Diffusion MRI reveals amyloid plaque–associated cellular signatures through spatial transcriptomic integration
Open accessAbstract
INTRODUCTION: Amyloid plaque deposition is a defining feature of Alzheimer's disease (AD), yet how plaque‐associated cellular alterations are reflected in noninvasive imaging remains unclear. METHODS: We integrated high‐resolution diffusion magnetic resonance imaging (dMRI) with spatial transcriptomics (ST) in an AD mouse model. Spatial correlations were assessed across the whole brain and in disease–control and plaque–non‐plaque comparisons. Neural network models were trained to predict AD‐associated cell types from dMRI metrics. RESULTS: dMRI metrics showed moderate spatial correlations ( r = −0.41 to 0.36) with oligodendrocyte‐ and neuron‐related gene expression. Disease–control comparisons revealed region‐specific MRI‐derived microstructural alterations aligned with transcriptional changes in microglia, astrocytes, oligodendrocytes, and neurons. Plaque‐rich regions exhibited distinct diffusion and susceptibility signatures associated with oligodendrocyte and microglial expression. Neural networks predicted spatial distributions of AD‐associated cell types with 80%–86% accuracy. DISCUSSION: dMRI captures plaque‐associated cellular pathology and supports biologically informed interpretation of imaging contrasts in AD.
AI snapshot
Written from the paper’s full text.
TL;DR
This study integrated high-resolution diffusion MRI (dMRI), quantitative susceptibility mapping (QSM), and spatial transcriptomics in 5xFAD and B6 mice to link imaging metrics with cell-type–associated transcriptional patterns. dMRI metrics showed spatial correlations with neuronal and oligodendrocyte gene expression and supported neural-network prediction of AD-associated cell types, but the work is spatially associative and proof-of-concept.
Key findings
- In 5xFAD mice, 1,059 genes showed significant whole-brain spatial correlations with dMRI metrics (r = −0.41 to 0.36); correlated genes were enriched for neuronal and oligodendrocyte cell types and associated with protein homeostasis, intracellular transport, and molecular binding.
- Regional comparisons showed lower global brain volume in 5xFAD mice (455.00 ± 8.52 mm³) than B6 controls (466.75 ± 6.29 mm³; −11.75 mm³ or −2.52%, p = 0.016), with the most robust MRI alterations in corpus callosum, including FDR-significant reductions in FA, AK, MKT, and RTK and increases in MD, RD, Rsoma, and QSM.
- QSM-defined plaque regions showed distinct diffusion signatures versus non-plaque tissue, with reduced MD and increased NDI among the largest differences (FDR-corrected p < 0.0001), alongside upregulated microglial and downregulated oligodendrocyte-associated genes.
- Neural networks predicted spatial distributions of AD-associated cell types with 80%–86% accuracy; advanced dMRI metrics improved over DTI-only models, with relative accuracy increases of 25.3% for astrocytes, 20.4% for GABAergic neurons, and 6.0% for glutamatergic neurons.
Methods
- Six 7.5-month-old female B6 mice and seven 7.5-month-old female 5xFAD mice underwent ex vivo 9.4T MRI; three brains per group were used for 10x Genomics Visium spatial transcriptomics. Multimodal registration aligned dMRI, QSM, histology, and Allen Mouse Brain Atlas labels, with Dice similarity 0.96 ± 0.03 for ST–MRI and 0.95 ± 0.02 for ST–AMBA alignment.
- Fourteen dMRI metrics from DTI, NODDI, DKI, and SANDI plus QSM were analyzed using Spearman correlations with spatial autocorrelation correction via BrainSMASH and BH-FDR, regional t-tests, plaque-versus-non-plaque Welch’s t-tests, and GO/cell-type enrichment.
- A supervised multilayer perceptron with leave-one-animal-out cross-validation predicted astrocytes, GABAergic neurons, glutamatergic neurons, and oligodendrocytes from dMRI metrics and atlas labels, and was compared with DTI-only, CNN, and ResNet models.
Limitations
- The paper states that analyses are spatially associative and not causal; findings from a transgenic mouse model may not generalize to human AD; QSM-based plaque mapping may be affected by partial-volume effects; long ex vivo dMRI scans were not drift-corrected; SANDI parameters are model-derived diffusion contrasts rather than direct soma/neurite measurements; and limited biological samples, especially for spatial transcriptomics, make the framework a proof of concept.
- The text also makes evident that ST-derived cell-type labels may not fully represent true biological distributions, and that prediction discrepancies may reflect MRI–ST scale mismatch, registration uncertainty, limited sample size, class imbalance, and the indirect relationship between diffusion signals and cell-type composition.
- The paper notes that although MLP was the primary architecture, ResNet achieved higher accuracy for astrocyte prediction, so superiority of a specific architecture is not established.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
50 of 74 cited sources checked · 13 supported · 1 flagged · accuracy not yet measured
- Premisecitation
Aβ plaque deposition generates detectable local magnetic susceptibility variations.
“generating detectable local magnetic susceptibility variations.”
Quote in the paperSubstantial claim2 of 3 sources checked- cites [30] Everett J, Lermyte F, Brooks J, et al. Biogenic metallic elements in the human brain?. Sci Adv. 2021; 7: eabf6707. PMC8189590 34108207 source
The Results and Discussion section, under “Discovery of ferromagnetic elemental iron” (page not provided), reports local magnetic differences within an amyloid plaque: one iron-rich region showed strong XMCD contrast, while another showed no clear dichroism. This supports detectable local magnetic variation associated with plaques, but the paper does not establish that plaque deposition generates magnetic susceptibility variations or report a susceptibility measurement. (LLM, conf=0.84)
Experimental — accuracy not yet measured - cites [31] van Duijn S, Bulk M, van Duinen SG, et al. Cortical iron reflects severity of Alzheimer's disease. J Alzheimers Dis. 2017; 60: 1533– 1545. PMC5676973 29081415 source
The Introduction reports MRI susceptibility changes that correlate with Aβ plaques and notes that plaques accumulate iron. It says plaque iron has been suggested to cause the susceptibility effects, but does not establish that plaque deposition itself generates them; the Conclusion also identifies other sources of iron-related MRI contrast. No page number is available in our copy. (LLM, conf=0.85)
Experimental — accuracy not yet measured - cites [32] Gong N‐J, Dibb R, Bulk M, van der Weerd L, Liu C. Imaging beta amyloid aggregation and iron accumulation in Alzheimer's disease using quantitative susceptibility mapping MRI. Neuroimage. 2019; 191: 176– 185. 30739060 source
- Premisecitation
Most prior work has focused on DTI metrics and linear associations, which may not fully capture compartment-specific microstructural changes.
“most prior work has focused on DTI metrics and linear associations, 39 , 40 , 41 , 42 which may not fully capture compartment‐specific microstructural changes.”
Quote in the paperSubstantial claim2 of 3 sources checked- cites [39] Shen Y, Shen Y, Wang M, et al. A spatial imaging‐transcriptomics paradigm for deciphering the molecular basis of microscopic MRI in the normal brain and Alzheimer's disease. Cell Rep. 2025; 44: 116073. doi: 10.1016/j.celrep.2025.116073 40742809 source
- cites [40] Adewale Q, Khan AF, Carbonell F, Iturria‐Medina Y, Initiative ADN. Integrated transcriptomic and neuroimaging brain model decodes biological mechanisms in aging and Alzheimer's disease. Elife. 2021; 10: e62589. PMC8131100 34002691 source
The Introduction (page unavailable) says an earlier multifactorial model did not consider microscopic properties of brain regions, and Materials and methods > Anatomical connectivity estimation (page unavailable) describes diffusion-based connectivity methods. Neither passage addresses whether most prior work used DTI metrics and linear associations or whether those approaches miss compartment-specific changes. Searches for those terms and paraphrases found no passage making that claim. (LLM, conf=0.76)
Experimental — accuracy not yet measured - cites [41] Torso M, Ridgway GR, Valotti M, Hardingham I, Chance SA, Initiative ADN. In vivo cortical diffusion imaging relates to Alzheimer's disease neuropathology. Alzheimers Res Ther. 2023; 15: 165. PMC10548768 37794477 source
The Methods (MRI analysis; Statistical analyses) show that this study used DTI-derived metrics and linear models for regional associations, so it supports an example of that approach. The paper does not establish that most prior work used it or say that such analyses fail to capture compartment-specific microstructural changes. (LLM, conf=0.83)
Experimental — accuracy not yet measured
- Study design and conductcitation
Aβ-plaque locations in 5xFAD mice were identified by the QSM-Plaque Aβ Detector, which was developed previously with reported precision of 0.86 and recall of 0.94.
“We identified Aβ‐plaque locations in 5xFAD mice by the QSM‐Plaque Aβ Detector which was developed previously with reported precision of 0.86 and recall of 0.94”
Quote in the paperSubstantial claim1 of 1 sources checked- cites [59] Chen J, Han X, Liu Z, et al. Detecting beta‐amyloid plaque via low rank based orthogonal projection and spatial‐spectrum detector using high‐resolution quantitative susceptibility mapping for preclinical studies. IEEE Trans Biomed Eng. 2026; 73(5): 1734– 1745. PMC12469857 40991598 source
The Dataset Description identifies 5xFAD mice, and the High-resolution QSM Slice Results describe plaque locations detected by the QSM-PLAQUE Aβ Detector. The paper reports evaluating precision and recall, but the passages in our copy do not establish the claimed values of 0.86 and 0.94; searches for both numbers did not locate them. Page numbers were not available in the indexed passages. (LLM, conf=0.82)
Experimental — accuracy not yet measured
- Study design and conductcitation
BrainSMASH was used to generate spatially autocorrelated surrogate maps to account for spatial autocorrelation in dMRI maps.
