CoChat
Sign Up
PreprintCited by 2Dementia and Cognitive Impairment ResearchAlzheimer's disease research and treatmentsAdvanced Neuroimaging Techniques and Applications

White matter microstructural abnormality precedes cortical volumetric decline in Alzheimer’s disease: evidence from data-driven disease progression modelling

Open access
Parker, CS, Weston, PSJ, Zhang, H, Oxtoby, NP, the Alzheimer’s Disease Neuroimaging Initiative·2022·bioRxiv (Cold Spring Harbor Laboratory)
DOI 10.1101/2022.07.12.499784
Open this paper in CoChat — free accountOpen PDF

Abstract

Sequencing the regional progression of neurodegeneration in Alzheimer’s disease (AD) informs disease mechanisms and facilitates identification and staging of individuals at greatest risk of imminent cognitive decline, which may aid the development of early therapeutic interventions. Previous attempts to sequence neurodegeneration have analysed measures of regional volume and identified the initial sites of atrophy. However, focal microstructural alterations in white matter have also been reported in early AD. Yet, the temporal ordering of abnormality in measures of white matter microstructure relative to grey matter volume has not been established. In this study we used event-based modelling of disease progression (EBM) to provide a data-driven evaluation of the temporal sequence of abnormality in markers of white matter microstructure relative to grey matter volume. Regional microstructural metrics derived from diffusion tensor imaging (DTI) and regional volumes from Freesurfer cortical parcellation were obtained from the Alzheimer’s disease Neuroimaging Initiative (ADNI) database for 441 amyloid-positive participants (81 AD-dementia, 159 mild cognitive impairment, 201 cognitively normal). The estimated sequence shows a series of abnormalities in markers of white matter microstructure, followed by sequential grey matter volumetric decline, with no overlap between the two. Analysis of positional variance and cross-validation supports the robustness of our findings. These results provide the first data-driven evidence that markers of white matter microstructural degeneration precede those of cortical volumetric decline in the AD cascade. This prompts a re-evaluation of the view that regional volumetric decline can be used to characterise the very earliest stages of AD neurodegeneration. Instead, we suggest that white matter microstructural markers provide an earlier window into AD neurodegeneration. An early staging system of AD neurodegeneration based on measures of brain microstructure may find application in selecting AD subjects with early but minimal brain damage for clinical trials that aim to prevent cognitive decline.

AI snapshot

Written from the paper’s full text.

TL;DR

Using event-based disease progression modelling in 441 amyloid-positive ADNI participants, white matter microstructural abnormalities were estimated to occur before grey matter volumetric decline, with no overlap between the two groups. The sequence was robust to positional uncertainty and cross-validation, suggesting white matter microstructure may provide an earlier marker of Alzheimer’s disease neurodegeneration.

Key findings

  • The estimated sequence contained 10 white matter microstructural events followed by 4 grey matter volumetric events, with no overlap: increased MD in the body of the fornix was first; decreased FA in the uncinate fasciculus was the last white matter event; decreased entorhinal cortex volume was the first grey matter event, followed by fusiform gyrus, mid-temporal gyrus, and hippocampus.
  • MCMC positional variance was low, and cross-validation across 450 resamples showed the median position of each biomarker event equal to its original position, with the relative ordering of white matter and grey matter markers preserved; entorhinal cortex consistently separated white matter microstructural events from later grey matter volumetric events.
  • Moderate positional variation occurred within white matter and grey matter groups: increased MD in the fornix body was sometimes positioned later, decreased FA in the uncinate fasciculus was sometimes positioned earlier, and reduced hippocampal volume was frequently positioned immediately after entorhinal cortex volume reductions.
  • The authors propose an updated AD cascade model splitting structural brain degeneration into earlier white matter microstructural abnormality and later grey matter volumetric decline, and suggest DTI-based staging is less expensive and non-invasive compared with PET.

