to improving our strategies to measure myelin in humans.”

May combine two claims.

Iron occurs in high concentrations in oligodendrocytes and is important for production and maintenance of myelin.

“Iron occurs in high concentrations in oligodendrocytes and is important for the production and maintenance of myelin (Möller et al., 2019).”

• cites Möller et al., 2019 sourceNot checked — no full text of the source

May combine two claims.

Iron affects some MRI metrics similarly to myelin and its distribution resembles myeloarchitecture.

“It affects some MRI metrics similarly to myelin (Möller et al., 2019) and its distribution resembles myeloarchitecture (Fukunaga et al., 2010).”

• cites Möller et al., 2019 sourceNot checked — no full text of the source
• cites Fukunaga et al., 2010 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S2 (checkable).

For the time being, using multiple microstructural imaging markers in parallel may be the best way to test hypotheses related to myelin in humans in vivo.

“This suggests that for the time being, using multiple microstructural imaging markers in parallel may be the best way to test hypotheses related to myelin in humans in vivo.”

• AssumptionRests on an assumption: Myelin histology provides a valid reference standard for myelin content in the validation studies included in the meta-analysis.

May combine two claims.

In healthy tissue, myelin content is often related to iron and axon density.

“In healthy tissue, myelin content is often related to iron and axon density.”

May combine two claims.

Multiple groups provide correlation with more power, but spurious correlations are a concern when dealing with multiple groups.

“While this provides the correlation with more power, spurious correlations are a concern when dealing with multiple groups.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Asymmetric funnel plots can be driven by many factors; selective outcome reporting and selective analysis reporting are most common.

“Asymmetric funnel plots can be driven by many factors. Selective outcome reporting and selective analysis reporting are the most common.”

May combine two claims.

Pooling multiple ROIs without modelling each contribution independently limits interpretability and fails to take nested structure into account.

“This limits the interpretability of the results, and fails to take the nested structure of the dataset into account.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Sampling variation and chance can lead to an asymmetric funnel plot without real publication bias.

“For instance, in some circumstances sampling variation and chance can lead to an asymmetric funnel plot without real publication bias.”

Probably not a substantial claim: may not pass S2 (checkable).

Using pre-registration may help establish more accurate effect sizes for correlations between MR metrics and myelin histology.

“Therefore, using pre-registration may help establish more accurate effect sizes for the correlations between MR metrics and myelin histology”

Probably not a substantial claim: may not pass S2 (checkable).

Taken together, these observations suggest that measures to prevent publication bias may be useful for future validation studies.

“Taken together, these observations suggest that measures to prevent publication bias may be useful for future validation studies.”

Classified as probably not a substantial claim.

The results on publication bias need to be interpreted with caution.

“These results need to be interpreted with caution”

May combine two claims.

Pre-registered protocols with pre-specified power analyses can strengthen robustness of effect size estimates, and even provide more accurate estimates than meta-analytic analyses themselves.

“Pre-registered protocols with pre-specified power analyses can strengthen the robustness of effect size estimates, and even provide more accurate estimates than meta-analytic analyses themselves (Kvarven et al., 2020).”

• cites Kvarven et al., 2020 sourceNot checked — no full text of the source