“BrainSMASH 67 was used to generate spatially autocorrelated surrogate maps (n = 1000 per dMRI metric) that matched the spatial dependence of the original dMRI data for each genotype separately”
Quote in the paperSubstantial claim1 of 1 sources checked- cites [67] Burt JB, Helmer M, Shinn M, Anticevic A, Murray JD. Generative modeling of brain maps with spatial autocorrelation. Neuroimage. 2020; 220: 117038. doi: 10.1016/j.neuroimage.2020.117038 32585343 source
- Premisecitation
In AD, spatial transcriptomics and single-cell techniques have revealed plaque-associated transcriptional niches and cell-state changes in situ.
“In AD, spatial transcriptomics and single‐cell techniques have revealed plaque‐associated transcriptional niches and cell‐state changes in situ. 6 , 8”
Quote in the paperSubstantial claim2 of 2 sources checked- cites [6] Mathys H, Davila‐Velderrain J, Peng Z, et al. Single‐cell transcriptomic analysis of Alzheimer's disease. Nature. 2019; 570: 332– 337. PMC6865822 31042697 source
The Abstract and Discussion report single-nucleus transcriptomic analysis identifying AD-associated cell subpopulations and transcriptional changes. Extended Data describes targeted RNA in situ validation, but the paper does not report spatial transcriptomics or establish plaque-associated transcriptional niches in situ. (LLM, conf=0.88)
Experimental — accuracy not yet measured - cites [8] Chen W‐T, Lu A, Craessaerts K, et al. Spatial transcriptomics and in situ sequencing to study Alzheimer's disease. Cell. 2020; 182: 976– 991. 32702314 source
The Introduction reports plaque-induced transcriptional changes measured in situ by spatial transcriptomics, with in situ sequencing providing cellular-resolution validation. But the paper says its spatial transcriptomics lacks single-cell resolution and distinguishes its methods from single-cell approaches; it does not show that single-cell techniques revealed these spatial findings. (LLM, conf=0.86)
Experimental — accuracy not yet measured
- Findingcitation
QSM showed a shift toward more diamagnetic susceptibility in plaque regions.
“QSM showed a shift toward more diamagnetic susceptibility, consistent with altered plaque composition and local susceptibility distribution. 32 , 59”
Quote in the paperNot a substantial claim1 of 2 sources checked- cites [32] Gong N‐J, Dibb R, Bulk M, van der Weerd L, Liu C. Imaging beta amyloid aggregation and iron accumulation in Alzheimer's disease using quantitative susceptibility mapping MRI. Neuroimage. 2019; 191: 176– 185. 30739060 source
- cites [59] Chen J, Han X, Liu Z, et al. Detecting beta‐amyloid plaque via low rank based orthogonal projection and spatial‐spectrum detector using high‐resolution quantitative susceptibility mapping for preclinical studies. IEEE Trans Biomed Eng. 2026; 73(5): 1734– 1745. PMC12469857 40991598 source
- Interpretationown data
Advanced diffusion models capture information relevant to cell-type organization not fully represented by tensor-based metrics.
“Advanced diffusion models therefore capture information relevant to cell‐type organization not fully represented by tensor‐based metrics, extending prior work showing greater sensitivity of SANDI than DTI to AD pathology 88 and spatial correlations between advanced diffusion models and spatial transcriptomics in the developing mouse brain. 43”
Quote in the paperSubstantial claim1 of 2 sources checked- cites [88] Limu Z, Gaughan A, Wang N. A comparative study of DTI and SANDI for discriminating Alzheimer's pathology in a 5xFAD mouse odel. ISMRM. 2024.
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites [43] Han X, Maharjan S, Chen J, et al. High‐resolution diffusion magnetic resonance imaging and spatial‐transcriptomic in developing mouse brain. Neuroimage. 2024; 297; 120734. doi: 10.1016/j.neuroimage.2024.120734 PMC11377129 39032791 source
- Findingown data
Neural networks predicted spatial distributions of AD-associated cell types with 80%-86% accuracy.
“Neural networks predicted spatial distributions of AD‐associated cell types with 80%–86% accuracy.”
Quote in the paperSubstantial claim - Findingown data
dMRI metrics showed moderate spatial correlations (r = −0.41 to 0.36) with oligodendrocyte- and neuron-related gene expression.
“dMRI metrics showed moderate spatial correlations (r = −0.41 to 0.36) with oligodendrocyte‐ and neuron‐related gene expression.”
Quote in the paperSubstantial claim - Findingown data
Plaque-rich regions exhibited distinct diffusion and susceptibility signatures associated with oligodendrocyte and microglial expression.
“Plaque‐rich regions exhibited distinct diffusion and susceptibility signatures associated with oligodendrocyte and microglial expression.”
Quote in the paperSubstantial claim - Findingown data
Disease-control comparisons revealed region-specific MRI-derived microstructural alterations aligned with transcriptional changes in microglia, astrocytes, oligodendrocytes, and neurons.
“Disease–control comparisons revealed region‐specific MRI‐derived microstructural alterations aligned with transcriptional changes in microglia, astrocytes, oligodendrocytes, and neurons.”
Quote in the paperSubstantial claim - Findingown data
Genes significantly associated with at least 10 MRI metrics showed broad spatial correlations across DTI, NODDI, DKI, and SANDI parameters.
“Genes significantly associated with at least 10 MRI metrics showed broad spatial correlations across DTI, NODDI, DKI, and SANDI parameters (Figure 4A).”
Quote in the paperSubstantial claim - Findingown data
The strongest MRI-transcriptomic association strengths obtained from the DKI/SANDI set strongly agreed with those obtained from the full 14-metric set.
“The following reduced‐model sensitivity analysis using only DKI and SANDI metrics showed that the strongest MRI–transcriptomic association strengths obtained from the DKI/SANDI set strongly agreed with those obtained from the full 14‐metric set (Figure S4B)”
Quote in the paperNot a substantial claim - Findingown data
Region-specific alterations in MRI measures were observed across multiple brain areas.
“Region‐specific alterations in MRI measures were observed across multiple brain areas (Figure 5A and B).”
Quote in the paperSubstantial claim - Findingown data
Cell-type enrichment analysis showed strong enrichment for oligodendrocyte cell types among MRI-associated genes in B6 mice.
“Cell‐type enrichment analysis further showed strong enrichment for oligodendrocyte cell types (Figure 4E).”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Findingown data
MRI-associated genes in B6 mice were related to processes involving cell motility, cell migration, cell polarity, and kinase/phosphatase-related molecular functions.
“GO enrichment analysis indicated that these MRI‐associated genes were related to processes involving cell motility, cell migration, cell polarity, and kinase/phosphatase‐related molecular functions (Figure 4D).”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Findingown data
Correlated genes were associated with biological processes related to protein homeostasis, intracellular transport, and molecular binding.
“Functional enrichment analyses indicated that the correlated genes were associated with biological processes related to protein homeostasis, intracellular transport, and molecular binding (Figure 3D)”
Quote in the paperNot a substantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Findingown data
Enrichment of correlated genes was predominantly observed in neuronal and oligodendrocyte cell types.
“with enrichment predominantly observed in neuronal and oligodendrocyte cell types (Figure 3E).”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Findingown data
In B6 mice, whole-brain MRI-transcriptomic correlations showed a distinct pattern dominated by oligodendrocyte-related genes.
“In B6 mice, whole‐brain MRI–transcriptomic correlations showed a distinct pattern dominated by oligodendrocyte‐related genes (Figure 4).”
Quote in the paperSubstantial claim - Findingown data
Global brain volume was modestly but significantly lower (−11.75 mm3 or −2.52%, p = 0.016) in 5xFAD mice (455.00 ± 8.52 mm3) than in B6 controls (466.75 ± 6.29 mm3).
“Global brain volume was modestly but significantly lower (−11.75 mm3 or −2.52%, p = 0.016) in 5xFAD mice (455.00 ± 8.52 mm3) than in B6 controls (466.75 ± 6.29 mm3).”
Quote in the paperNot a substantial claim - Findingown data
A total of 1,059 genes showed significant spatial correlations (r = −0.41∼0.36) across all examined dMRI metrics.
“A total of 1,059 genes showed significant spatial correlations (r = −0.41∼0.36) across all examined dMRI metrics.”
Quote in the paperSubstantial claim - Findingown data
Across the whole brain, MRI metrics showed robust spatial associations with both plaque-upregulated and plaque-downregulated gene sets.
“Across the whole brain, MRI metrics showed robust spatial associations with both plaque‐upregulated and plaque‐downregulated gene sets (Figure S8).”
Quote in the paperSubstantial claim- Rests on an assumption: QSM-defined plaque regions correspond to amyloid plaques.
- Findingown data
Upregulated genes in 5xFAD vs B6 were predominantly associated with microglial and astrocytic populations.
“with upregulated genes predominantly associated with microglial and astrocytic populations (Figure S5)”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Findingown data
The MLP model achieved overall comparable performance across the four cell types compared with CNN and ResNet architectures.
“In addition, comparison with CNN and ResNet architectures showed that the MLP model achieved overall comparable performance across the four cell types.”
Quote in the paperNot a substantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Findingown data
Multiple MRI metrics, including AD, MD, RD, NDI, AK, MKT, RTK, Fextra, Fneurite, Fsoma, and Rsoma, differed significantly between plaque and non-plaque regions.