Methods

  • Data came from ADNI: 441 amyloid-positive participants (81 AD dementia, 159 mild cognitive impairment, 201 cognitively normal). Regional DTI metrics and Freesurfer cortical volumes were obtained from 3T GE scans; DTI included 5 non-diffusion-weighted images and 41 diffusion-weighted images at b=1000 s/mm², and white matter metrics were averaged in 52 Eve atlas ROIs, with 10 white matter ROIs and 4 grey matter ROIs used in the final model.
  • Event-based modelling estimated a sequence of biomarker abnormalities by maximum likelihood, using 50 stochastic greedy ascent realisations with 1000 iterations each; event distributions were fitted with a kernel density mixture model.
  • Positional uncertainty was quantified with MCMC sampling of 100,000 sequences, and sequence variability was assessed with 450 resamples using stratified repeated 9-fold cross-validation. Left and right hemispheres were averaged, grey matter volumes were normalised by intracranial volume, and for each ROI the DTI metric with earliest abnormality was selected from FA, MD, RD, or AxD.

Limitations

  • Stated: sample-dependent positional variation may be due to AD subtypes with differing spatial patterns of spread, and future work can explore AD neurodegeneration subtypes.
  • Stated: grey matter microstructure was not included because of the confounding effect of CSF partial volume, and future joint analysis of white and grey matter changes in microstructure and volume is needed.
  • Stated/evident: there is no established method for determining the earliest DTI metric per ROI, so the authors selected the metric in the earliest sequence position following ROI- and DTI-specific EBM; the biomarker set was also restricted to 10 white matter ROIs and 4 grey matter ROIs selected from prior studies.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1101/2022.07.12.499784

Cites

53 of 53 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

7 of 15 cited sources checked · 2 supported · 0 flagged · accuracy not yet measured

  • Premisecitation

    Neurodegeneration is the major driver of cognitive decline in Alzheimer's disease.

    “Neurodegeneration, which encompasses brain structural damage at the micro-and macroscopic scale, occurs in both white and grey matter, and is the major driver of cognitive decline in AD (Englund 1998, Jack Jr. et al 1997, Frisoni et al 2002, McDonald et al 2012).”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites Englund 1998 source
    • cites Jack Jr. et al 1997 source

      The Discussion (page unavailable) links greater hippocampal atrophy with more severe memory impairment in AD, describing hippocampal volume as a marker of its degenerative neuroanatomic substrate. It does not establish that neurodegeneration is the major driver of cognitive decline across AD, or assess that causal claim against other contributors. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites Frisoni et al 2002

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

  • Premisecitation

    Cognitive decline is more closely correlated to neurodegeneration than tau accumulation.

    “Firstly, cognitive decline is more closely correlated to neurodegeneration than tau accumulation (Jack Jr. et al 2010).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Jack Jr. et al 2010 source
  • Findingown data

    In the estimated event-based model sequence, white matter microstructural abnormality precedes grey matter volumetric abnormality in Alzheimer's disease.

    “The estimated sequence shows a series of abnormalities in markers of white matter microstructure, followed by sequential grey matter volumetric decline, with no overlap between the two.”
    • Also stated in: § 3.1. The sequence of white matter microstructural abnormality and grey matter volumetric decline in AD · hedged
      “The sequence of white matter microstructural and grey matter volumetric neurodegeneration in AD is shown in Fig. 1, and suggests that white matter microstructural abnormality precedes grey matter volumetric abnormality.”
    • Also stated in: § 4. Conclusion · boosted
      “Our study provides the first data-driven evidence that markers of white matter microstructure become abnormal prior to those of grey matter volume in the AD neurodegenerative cascade.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.
  • Findingown data

    The entorhinal cortex consistently separated white matter microstructural events from later grey matter volumetric events.

    “The entorhinal cortex consistently separated white matter microstructural events from later grey matter volumetric events.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.
  • Interpretationown data

    Based on the observed sequence, abnormalities in markers of white matter microstructure precede measures of volumetric decline in an updated hypothesis of the temporal evolution of brain structural neurodegeneration in Alzheimer's disease.