“Using these QSM‐defined plaque and non‐plaque regions, multiple MRI metrics, including AD, MD, RD, NDI, AK, MKT, RTK, Fextra, Fneurite, Fsoma, and Rsoma, differed significantly between plaque and non‐plaque regions (Figure S7; Table S4).”
Quote in the paperSubstantial claim- Rests on an assumption: QSM-defined plaque regions correspond to amyloid plaques.
- Findingown data
Within individual brain regions, MRI metric alterations showed significant directional concordance with up- or down-regulated gene sets.
“Within individual brain regions, MRI metric alterations showed significant directional concordance with up‐ or down‐regulated gene sets, as determined by permutation testing with BH‐FDR correction (Figure 5D).”
Quote in the paperSubstantial claim - Findingown data
Downregulated genes in 5xFAD vs B6 were enriched for oligodendrocyte-, astrocyte-, and neuron-related markers.
“and downregulated genes enriched for oligodendrocyte‐, astrocyte‐, and neuron‐related markers (Figure S6).”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Findingown data
Combinations of Ctss and Mobp expression patterns and dMRI alterations showed partially overlapping but shifted distributions between QSM-defined plaque and non-plaque regions.
“Combinations of their expression patterns and dMRI alterations showed partially overlapping but shifted distributions between QSM‐defined plaque and non‐plaque regions (Figure 6H).”
Quote in the paperSubstantial claim- Rests on an assumption: QSM-defined plaque regions correspond to amyloid plaques.
- Findingown data
Models trained with advanced MRI metrics consistently outperformed those using DTI metrics alone across four AD-relevant cell types (astrocytes, GABAergic neurons, glutamatergic neurons, and oligodendrocytes).
“Across four AD‐relevant cell types—astrocytes, GABAergic neurons, glutamatergic neurons, and oligodendrocytes—models trained with advanced MRI metrics consistently outperformed those using DTI metrics alone (Figure 7), with notable improvements observed in the Cortex (white arrows), CC (yellow arrows), Thalamus (green arrows), and HPF (blue arrows).”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Findingown data
Joint visualization of dMRI metrics and plaque-associated gene expression patterns showed distinct profiles between QSM-defined plaques and non-plaque regions.
“Joint visualization of dMRI metrics and plaque‐associated gene expression patterns showed distinct profiles between QSM‐defined plaques and non‐plaque regions.”
Quote in the paperSubstantial claim- Rests on an assumption: QSM-defined plaque regions correspond to amyloid plaques.
- Findingown data
Plaque-associated transcriptional differences were observed.
“In parallel, plaque‐associated transcriptional differences were observed (Figure 6E, F).”
Quote in the paperSubstantial claim- Rests on an assumption: QSM-defined plaque regions correspond to amyloid plaques.
- Findingown data
Plaque regions were identified using the QSM-based plaque detector, based on the more diamagnetic susceptibility of amyloid plaques relative to surrounding non-plaque tissue.
“Plaque regions were identified using the QSM‐based plaque detector (Figure 6A), based on the more diamagnetic susceptibility of amyloid plaques relative to surrounding non‐plaque tissue (Figure 6C).”
Quote in the paperSubstantial claim- Rests on an assumption: QSM-defined plaque regions correspond to amyloid plaques.
- Findingown data
Diffusion metrics derived from different models were substantially correlated, confirming partially overlapping information.
“Diffusion metrics derived from different models were also substantially correlated, confirming partially overlapping information.”
Quote in the paperNot a substantial claim - Findingreasoning
Substantial differences remained between predicted and ST-derived maps.
“Performance depends on annotation accuracy, sensitivity, and spatial coverage, and substantial differences remained between predicted and ST‐derived maps.”
Quote in the paperNot a substantial claim - Findingown data
Whole-brain spatial correlation analyses revealed genotype-dependent associations between dMRI metrics and gene expression.
“Whole‐brain spatial correlation analyses revealed genotype‐dependent associations between dMRI metrics and gene expression.”
Quote in the paperSubstantial claim - Findingown data
Across whole-brain, regional, and plaque-associated scales, diffusion and susceptibility metrics showed consistent spatial associations with transcriptional signatures related to neurons, microglia, and oligodendrocytes in white and gray matter.
“Across whole‐brain, regional, and plaque‐associated scales, diffusion and susceptibility metrics showed consistent spatial associations with transcriptional signatures related to neurons, microglia, and oligodendrocytes in white and gray matter.”
Quote in the paperSubstantial claim - Findingown data
Group comparisons demonstrated region-dependent MRI and transcriptional alterations between 5xFAD and B6 mice.
“Group comparisons further demonstrated region‐dependent MRI and transcriptional alterations between 5xFAD and B6 mice.”
Quote in the paperSubstantial claim - Findingown data
Deep learning analyses demonstrated that incorporating advanced diffusion metrics beyond conventional DTI improved spatial prediction of major brain cell types.
“Deep learning analyses demonstrated that incorporating advanced diffusion metrics beyond conventional DTI improved spatial prediction of major brain cell types.”
Quote in the paperSubstantial claim - Findingown data
At the plaque level, plaque-associated regions showed distinct diffusion and susceptibility signatures compared with non-plaque tissue.
“At the plaque level, plaque‐associated regions showed distinct diffusion and susceptibility signatures compared with non‐plaque tissue.”
Quote in the paperSubstantial claim - Findingown data
Deep learning analyses showed that advanced diffusion metrics improved spatial prediction of AD-relevant cell-type distributions beyond conventional tensor-based measures.
“Deep learning analyses further showed that advanced diffusion metrics improved spatial prediction of AD‐relevant cell‐type distributions beyond conventional tensor‐based measures.”
Quote in the paperSubstantial claim - Premisecitation
DTI metrics are not specific to particular cellular or molecular processes.
“However, DTI metrics are not specific to particular cellular or molecular processes.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [16] Jones DK, Knösche TR, Turner R. White matter integrity, fiber count, and other fallacies: the do's and don'ts of diffusion MRI. Neuroimage. 2013; 73: 239– 254. 22846632 source
- Premisereasoning
Spatial transcriptomics approaches require tissue extraction and therefore cannot be applied longitudinally in living patients.
“However, these approaches require tissue extraction and therefore cannot be applied longitudinally in living patients.”
Quote in the paperNot a substantial claim - Interpretationown data
DKI and SANDI preserve most of the MRI-transcriptomic association strength observed with the full metric set.
“indicating that DKI and SANDI preserve most of the MRI–transcriptomic association strength observed with the full metric set.”
Quote in the paperSubstantial claim - Interpretationown data
The high degree of multimodal registration supports the spatial accuracy of subsequent analyses linking dMRI-derived microstructural metrics with cell-type-specific transcriptional profiles.
“This high degree of multimodal registration supports the spatial accuracy of subsequent analyses linking dMRI–derived microstructural metrics with cell‐type–specific transcriptional profiles.”
Quote in the paperSubstantial claim - Interpretationreasoning
The spatially restricted appearance of ST-derived label maps likely reflects the sensitivity and thresholding of the ST-based annotation procedure, spot-level sampling, and registration-based exclusion of unmatched pixels, rather than the complete biological distribution of each cell type.
“The spatially restricted appearance of these maps likely reflects the sensitivity and thresholding of the ST‐based annotation procedure, spot‐level sampling, and registration‐based exclusion of unmatched pixels, rather than the complete biological distribution of each cell type.”
Quote in the paperNot a substantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Study design and conductown data
The full MRI cohort was primarily powered to detect large effects.
“The sensitivity power analysis showed that the full MRI cohort was primarily powered to detect large effects (Table S3).”
Quote in the paperNot a substantial claim - Interpretationown data
Plaque-containing regions in 5xFAD mice are characterized by distinct MRI microstructural and susceptibility profiles that systematically co-vary with microglia- and oligodendrocyte-associated transcriptional alterations.
“Collectively, these results indicate that plaque‐containing regions in 5xFAD mice are characterized by distinct MRI microstructural and susceptibility profiles that systematically co‐vary with microglia‐ and oligodendrocyte‐associated transcriptional alterations.”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Rests on an assumption: QSM-defined plaque regions correspond to amyloid plaques.
- Study design and conductown data
Accurate multimodal co-registration enabled spatially resolved analyses linking diffusion and susceptibility metrics to molecular populations across the whole brain.
“Accurate multimodal co‐registration enabled spatially resolved analyses linking diffusion and susceptibility metrics to molecular populations across the whole brain.”
Quote in the paperSubstantial claim - Interpretationown data
Advanced dMRI metrics can support cell-type prediction across multiple model architectures.
“These results also suggest that advanced dMRI metrics can support cell‐type prediction across multiple model architectures.”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Interpretationown data
Group-level differences between 5xFAD and B6 mice are characterized by region-dependent MRI microstructural changes that systematically co-vary with distinct cell-type-associated transcriptional patterns.
“Together, these results indicate that group‐level differences between 5xFAD and B6 mice are characterized by region‐dependent MRI microstructural changes that systematically co‐vary with distinct cell‐type–associated transcriptional patterns.”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Interpretationreasoning
MRI-derived microstructural alterations spatially correspond to transcriptomic changes in cellular composition or state associated with AD pathology, providing an indirect link to underlying molecular changes, rather than MRI directly detecting gene expression.
“MRI–transcriptomic correspondence likely reflects spatial links between MRI‐derived tissue microstructural alterations and transcriptomic changes in cellular composition or state associated with AD pathology, rather than direct detection of gene expression by MRI.”
- Also stated in: § DISCUSSION · hedged
“Although MRI does not directly measure gene expression, MRI‐derived microstructural alterations may spatially correspond to transcriptomic changes in cellular composition or state, providing an indirect link to underlying molecular changes.”
Quote in the paperSubstantial claim - Also stated in: § DISCUSSION · hedged
- Interpretationreasoning
Spatial correspondence between MRI-derived microstructural alterations and plaque-associated cellular changes suggests coordination within the plaque microenvironment.
“Their spatial correspondence suggests coordination between MRI‐derived microstructural alterations and plaque‐associated cellular changes within the plaque microenvironment.”
Quote in the paperSubstantial claim - Interpretationreasoning
Neuron-associated transcriptional alterations may correspond to dendritic and axonal injury, influencing DKI-, NODDI-, and SANDI-derived measures of microstructural complexity, neurite organization, and soma- or neurite-associated signal fractions.
“Neuron‐associated transcriptional alterations may correspond to dendritic and axonal injury, influencing DKI‐, NODDI‐, and SANDI‐derived measures of microstructural complexity, neurite organization, and soma‐ or neurite‐associated signal fractions.”
Quote in the paperSubstantial claim - Interpretationreasoning
SANDI-derived parameters are model-derived diffusion contrasts rather than direct soma- or neurite-specific biological quantities.
“These parameters should therefore be interpreted as model‐derived diffusion contrasts rather than direct soma‐ or neurite‐specific biological quantities.”
Quote in the paperSubstantial claim - Interpretationreasoning
MRI contrasts are sensitive to coordinated, multiscale AD-associated changes.
“These findings show that MRI contrasts are sensitive to coordinated, multiscale AD‐associated changes.”
Quote in the paperSubstantial claim - Interpretationown data
Diffusion metrics from different models are related measurements of overlapping aspects of tissue restriction, anisotropy, compartmental organization, and neurite-associated signal contributions rather than independent biological readouts.
“Individual metrics therefore should be interpreted as related measurements of overlapping aspects of tissue restriction, anisotropy, compartmental organization, and neurite‐associated signal contributions rather than independent biological readouts.”
Quote in the paperSubstantial claim - Interpretationown data
Neural networks may be able to estimate spatial distributions of neuronal and glial cell types in neurodegenerative disease.
“Compared with deep learning approaches focused on healthy brains, 89 our results provide preliminary evidence that neural networks may be able to estimate spatial distributions of neuronal and glial cell types in neurodegenerative disease.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [89] Han X, Hu R, Liu Z, et al. High‐resolution mri guided whole mouse brain neuronal cell type atlas using deep learning. Commun Biol. 2026. doi: 10.1038/s42003-026-10608-y 42414549 source
- Study design and conductreasoning
The models were trained and validated using ST-derived labels, which may not fully represent true biological distributions of broadly distributed cell populations.
“However, the models were trained and validated using ST‐derived labels, which may not fully represent true biological distributions of broadly distributed cell populations.”
Quote in the paperSubstantial claim - Interpretationreasoning
MRI-transcriptomic associations indicate spatial convergence between molecular pathology and tissue-level imaging phenotypes.
“Thus, MRI–transcriptomic associations indicate spatial convergence between molecular pathology and tissue‐level imaging phenotypes.”
Quote in the paperSubstantial claim
- Classified as probably not a substantial claim. 28
- May combine two claims. 62
Classified as probably not a substantial claim.
Study design and conductcitationOnly genes previously identified as plaque-pathology-associated genes were included in the ST differential expression analysis.
“only genes that are previously identified as plaque‐pathology‐associated genes 8 were included.”
Quote in the paperSubstantial claim1 of 1 sources checked- cites [8] Chen W‐T, Lu A, Craessaerts K, et al. Spatial transcriptomics and in situ sequencing to study Alzheimer's disease. Cell. 2020; 182: 976– 991. 32702314 source
The Introduction (page unavailable) describes ST as measuring genome-wide transcriptomic changes induced by amyloid plaques, not as restricting genes to a previously identified plaque-associated set. The Differential expression analysis methods (page unavailable) describe models based on Aβ intensity and genotype, with no such prior-gene selection. (LLM, conf=0.88)
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measured
May combine two claims.
PremisecitationAβ plaque deposition is accompanied by localized cellular responses involving microglial activation, astrocytic reactivity, neuronal stress, and myelin disruption.
“is accompanied by localized cellular responses involving microglial activation, astrocytic reactivity, neuronal stress, and myelin disruption”
Quote in the paperSubstantial claim3 of 3 sources checked- cites [6] Mathys H, Davila‐Velderrain J, Peng Z, et al. Single‐cell transcriptomic analysis of Alzheimer's disease. Nature. 2019; 570: 332– 337. PMC6865822 31042697 source
The ‘Identification of AD-associated cellular subpopulations’ section reports AD-pathology-associated microglial and astrocyte states, neuronal changes, and stress-related signatures. It does not establish that these responses are localized around Aβ plaques; it describes demyelination only as a possible consequence of CRYAB dysfunction, not an observed myelin disruption. Page numbers are unavailable in our copy. (LLM, conf=0.87)
Experimental — accuracy not yet measured - cites [8] Chen W‐T, Lu A, Craessaerts K, et al. Spatial transcriptomics and in situ sequencing to study Alzheimer's disease. Cell. 2020; 182: 976– 991. 32702314 source
The Discussion (page not available) reports a localized, multicellular response around amyloid plaques involving astrocytes and microglia. The Introduction and oligodendrocyte section report that myelination-related gene expression falls where Aβ accumulation is highest. The copy mentions altered neuronal responses but does not establish neuronal stress specifically, and it shows changes in myelin-related genes rather than direct myelin disruption. (LLM, conf=0.84)
Experimental — accuracy not yet measured - cites [29] Keren‐Shaul H, Spinrad A, Weiner A, et al. A unique microglia type associated with restricting development of Alzheimer's disease. Cell. 2017; 169: 1276– 1290. 28602351 source
The Results (page unavailable) show disease-associated microglia clustered near Aβ plaques in mouse and human AD tissue, supporting a localized microglial response. The passages do not establish plaque-associated astrocytic reactivity, neuronal stress, or myelin disruption. (LLM, conf=0.88)
Experimental — accuracy not yet measured
May combine two claims.
PremisecitationAD is not driven by a single pathological process but by complex interactions among neurons and glial cells-including astrocytes, microglia, and oligodendrocytes-that contribute to synaptic dysfunction, neuroinflammation, white matter injury, and neuronal loss.
“AD is not driven by a single pathological process but rather by complex interactions among neurons and glial cells—including astrocytes, microglia, and oligodendrocytes—that contribute to synaptic dysfunction, neuroinflammation, white matter injury, and neuronal loss.”
Quote in the paperSubstantial claim2 of 3 sources checked- cites [3] Rodrigue KM, Kennedy KM, Park DC. Beta‐amyloid deposition and the aging brain. Neuropsychol Rev. 2009; 19: 436– 450. PMC2844114 19908146 source
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 - cites [4] De Strooper B, Karran E. The cellular phase of Alzheimer's disease. Cell. 2016; 164: 603– 615. 26871627 source
- cites [5] Guo T, Zhang D, Zeng Y, Huang TY, Xu H, Zhao Y. Molecular and cellular mechanisms underlying the pathogenesis of Alzheimer's disease. Mol Neurodegener. 2020; 15: 40. doi: 10.1186/s13024‐020‐00391‐7 PMC7364557 32677986 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
Classified as probably not a substantial claim.
PremisecitationSANDI extends this framework by explicitly modeling soma- and neurite-associated compartments, allowing estimation of soma fraction, neurite fraction, and apparent soma size.
“SANDI extends this framework by explicitly modeling soma‐ and neurite‐associated compartments, allowing estimation of soma fraction, neurite fraction, and apparent soma size.”
Quote in the paperSubstantial claim1 of 1 sources checked- cites [28] Palombo M, Ianus A, Guerreri M, et al. SANDI: a compartment‐based model for non‐invasive apparent soma and neurite imaging by diffusion MRI. Neuroimage. 2020; 215: 116835. PMC8543044 32289460 source
Classified as probably not a substantial claim. May combine two claims.
PremisecitationDKI extends DTI by quantifying non-Gaussian diffusion and is therefore sensitive to microstructural heterogeneity and restricted diffusion arising from complex tissue architecture.
“DKI extends DTI by quantifying non‐Gaussian diffusion and is therefore sensitive to microstructural heterogeneity and restricted diffusion arising from complex tissue architecture.”
Quote in the paperSubstantial claim2 of 2 sources checked- cites [25] Jensen JH, Helpern JA, Ramani A, Lu H, Kaczynski K. Diffusional kurtosis imaging: the quantification of non‐gaussian water diffusion by means of magnetic resonance imaging. Magn Reson Med. 2005; 53: 1432– 1440. 15906300 source
The Conclusion describes DKI as quantifying non-Gaussian diffusion to provide a sensitive measure of tissue structure, and the Discussion links kurtosis to structural complexity and diffusion barriers. But the paper calls DKI an extension of diffusion-weighted imaging (DWI), not diffusion tensor imaging (DTI); it discusses DTI as a distinct technique with different sensitivities. No page numbers are available in our copy. (LLM, conf=0.86)
Experimental — accuracy not yet measured - cites [26] Fieremans E, Jensen JH, Helpern JA. White matter characterization with diffusional kurtosis imaging. Neuroimage. 2011; 58: 177– 188. PMC3136876 21699989 source
Classified as probably not a substantial claim.
PremisecitationSingle-cell RNA sequencing and spatial transcriptomics have transformed understanding of cellular heterogeneity in AD by revealing disease-associated cell states and spatially localized inflammatory responses.