    “Based on the observed sequence, we provide an updated hypotheses on the temporal evolution of brain structural neurodegeneration in AD, where abnormalities in markers of white matter microstructure precede measures of volumetric decline.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.
  • Classified as probably not a substantial claim. 10
  • May combine two claims. 11

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

    Interpretationcitation

    The early positioning of microstructural abnormality complements the hypothesis that microstructural changes in grey matter precede volumetric decline.

    “The early positioning of microstructural abnormality complements the hypothesis of (Weston et al 2015), who used systematic review to suggest that microstructural changes in grey matter precedes volumetric decline.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Weston et al 2015 source
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.

  • May combine two claims.

    Interpretationcitation

    Reduced volume of the hippocampus was frequently positioned immediately following entorhinal cortex volume reductions, which is consistent with it being recognised as one of the preliminary sites of grey matter volumetric decline in Alzheimer's disease.

    “Furthermore, reduced volume of the hippocampus was frequently positioned immediately following entorhinal cortex volume reductions, which is consistent with it being recognised as one of the preliminary sites of grey matter volumetric decline in AD (Jack Jr. et al 1997).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Jack Jr. et al 1997 source

      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
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.

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

    Premisecitation

    Volumetric decline occurs initially in medial temporal lobe grey matter, following tau protein deposition, and is closely correlated to cognitive decline.

    “These investigations have established that volumetric decline occurs initially in medial temporal lobe grey matter, following tau protein deposition, and is closely correlated to cognitive decline.”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Jack Jr. et al 2010 source

      In “Use of biomarkers to stage AD,” the paper says cerebral atrophy begins in medial temporal limbic areas. “Biomarkers of neurodegeneration” reports a strong correlation between atrophy and cognitive decline. The temporal-ordering section places abnormal CSF tau before MRI atrophy, but does not establish that tau protein deposition itself precedes the volumetric decline. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
    • cites Young et al 2014 source

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

    Premisecitation

    Stage-based selection can identify individuals with similar underlying biochemistry and disease who are likely to respond consistently to a single treatment, thus improving treatment detectability.

    “Furthermore, stage-based selection can identify individuals with similar underlying biochemistry and disease who are likely to respond consistently to a single treatment, thus improving treatment detectability(Oxtoby & Alexander 2017).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Oxtoby & Alexander 2017 source

      In 'Enhancing existing models' (page unavailable), the paper says subtype models could identify more homogeneous cohorts with coherent underlying biochemistry and disease that may respond consistently to one treatment. The claim changes subtype-based selection to stage-based selection. The paper discusses potential benefits for clinical trials, but does not specifically say this would improve treatment detectability. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

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

    Premisecitation

    On the regional level, abnormality is evident in medial temporal lobe areas first (hippocampus and entorhinal cortex), then in lateral temporal lobe areas (mid-temporal gyrus and fusiform gyrus) and finally in frontal lobe areas.

    “On the regional level, abnormality is evident in medial temporal lobe areas first (hippocampus and entorhinal cortex), then in lateral temporal lobe areas (mid-temporal gyrus and fusiform gyrus) and finally in frontal lobe areas (Jack Jr. et al 2010, Young et al 2014).”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites Jack Jr. et al 2010 source

      In “Biomarker abnormalities precede clinical symptoms” (page unavailable), the paper says neurofibrillary tangles in asymptomatic individuals tend to be confined to the entorhinal cortex, whereas they are more widespread in symptomatic individuals. “Biomarkers of neurodegeneration” (page unavailable) says the distribution of MRI atrophy maps onto Braak staging. These passages support an early entorhinal abnormality and later wider involvement, but do not establish the claimed sequence through the mid-temporal and fusiform gyri and finally the frontal lobe. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
    • cites Young et al 2014 source

  • May combine two claims.

    Findingown data

    The estimated event-based model sequence orders the 14 biomarker events as: increased MD in the body of the fornix first, followed by increased AxD in the body of the corpus callosum, increased AxD in the fornix striaterminalis, increased MD in the sagittal stratum, increased AxD in the splenium of the corpus callosum, increased MD in the cingulate gyrus, increased RD in the superior longitudinalf

    “1. increased MD in the body of the fornix, 2. increased AxD in the body of the corpus callosum, 3. increased AxD in the fornix striaterminalis, 4. increased MD in the sagittal stratum, 5. increased AxD in the splenium of the corpus callosum, 6. increased MD in the cingulate gyrus, 7. increased RD in the superior longitudinal fasciculus, 8. decreased FA in the genu of the corpus callosum, 9. incre”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.