“Recent advances in single‐cell RNA sequencing and spatial transcriptomics (ST) have transformed our understanding of cellular heterogeneity in AD by revealing disease‐associated cell states and spatially localized inflammatory responses.”
Quote in the paperSubstantial claim2 of 3 sources checked- cites [6] Mathys H, Davila‐Velderrain J, Peng Z, et al. Single‐cell transcriptomic analysis of Alzheimer's disease. Nature. 2019; 570: 332– 337. PMC6865822 31042697 source
The Abstract and ‘Identification of AD-associated cellular subpopulations’ section (page numbers unavailable) report that single-nucleus RNA sequencing revealed AD-associated cell subpopulations, including a distinct microglial state. The paper does not report spatial transcriptomics or use it to reveal spatially localized inflammatory responses; it describes immunohistochemistry of selected cells in white matter instead. (LLM, conf=0.88)
Experimental — accuracy not yet measured - cites [7] Grubman A, Chew G, Ouyang JF, et al. A single‐cell atlas of entorhinal cortex from individuals with Alzheimer's disease reveals cell‐type‐specific gene expression regulation. Nat Neurosci. 2019; 22: 2087– 2097. 31768052 source
- cites [8] Chen W‐T, Lu A, Craessaerts K, et al. Spatial transcriptomics and in situ sequencing to study Alzheimer's disease. Cell. 2020; 182: 976– 991. 32702314 source
May combine two claims.
PremisecitationAβ plaque deposition alters tissue microstructure and metal homeostasis.
“altering tissue microstructure and metal homeostasis”
Quote in the paperSubstantial claim3 of 3 sources checked- cites [6] Mathys H, Davila‐Velderrain J, Peng Z, et al. Single‐cell transcriptomic analysis of Alzheimer's disease. Nature. 2019; 570: 332– 337. PMC6865822 31042697 source
The Discussion and the section on cell-type-specific changes report transcriptional and myelination-related responses to AD pathology, not changes in tissue microstructure or metal homeostasis caused by Aβ plaques. Searches for “metal homeostasis” and “tissue microstructure” found no passage addressing those effects; page numbers were unavailable in our copy. (LLM, conf=0.77)
Experimental — accuracy not yet measured - cites [8] Chen W‐T, Lu A, Craessaerts K, et al. Spatial transcriptomics and in situ sequencing to study Alzheimer's disease. Cell. 2020; 182: 976– 991. 32702314 source
The Introduction and the section ‘Oligodendrocytes are part of the cellular phase in the amyloid plaques’ report plaque-associated changes in cells and in expression of myelination genes, which bear on tissue organization. They do not establish a change in tissue microstructure itself, and the searched passages do not address metal homeostasis. Page numbers were unavailable in our copy. (LLM, conf=0.76)
Experimental — accuracy not yet measured - cites [29] Keren‐Shaul H, Spinrad A, Weiner A, et al. A unique microglia type associated with restricting development of Alzheimer's disease. Cell. 2017; 169: 1276– 1290. 28602351 source
The Introduction (page unavailable) describes Aβ plaque deposition, and the Results (page unavailable) describe microglia clustered near plaques. Neither passage says that plaque deposition alters tissue microstructure or metal homeostasis; searches for those terms and for iron, copper, and zinc found no supporting passage in our copy. (LLM, conf=0.73)
Experimental — accuracy not yet measured
May combine two claims.
PremisecitationAdvanced diffusion models-including DKI, NODDI, and SANDI-offer improved sensitivity to microstructural complexity and may provide more biologically informative markers than DTI.
“Advanced diffusion models—including diffusion kurtosis imaging (DKI), neurite orientation dispersion and density imaging (NODDI), and soma and neurite density imaging (SANDI) —offer improved sensitivity to microstructural complexity and may provide more biologically informative markers than DTI.”
Quote in the paperSubstantial claim3 of 3 sources checked- cites [20] Praet J, Manyakov NV, Muchene L, et al. Diffusion kurtosis imaging allows the early detection and longitudinal follow‐up of amyloid‐β‐induced pathology. Alzheimers Res Ther. 2018; 10: 1. PMC6389136 29370870 source
The Background says DKI is more sensitive than DTI to microstructural changes, and the Results report correlations between DKI metrics and underlying amyloid-β pathology. The paper addresses DKI, but not NODDI or SANDI, so it does not support extending that comparison to all three models. Page numbers are unavailable in our copy. (LLM, conf=0.90)
Experimental — accuracy not yet measured - cites [21] Falangola MF, Nie X, Ward R, et al. Diffusion MRI detects basal forebrain cholinergic abnormalities in the 3xTg‐AD mouse model of Alzheimer's disease. Magn Reson Imaging. 2021; 83: 1– 13. PMC8449815 34229088 source
The Introduction (page unavailable) says DKI adds measures associated with tissue microstructural complexity beyond DTI, and Methods says it enables a more comprehensive assessment of the diffusion microenvironment. The paper does not address NODDI or SANDI, so it does not support the claim for all three named models. (LLM, conf=0.88)
Experimental — accuracy not yet measured - cites [22] Whittaker HT, Zhu S, Di Curzio DL, et al. T1, diffusion tensor, and quantitative magnetization transfer imaging of the hippocampus in an Alzheimer's disease mouse model. Magn Reson Imaging. 2018; 50: 26– 37. PMC6461204 29545212 source
The Introduction discusses DTI measures, and Methods > Imaging Data Analysis describes fitting diffusion tensors; neither addresses the claimed advantage of advanced diffusion models. Searches for DKI, NODDI, SANDI, and comparisons of their microstructural sensitivity with DTI found no relevant passage in our copy. Page numbers were not provided. (LLM, conf=0.76)
Experimental — accuracy not yet measured
Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
Study design and conductcitationNo significant signal drift was observed in the dMRI data, so no additional retrospective drift correction was applied.
“No significant signal drift was observed as described in the previous study 49 ; therefore, no additional retrospective drift correction was applied.”
Quote in the paperNot a substantial claim1 of 1 sources checked- cites [49] Wang N, Zhang J, Cofer G, et al. Neurite orientation dispersion and density imaging of mouse brain microstructure. Brain Struct Funct. 2019; 224: 1797– 1813. PMC6565502 31006072 source
May combine two claims.
PremisecitationEx vivo MRI achieves higher anatomic resolution and better tissue contrast than in vivo study, improving detection of microstructural changes, and maintains consistency with ex vivo spatial transcriptomics data.
“Ex vivo MRI was chosen over an in vivo study to achieve higher anatomic resolution and better tissue contrast, which improves the detection of microstructural changes, 45 , 46 and to maintain consistency with ex vivo spatial transcriptomics data.”
Quote in the paperSubstantial claim2 of 2 sources checked- cites [45] To XV, Kurniawan ND, Cumming P, Nasrallah FA. A cross‐comparative analysis of in vivo versus ex vivo MRI indices in a mouse model of concussion. Brain Res. 2023; 1820: 148562. 37673379 source
- cites [46] Holmes HE, Powell NM, Ma D, et al. Comparison of in vivo and ex vivo MRI for the detection of structural abnormalities in a mouse model of tauopathy. Front Neuroinform. 2017; 11: 20. PMC5374887 28408879 source
The Results report higher ex vivo spatial resolution, and the Discussion reports greater gray–white matter contrast than in vivo MRI. The Conclusion says ex vivo MRI may be more sensitive to subtle morphological changes, but the paper does not establish improved detection of microstructural changes. The spatial-transcriptomics rationale follows the citations in the citing sentence and is not attributed to this source. (LLM, conf=0.84)
Experimental — accuracy not yet measured
Probably not a substantial claim: may not pass S3 (load-bearing).
InterpretationcitationThis modest global volume reduction is consistent with prior reports of late-stage morphometric and neurodegenerative changes in 5xFAD mice.
“This modest global volume reduction is consistent with prior reports of late‐stage morphometric and neurodegenerative changes in 5xFAD mice. 75 , 76”
Quote in the paperNot a substantial claim2 of 2 sources checked- cites [75] Yoo C‐H, Kim J, Baek H‐M, Chang K‐A, Choe B‐Y. Neurodegenerative changes in the brains of the 5xFAD Alzheimer's disease model mice investigated by high‐field and high‐resolution magnetic resonance imaging and multi‐nuclei magnetic resonance spectroscopy. Int J Mol Sci. 2023; 24: 5073. PMC10049146 36982146 source
- cites [76] Eimer WA, Vassar R. Neuron loss in the 5XFAD mouse model of Alzheimer's disease correlates with intraneuronal Aβ42 accumulation and Caspase‐3 activation. Mol Neurodegener. 2013; 8: 2. PMC3552866 23316765 source
The Results section (page unavailable) reports significant neuron loss in 5XFAD cortical Layer 5 at 9 and 12 months, supporting the late-stage neurodegenerative part of the claim. The paper counts neurons in specific regions; it does not report global brain-volume reduction or a separate late-stage morphometric change. The modest global reduction is the citing authors’ observation, not a finding attributed to this source. (LLM, conf=0.85)
Experimental — accuracy not yet measured
May combine two claims.
FindingcitationUpregulated genes showed substantial overlap with previously described plaque-induced gene sets, whereas downregulated genes were predominantly drawn from predefined oligodendrocyte-related gene sets.
“Upregulated genes showed substantial overlap with previously described plaque‐induced gene sets, whereas downregulated genes were predominantly drawn from predefined oligodendrocyte‐related gene sets. 8”
Quote in the paperSubstantial claim1 of 1 sources checked- cites [8] Chen W‐T, Lu A, Craessaerts K, et al. Spatial transcriptomics and in situ sequencing to study Alzheimer's disease. Cell. 2020; 182: 976– 991. 32702314 source
The Introduction describes plaque-induced genes (PIGs) and an oligodendrocyte-enriched OLIG network; the Discussion and the section on oligodendrocytes report that OLIG genes become downregulated where amyloid accumulation is highest. These passages support the identity and directional behavior of the cited gene sets, but do not establish that the citing study’s upregulated and downregulated genes overlap those sets to the stated extent. (LLM, conf=0.84)
Experimental — accuracy not yet measured
- Rests on an assumption: QSM-defined plaque regions correspond to amyloid plaques.
May combine two claims.
PremisecitationInterpreting MRI alterations in AD remains challenging because MRI contrasts reflect complex, combined contributions from cellular composition, microstructural organization, and biochemical changes rather than disease-specific processes alone, and provide only indirect measurements of underlying physiological mechanisms.
“Interpreting MRI alterations in AD remains challenging because MRI contrasts reflect complex, combined contributions from cellular composition, microstructural organization, and biochemical changes rather than disease‐specific processes alone, 26 , 79 , 80 and provide only indirect measurements of underlying physiological mechanisms. 81”
Quote in the paperSubstantial claim3 of 4 sources checked- cites [26] Fieremans E, Jensen JH, Helpern JA. White matter characterization with diffusional kurtosis imaging. Neuroimage. 2011; 58: 177– 188. PMC3136876 21699989 source
The Introduction (page unavailable) says diffusion MRI is affected by tissue barriers, their orientation and permeability, and various cell types; it also notes that diffusion tensor imaging lacks specificity to histological features. This supports the microstructural-complexity point, but the paper does not establish the claim about biochemical changes or interpreting MRI alterations specifically in Alzheimer’s disease. The separate claim about indirect measurements is attached to citation 81, not this source. (LLM, conf=0.83)
Experimental — accuracy not yet measured - cites [79] Lampinen B, Szczepankiewicz F, Lätt J, et al. Probing brain tissue microstructure with MRI: principles, challenges, and the role of multidimensional diffusion‐relaxation encoding. Neuroimage. 2023; 282: 120338. 37598814 source
- cites [80] Falangola MF, Guilfoyle DN, Tabesh A, et al. Histological correlation of diffusional kurtosis and white matter modeling metrics in cuprizone‐induced corpus callosum demyelination. NMR Biomed. 2014; 27: 948– 957. PMC5297373 24890981 source
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 - cites [81] Arnatkeviciute A, Fulcher BD, Bellgrove MA, Fornito A. Imaging transcriptomics of brain disorders. Biol Psychiatry Glob Open Sci. 2022; 2: 319– 331. PMC9616271 36324650 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
Classified as probably not a substantial claim. May combine two claims.
PremisecitationStudies integrating neuroimaging with transcriptomic data have shown that large-scale imaging features align with spatial gene-expression patterns, providing molecular context for interpreting neuroimaging findings in health and disease.
“Studies integrating neuroimaging with transcriptomic data have shown that large‐scale imaging features align with spatial gene‐expression patterns, providing molecular context for interpreting neuroimaging findings in health and disease. 82 , 83”
Quote in the paperSubstantial claim2 of 2 sources checked- cites [82] Richiardi J, Altmann A, Milazzo A‐C, et al. Correlated gene expression supports synchronous activity in brain networks. Science. 1979; 348: 1241– 1244. PMC4829082 26068849 source
The Abstract (page unavailable) reports that resting-state fMRI networks can be recapitulated from correlated gene expression, supporting the claimed alignment and molecular context. It also notes disease associations for the identified genes, but the imaging analyses used healthy participants; it does not demonstrate that the alignment holds for neuroimaging findings in disease. (LLM, conf=0.83)
Experimental — accuracy not yet measured - cites [83] Burt JB, Demirtaş M, Eckner WJ, et al. Hierarchy of transcriptomic specialization across human cortex captured by structural neuroimaging topography. Nat Neurosci. 2018; 21: 1251– 1259. PMC6119093 30082915 source
Classified as probably not a substantial claim.
PremisecitationSpatial transcriptomics has enabled characterization of fine-scale, cell-type-specific transcriptional heterogeneity within intact tissue, revealing molecular gradients not captured by region-averaged approaches.
“More recently, spatial transcriptomics has enabled characterization of fine‐scale, cell‐type–specific transcriptional heterogeneity within intact tissue, revealing molecular gradients not captured by region‐averaged approaches. 62 , 84”
Quote in the paperSubstantial claim2 of 2 sources checked- cites [62] Yao Z, van Velthoven CTJ, Kunst M, et al. A high‐resolution transcriptomic and spatial atlas of cell types in the whole mouse brain. Nature. 2023; 624: 317– 332. PMC10719114 38092916 source
The Discussion reports a strong correspondence between cell-type transcriptomic identity and spatial location, and the Non-neuronal and immature neuronal cell types section describes spatial gradients accompanied by changes in astrocyte gene expression. These passages support fine-scale, cell-type-specific heterogeneity in the mouse brain, but our copy does not compare those gradients with region-averaged approaches. (LLM, conf=0.82)
Experimental — accuracy not yet measured - cites [84] Asp M, Bergenstråhle J, Lundeberg J. Spatially resolved transcriptomes—next generation tools for tissue exploration. BioEssays. 2020; 42: 1900221. 32363691 source
Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
Study design and conductcitationAcquisition was not actively drift-compensated, and no retrospective correction was applied because previous analysis showed no significant drift.
“Acquisition was not actively drift‐compensated, and no retrospective correction was applied because previous analysis showed no significant drift 49 ; future studies could nevertheless incorporate drift correction. 90 , 91 , 92”
Quote in the paperNot a substantial claim1 of 1 sources checked- cites [49] Wang N, Zhang J, Cofer G, et al. Neurite orientation dispersion and density imaging of mouse brain microstructure. Brain Struct Funct. 2019; 224: 1797– 1813. PMC6565502 31006072 source
The Methods (MR histology protocol; page unavailable) report that b0 signal varied by about 2.3% during scans, with no notable signal drift. The paper does not say whether acquisition was actively drift-compensated, whether retrospective correction was applied, or whether a drift analysis motivated either decision. (LLM, conf=0.86)
Experimental — accuracy not yet measured
May combine two claims.
InterpretationcitationSignificant directional alignment between MRI and transcriptional changes likely reflects co-occurring neuronal loss, dendritic degeneration, axonal disruption, and myelin damage, together with cellular responses to neuroinflammation and neuropathology.
“Significant directional alignment between MRI and transcriptional changes likely reflects co‐occurring neuronal loss, dendritic degeneration, axonal disruption, and myelin damage, 85 , 86 , 87 together with cellular responses to neuroinflammation and neuropathology. 69”
Quote in the paperSubstantial claim2 of 3 sources checked- cites [85] Qin Y‐Y, Li M‐W, Zhang S, et al. In vivo quantitative whole‐brain diffusion tensor imaging analysis of APP/PS1 transgenic mice using voxel‐based and atlas‐based methods. Neuroradiology. 2013; 55: 1027– 1038. 23644540 source
- cites [86] Müller H‐P, Roselli F, Rasche V, Kassubek J. Diffusion tensor imaging‐based studies at the group‐level applied to animal models of neurodegenerative diseases. Front Neurosci. 2020; 14: 734. PMC7487414 32982659 source
Citation 69 is attached to the proposed cellular responses to neuroinflammation and neuropathology. The Introduction (Neuroimaging of Neurodegenerative Diseases in Humans and Preclinical Models; page unavailable) says cellular mechanisms are not completely understood, while the Applications section links DTI abnormalities to tissue pathology. Searches for neuroinflammation, cellular responses, and MRI–transcriptional alignment found no passage making the attributed connection. (LLM, conf=0.72)
Experimental — accuracy not yet measured - cites [87] Sun S‐W, Liang H‐F, Mei J, Xu D, Shi W‐X. In vivo diffusion tensor imaging of amyloid‐β‐induced white matter damage in mice. J Alzheimers Dis. 2013; 38: 93– 101. PMC3962204 24077431 source
The Results and Discussion report that DTI abnormalities in Aβ-treated mice were associated with axonal loss and myelin damage. They do not establish neuronal loss or dendritic degeneration, and the study did not compare MRI with transcriptional changes. The separate claim about cellular responses is attached to citation 69, not this source. (LLM, conf=0.84)
Experimental — accuracy not yet measured
May combine two claims.
PremisecitationExchange-sensitive NEXI and SMEX/SANDIX models suggest that inter-compartment water exchange and diffusion-time dependence substantially affect gray-matter diffusion signals but are not explicitly modeled by conventional SANDI.
“Although they correlate with the Allen Brain Atlas, 93 exchange‐sensitive NEXI 55 and SMEX/SANDIX 54 models suggest that inter‐compartment water exchange and diffusion‐time dependence substantially affect gray‐matter diffusion signals but are not explicitly modeled by conventional SANDI.”