  • May combine two claims.

    Findingown data

    Analysis of sequence uncertainty using MCMC shows that positional variance is low, suggesting high confidence in the ordering.

    “Analysis of sequence uncertainty using MCMC shows that positional variance is low, suggesting high confidence in the ordering (Fig. 1).”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.

  • May combine two claims.

    Findingown data

    Cross-validation shows that the estimated sequence is consistent, with the median position of each biomarker event being equal to its position estimated on the original dataset.

    “Cross-validation shows that the estimated sequence is consistent (Fig. 2), with the median position of each biomarker event being equal to its position estimated on the original dataset.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.

  • May combine two claims.

    Findingown data

    In cross-validation, the relative position of markers of white matter microstructure and markers of grey matter volume was preserved, supporting the generalisability of the finding that white matter microstructural abnormality precedes grey matter volume abnormality.

    “Importantly, the relative position of markers of white matter microstructure and markers of grey matter volume was preserved, supporting the generalisability of our findings that WM microstructural abnormality precedes GM volume abnormality.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.

  • May combine two claims.

    Interpretationown data

    The finding that markers of white matter microstructure become abnormal earlier in the Alzheimer's disease cascade than grey matter volumetric decline suggests that methods to stage Alzheimer's disease based on regional grey matter volumes may be omitting earlier microscopic neurodegenerative processes occurring in white matter that are useful for pre-symptomatic staging.

    “This finding suggests that methods to stage AD based on regional grey matter volumes may be omitting earlier microscopic neurodegenerative processes occurring in white matter that is useful for pre-symptomatic staging.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.

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

    Interpretationcitation

    The early positioning of the entorhinal cortex is consistent with post-mortem Braak staging, which shows that entorhinal cortex is the earliest site of tau deposition and one of the earliest grey matter regions to undergo neurodegeneration in Alzheimer's disease.

    “Among the grey matter regions, the early positioning of the entorhinal cortex is consistent with post-mortem Braak staging which shows that entorhinal cortex is the earliest site of tau deposition and one of the earliest grey matter regions to undergo neurodegeneration in AD (Braak, H & Braak, E 1991).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Braak, H & Braak, E 1991 source
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.

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

    Interpretationown data

    Our results suggest that measures of microstructure are essential for investigation of the earliest neurodegeneration in Alzheimer's disease.

    “Indeed, our results suggest that measures of microstructure are essential for investigation of the earliest neurodegeneration in AD.”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: The event-based model likelihood assumes an equal probability of each stage in the sequence.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of subjects' data.
    • The authors state this as an assumption: The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.
    • The authors state this as an assumption: There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.

  • Classified as probably not a substantial claim.

    Premisenone given

    The temporal progression of regional white matter microstructural abnormality relative to grey matter volumetric decline has not been established.

    “However, the temporal progression of regional white matter microstructural abnormality relative to grey matter volumetric decline has not been established.”
    • Also stated in: § 3.3. Sequencing white matter microstructural neurodegeneration in AD: novelty, mechanisms and applications · plain
      “However, none have so far determined their positioning in the AD cascade relative to grey matter volumetric decline.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Measures of brain microstructure derived from diffusion tensor imaging are abnormal in white matter regions in prodromal and early phases of Alzheimer's disease.

    “Previous studies have shown that measures of brain microstructure derived from diffusion tensor imaging (DTI) (Basser et al 1994) are abnormal in white matter regions in prodromal and early phases of AD (Sexton et al 2011, Douaud 2013, Yu et al 2017, Fellgiebel et al 2018), suggesting such measures are sensitive to early AD neurodegeneration.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Sexton et al 2011 source
    • cites Douaud 2013

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

    • cites Yu et al 2017

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

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Premisenone given

    Staging markers of regional white matter microstructure has the additional benefit of allowing fine-grained staging, which may allow more detailed monitoring and prognosis of individual disease progression.