Quote in the paperNot a substantial claim2 of 3 sources checked- cites [93] Ianuş A, Carvalho J, Fernandes FF, et al. Soma and neurite density MRI (SANDI) of the in‐vivo mouse brain and comparison with the Allen Brain Atlas. Neuroimage. 2022; 254: 119135. 35339686 source
- cites [55] Jelescu IO, de Skowronski A, Geffroy F, Palombo M, Novikov DS. Neurite Exchange Imaging (NEXI): a minimal model of diffusion in gray matter with inter‐compartment water exchange. Neuroimage. 2022; 256: 119277. PMC10363376 35523369 source
- cites [54] Olesen JL, Østergaard L, Shemesh N, Jespersen SN. Diffusion time dependence, power‐law scaling, and exchange in gray matter. Neuroimage. 2022; 251: 118976. PMC8961002 35168088 source
The Methods (Models) describe SMEX/NEXI and SANDIX as exchange models, and the Results (Power-law scaling and time dependence in ex vivo brain) say SANDI is incompatible with the observed time dependence while SMEX describes it well. But the Methods (Diffusion time dependence as an exchange signature) explicitly describe time dependence in conventional SANDI. The claim is therefore supported for exchange and for SANDI’s failure to fit the observed time dependence, not for the assertion that SANDI does not explicitly model time dependence. Page numbers were not provided in our copy. (LLM, conf=0.86)
Experimental — accuracy not yet measured
May combine two claims.
Findingown dataST revealed increased microglial and astrocytic gene expression and decreased oligodendrocyte- and neuron-associated signatures.
“ST revealed increased microglial and astrocytic gene expression and decreased oligodendrocyte‐ and neuron‐associated signatures, consistent with previous immunohistochemical and transcriptomic studies. 6 , 8 , 77 , 78”
Quote in the paperSubstantial claim2 of 3 sources checked- cites [6] Mathys H, Davila‐Velderrain J, Peng Z, et al. Single‐cell transcriptomic analysis of Alzheimer's disease. Nature. 2019; 570: 332– 337. PMC6865822 31042697 source
In “Systematic differential expression analysis” (page not provided), the paper reports predominantly increased expression among differentially expressed genes in microglia and astrocytes, and decreased expression in neurons. It reports predominantly increased—not decreased—expression in oligodendrocytes. The cited study used single-cell transcriptomics, so it supports these findings as a previous transcriptomic study rather than as the source of the ST result. (LLM, conf=0.90)
Experimental — accuracy not yet measured - cites [8] Chen W‐T, Lu A, Craessaerts K, et al. Spatial transcriptomics and in situ sequencing to study Alzheimer's disease. Cell. 2020; 182: 976– 991. 32702314 source
The Identification of the Plaque-Induced Genes module section links the plaque response to activated microglia and inflammatory astrocytes. The Introduction says the oligodendrocyte-enriched OLIG module is depleted under high amyloid accumulation, but activated under mild stress; the Reproducibility section says one neuronal module is downregulated near plaques while another is unchanged. The claim captures the direction of some responses but presents the oligodendrocyte and neuronal decreases too generally. (LLM, conf=0.88)
Experimental — accuracy not yet measured - cites [77] Olabarria M, Noristani HN, Verkhratsky A, Rodríguez JJ. Concomitant astroglial atrophy and astrogliosis in a triple transgenic animal model of Alzheimer's disease. Glia. 2010; 58: 831– 838. 20140958 source
May combine two claims.
FindingcitationST identified upregulation of microglia-associated genes and downregulation of oligodendrocyte-associated genes in plaque regions.
“ST identified upregulation of microglia‐associated genes and downregulation of oligodendrocyte‐associated genes, corresponding to reported enrichment of immune‐response, oxidative‐stress, lysosomal‐function, and myelination pathways. 6 , 8”
Quote in the paperSubstantial claim2 of 2 sources checked- cites [6] Mathys H, Davila‐Velderrain J, Peng Z, et al. Single‐cell transcriptomic analysis of Alzheimer's disease. Nature. 2019; 570: 332– 337. PMC6865822 31042697 source
The Identification of AD-associated cellular subpopulations section reports AD-associated microglial immune markers, and the Discussion reports altered myelination-related genes in oligodendrocytes. But this is a single-cell study, not a spatial transcriptomics (ST) study; the cited passages do not establish gene-expression changes in plaque regions or a general downregulation of oligodendrocyte-associated genes. (LLM, conf=0.85)
Experimental — accuracy not yet measured - cites [8] Chen W‐T, Lu A, Craessaerts K, et al. Spatial transcriptomics and in situ sequencing to study Alzheimer's disease. Cell. 2020; 182: 976– 991. 32702314 source
May combine two claims.
Findingown dataDiffusion changes were most pronounced in corpus callosum, with additional alterations in cortex, hippocampal formation, and olfactory regions.
“Diffusion changes were most pronounced in corpus callosum, with additional alterations in cortex, hippocampal formation, and olfactory regions, consistent with a prior dMRI study of AD mouse models. 48”
Quote in the paperNot a substantial claim1 of 1 sources checked- cites [48] Maharjan S, Tsai AP, Lin PB, et al. Age‐dependent microstructure alterations in 5xFAD mice by high‐resolution diffusion tensor imaging. Front Neurosci. 2022; 16: 964654. doi: 10.3389/fnins.2022.964654 PMC9428354 36061588 source
The Results (Figures 6–7; page unavailable) report age-dependent diffusion changes in the corpus callosum, several cortical regions, and hippocampal subregions. They do not identify the corpus callosum as the most affected region, and a search for olfactory regions found no such finding in our copy. (LLM, conf=0.86)
Experimental — accuracy not yet measured
May combine two claims.
Findingown dataThe top 200 genes displayed positive associations with AD, MD, RD, ODI, Fextra, Fsoma, and Rsoma, and negative associations with FA, NDI, AK, KFA, MKT, RTK, and Fneurite.
“These genes displayed positive associations with AD, MD, RD, ODI, Fextra, Fsoma, and Rsoma, and negative associations with FA, NDI, AK, KFA, MKT, RTK, and Fneurite.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataThe Dice similarity coefficient was 0.96 ± 0.03 for ST-MRI alignment and 0.95 ± 0.02 for ST-AMBA label alignment (n = 6).
“The Dice similarity coefficient (n = 6) was 0.96 ± 0.03 for ST–MRI alignment and 0.95 ± 0.02 for ST–AMBA label alignment.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataIn the HPF of 5xFAD mice, FA was reduced, whereas MD, RD, and Rsoma were increased.
“in the HPF, FA was reduced, whereas MD, RD, and Rsoma were increased;”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataThe most robust animal-level alterations were observed in the CC, where 5xFAD mice showed FDR-significant reductions in FA, AK, MKT, and RTK, together with increases in MD, RD, Rsoma, and QSM.
“The most robust animal‐level alterations were observed in the CC, where 5xFAD mice showed FDR‐significant reductions in FA, AK, MKT, and RTK, together with increases in MD, RD, Rsoma, and QSM.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataRepresentative neuronal marker genes Ppp1r9b and Camk2a exhibited strong spatial correspondence with RTK and MD and significant linear relationships in pixel-wise correlation analyses.
“Representative neuronal marker genes, including Ppp1r9b and Camk2a, exhibited strong spatial correspondence with RTK and MD (Figure 3B) and significant linear relationships in pixel‐wise correlation analyses (Figure 3C).”
Quote in the paperNot a substantial claimMay combine two claims.
Findingown dataSeveral dMRI metrics showed substantial correlations, indicating that the full metric set contains partially overlapping information.
“Several metrics showed substantial correlations, indicating that the full metric set contains partially overlapping information.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataIn the cortex of 5xFAD mice, FA and NDI were reduced, RD and Rsoma were increased, and QSM values were shifted toward lower paramagnetism.
“In the cortex, FA and NDI were reduced, RD and Rsoma were increased, and QSM values were shifted toward lower paramagnetism;”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataData preprocessing, including denoising, improved image quality and tissue contrast, thereby supporting accurate downstream multimodal image registration.
“Data preprocessing, including denoising, improved image quality and tissue contrast, thereby supporting accurate downstream multimodal image registration (Figure S3).”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataIn the OLF of 5xFAD mice, NDI was reduced, while MD, RD, and Rsoma were increased.
“in the OLF, NDI was reduced, while MD, RD, and Rsoma were increased.”
Quote in the paperNot a substantial claimMay combine two claims.
Findingown dataRepresentative oligodendrocyte-associated genes Plp1 and Mbp showed clear spatial correspondence with diffusion metrics, including positive association between Plp1 and NDI and negative association between Mbp and RD.
“Representative oligodendrocyte‐associated genes, including Plp1 and Mbp, showed clear spatial correspondence with diffusion metrics, including a positive association between Plp1 expression and NDI and a negative association between Mbp expression and RD (Figure 4B, C).”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataDiffusivity-related metrics showed strong positive correlations with each other, whereas kurtosis- and neurite-associated metrics formed a separate correlated feature group.
“In particular, diffusivity‐related metrics showed strong positive correlations with each other, whereas kurtosis‐ and neurite‐associated metrics formed a separate correlated feature group.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataAdditional CC changes, including reduced NDI and increased AD and ODI, were observed in the regional comparison.
“Additional CC changes, including reduced NDI and increased AD and ODI, were observed in the regional comparison.”
Quote in the paperNot a substantial claimMay combine two claims.
Findingown dataImprovements in F1-score were most pronounced for astrocytes (68.9%) and oligodendrocytes (7.2%), while macro-averaged F1-scores increased by 29.3% for astrocytes, 17.4% for GABAergic neurons, 7.6% for glutamatergic neurons, and 3.6% for oligodendrocytes.
“Improvements in F1‐score were most pronounced for astrocytes (68.9%) and oligodendrocytes (7.2%), while macro‐averaged F1‐scores increased by 29.3% for astrocytes, 17.4% for GABAergic neurons, 7.6% for glutamatergic neurons, and 3.6% for oligodendrocytes (Figure 7B).”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
May combine two claims.