    “Staging markers of regional white matter microstructure has the additional benefit of allowing fine-grained staging, which may allow more detailed monitoring and prognosis of individual disease progression.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Premisenone given

    Microstructural imaging uses diffusion-weighted magnetic resonance imaging, which is less expensive than PET imaging and is also non-invasive, making it easier to adopt in clinical settings.

    “Secondly, microstructural imaging uses diffusion-weighted magnetic resonance imaging, which is less expensive than PET imaging and is also non-invasive, which makes it easier to adopt in clinical settings.”
    Quote in the paperNot a substantial claim
16 of 16

  • The authors state this as an assumption.

    The event-based model likelihood assumes an equal probability of each stage in the sequence.

    “Assuming (i) an equal probability of each stage, k, in the sequence, (ii) independence of subjects' data, and (iii) independence of the different biomarker observations at each sequence stage, the likelihood of the sequence is:”

  • The authors state this as an assumption.

    The event-based model likelihood assumes independence of subjects' data.

    “Assuming (i) an equal probability of each stage, k, in the sequence, (ii) independence of subjects' data, and (iii) independence of the different biomarker observations at each sequence stage, the likelihood of the sequence is:”

  • The authors state this as an assumption.

    The event-based model likelihood assumes independence of the different biomarker observations at each sequence stage.

    “Assuming (i) an equal probability of each stage, k, in the sequence, (ii) independence of subjects' data, and (iii) independence of the different biomarker observations at each sequence stage, the likelihood of the sequence is:”

  • The authors state this as an assumption.

    There are no hemisphere-specific disease effects on the diffusion tensor imaging metrics.

    “Regional white matter microstructure biomarkers derived from DTI were averaged over left and right hemispheres, as we did not expect any hemisphere-specific disease effects.”
4 of 4

Cites

Flagged first, then by citations

Research works printed but not linked: 3

  1. Braak, H. and Braak, E. 1991. Neuropathological stageing of Alzheimer-related changes. Acta neuropathologica, 82(4) · DOI · Paper page
  2. Fellgiebel, A. , Schermuly, I. , Gerhard, A. , Keller, I. , Albrecht, J. , Weibrich, C. , Müller, M.J. and Stoeter, P. . Functional relevant loss of long association fibre tracts integrity in early Alzheimer’s disease. Neuropsychologia, 46(6) (2008) · DOI
  3. Lee, S.H. , Coutu, J.P. , Wilkens, P. , Yendiki, A. , Rosas, H.D. , Salat, D.H. and Alzheimer’s disease Neuroimaging Initiative (ADNI). Tract-based analysis of white matter degeneration in Alzheimer’s disease. Neuroscience, 301 (2015) · DOI

Cited by

Sorted by citations

Related

Authors
Parker, C., Weston, P., Zhang, H., Oxtoby, N., & Initiative, T. A. D. N.
Affiliations
  • University College London (GB)
  • UK Dementia Research Institute (GB)
Corresponding author
Parker, CS
ORCID
Zhang, H
Venue
bioRxiv (Cold Spring Harbor Laboratory)
Publisher
openRxiv
Published
2022-07-14
DOI
10.1101/2022.07.12.499784
OpenAlex
W4285594904
Semantic Scholar
8d71628edfe46e7435e0e5805c74abbd533bd0d2
Cited by
2
References
50
Citations by year
2025: 1 · 2023: 1
Type
posted content
Language
en
Open access
Licence
cc-by-nc-nd
Locations
bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
Funders
Medical Research Council (MR/T046473/1)
Topics
  • Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
  • Health Sciences › Medicine › Physiology › Alzheimer's disease research and treatments
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Advanced Neuroimaging Techniques and Applications
Keywords
Alzheimer's disease, white matter microstructure, event-based modeling, diffusion tensor imaging, gray matter atrophy, amyloid-positive, mild cognitive impairment, ADNI

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-11.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-11.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review