Findingown dataCell-type enrichment analysis indicated that upregulated genes were primarily linked to microglial populations, while downregulated genes were associated with oligodendrocytes.
“Cell‐type enrichment analysis indicated that upregulated genes were primarily linked to microglial populations, while downregulated genes were associated with oligodendrocytes (Figure 6G).”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
- Rests on an assumption: QSM-defined plaque regions correspond to amyloid plaques.
May combine two claims.
Findingown dataThe use of advanced MRI metrics yielded relative increases in overall accuracy of 25.3% for astrocytes, 20.4% for GABAergic neurons, and 6.0% for glutamatergic neurons.
“In terms of overall accuracy, the use of advanced MRI metrics yielded relative increases of 25.3% for astrocytes, 20.4% for GABAergic neurons, and 6.0% for glutamatergic neurons.”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
May combine two claims.
Findingown dataPlaque-related neuroinflammation-associated gene Ctss and myelin-associated gene Mobp are up-regulated and down-regulated in plaque regions, respectively.
“For example, plaque‐related neuroinflammation‐associated gene Ctss and myelin‐associated gene Mobp are up‐regulated and down‐regulated in plaque regions, respectively.”
Quote in the paperNot a substantial claim- Rests on an assumption: QSM-defined plaque regions correspond to amyloid plaques.
May combine two claims.
Findingown dataResNet achieved higher accuracy for astrocyte prediction, whereas MLP showed comparable or slightly higher performance for the other cell types.
“ResNet achieved higher accuracy for astrocyte prediction, whereas MLP showed comparable or slightly higher performance for the other cell types (Table S5).”
Quote in the paperNot a substantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
May combine two claims.
FindingcitationEspecially in cortex and HPF, downregulated genes were also enriched for astrocytes, consistent with previous studies on astrodegeneration in cortex and hippocampus.
“Especially in cortex and HPF, downregulated genes were also enriched for astrocytes, which is consistent with previous studies on astrodegeneration in cortex and hippocampus. 77 , 78”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [77] Olabarria M, Noristani HN, Verkhratsky A, Rodríguez JJ. Concomitant astroglial atrophy and astrogliosis in a triple transgenic animal model of Alzheimer's disease. Glia. 2010; 58: 831– 838. 20140958 source
- cites [78] Kulijewicz‐Nawrot M, Verkhratsky A, Chvatal A, Sykova E, Rodríguez JJ. Astrocytic cytoskeletal atrophy in the medial prefrontal cortex of a triple transgenic mouse model of Alzheimer's disease. J Anat. 2012; 221: 252– 262. PMC3458630 22738374 source
- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
May combine two claims.
Findingown dataAmong dMRI metrics, NDI and MD showed the biggest differences, with plaque regions exhibiting reduced MD and increased NDI relative to non-plaque tissue.
“Among the dMRI metrics, NDI and MD showed the biggest differences, with plaque regions exhibiting reduced MD and increased NDI relative to non‐plaque tissue (Figure 6B, D).”
Quote in the paperSubstantial claim- Rests on an assumption: QSM-defined plaque regions correspond to amyloid plaques.
May combine two claims.
Findingown dataIn B6 mice, correlated genes tended to show lower expression in gray matter and higher expression in white matter and were primarily oligodendrocyte-related, with weaker correlations.
“In B6 mice, correlated genes tended to show lower expression in gray matter and higher expression in white matter and were primarily oligodendrocyte‐related, with weaker correlations.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataModels trained on advanced MRI-derived microstructural features achieved higher classification accuracy for astrocytic, GABAergic, glutamatergic, and oligodendrocyte populations, with notable improvements in cortical, hippocampal, thalamic, and white matter regions.
“Models trained on advanced MRI‐derived microstructural features achieved higher classification accuracy for astrocytic, GABAergic, glutamatergic, and oligodendrocyte populations, with notable improvements in cortical, hippocampal, thalamic, and white matter regions.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataPlaque regions exhibited increased diffusion restriction, with reduced MD, AD, RD, Fextra, and Fsoma and increased NDI.
“Plaque regions exhibited increased diffusion restriction, with reduced MD, AD, RD, Fextra, and Fsoma and increased NDI.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataIn 5xFAD mice, genes with higher expression in gray matter and lower expression in white matter showed significant correlations with diffusion metrics and were predominantly neuronal and oligodendrocyte-associated.
“In 5xFAD mice, genes with higher expression in gray matter and lower expression in white matter showed significant correlations with diffusion metrics and were predominantly neuronal and oligodendrocyte‐associated, consistent with previous findings. 39”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [39] Shen Y, Shen Y, Wang M, et al. A spatial imaging‐transcriptomics paradigm for deciphering the molecular basis of microscopic MRI in the normal brain and Alzheimer's disease. Cell Rep. 2025; 44: 116073. doi: 10.1016/j.celrep.2025.116073 40742809 source
May combine two claims.
Findingown dataDecreased FA and increased MD were observed across white and gray matter.
“Decreased FA and increased MD across white and gray matter support sensitivity of dMRI to widespread microstructural abnormalities while reinforcing its greater sensitivity in white matter.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
Findingnone givenAlthough MLP was the primary architecture, ResNet performed better for astrocyte prediction.
“Although MLP was the primary architecture, ResNet performed better for astrocyte prediction.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Premisenone givenPrior integrative work has largely focused on diffusion tensor imaging and correlation-based analyses rather than predictive modeling.
“Prior integrative work has largely focused on diffusion tensor imaging and correlation‐based analyses rather than predictive modeling.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationBoth preclinical and clinical studies have reported reduced fractional anisotropy (FA) and increased mean diffusivity (MD) in AD.
“Both preclinical and clinical studies have reported reduced fractional anisotropy (FA) and increased mean diffusivity (MD) in AD.”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [13] Kim KW, MacFall JR, Payne ME. Classification of white matter lesions on magnetic resonance imaging in elderly persons. Biol Psychiatry. 2008; 64: 273– 280. PMC2593803 18471801 source
- cites [17] Chua TC, Wen W, Slavin MJ, Sachdev PS. Diffusion tensor imaging in mild cognitive impairment and Alzheimer's disease: a review. Curr Opin Neurol. 2008; 21: 83– 92. 18180656 source
Classified as probably not a substantial claim.
Premisenone givenAmyloid-β plaque deposition remains a defining pathological feature of AD.
“Amyloid‐β (Aβ) plaque deposition remains a defining pathological feature of AD”
Quote in the paperSubstantial claimClassified as probably not a substantial claim. May combine two claims.
Premisenone givenAdvanced diffusion parameters remain indirect imaging markers rather than direct cell-type measurements, and their interpretation depends on model assumptions, acquisition conditions, tissue fixation, and potential parameter degeneracy.
“Nevertheless, these parameters remain indirect imaging markers rather than direct cell‐type measurements, and their interpretation depends on model assumptions, acquisition conditions, tissue fixation, and potential parameter degeneracy.”
Quote in the paperSubstantial claimClassified as probably not a substantial claim.
PremisecitationNODDI uses a compartment-based model to estimate neurite density and orientation dispersion, providing indices related to axonal and dendritic organization.
“NODDI further uses a compartment‐based model to estimate neurite density and orientation dispersion, providing indices related to axonal and dendritic organization.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [27] Zhang H, Schneider T, Wheeler‐Kingshott CA, Alexander DC. NODDI: practical in vivo neurite orientation dispersion and density imaging of the human brain. Neuroimage. 2012; 61: 1000– 1016. 22484410 source
Classified as probably not a substantial claim.
Premisenone givenStudies directly linking imaging with spatial gene expression remain limited, particularly for amyloid plaque-associated pathology.
“studies directly linking imaging with spatial gene expression remain limited, particularly for amyloid plaque–associated pathology.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationQSM has shown sensitivity to plaque-related iron accumulation and associated tissue changes.
“Quantitative susceptibility mapping (QSM), which measures magnetic susceptibility, has shown sensitivity to plaque‐related iron accumulation and associated tissue changes.”
Quote in the paperNot a substantial claim0 of 3 sources checked- cites [32] Gong N‐J, Dibb R, Bulk M, van der Weerd L, Liu C. Imaging beta amyloid aggregation and iron accumulation in Alzheimer's disease using quantitative susceptibility mapping MRI. Neuroimage. 2019; 191: 176– 185. 30739060 source
- cites [33] Deistung A, Schweser F, Reichenbach JR. Overview of quantitative susceptibility mapping. NMR Biomed. 2017; 30: e3569. 27434134 source
- cites [34] Haacke EM, Liu S, Buch S, Zheng W, Wu D, Ye Y. Quantitative susceptibility mapping: current status and future directions. Magn Reson Imaging. 2015; 33: 1– 25. 25267705 source
May combine two claims.
Interpretationown datadMRI captures plaque-associated cellular pathology and supports biologically informed interpretation of imaging contrasts in AD.
“dMRI captures plaque‐associated cellular pathology and supports biologically informed interpretation of imaging contrasts in AD.”
Quote in the paperSubstantial claim- Rests on an assumption: Findings from the 5xFAD mouse model generalize to human Alzheimer's disease.
May combine two claims.
Premisenone givenThe cellular determinants of plaque-associated diffusion and susceptibility signals remain poorly understood, limiting the interpretability of MRI findings.
“Yet, the cellular determinants of plaque‐associated diffusion and susceptibility signals remain poorly understood, limiting the interpretability of MRI findings.”
Quote in the paperSubstantial claimClassified as probably not a substantial claim.
PremisereasoningWhether advanced diffusion models combined with data-driven approaches can improve cell-type-specific inference remains largely unexplored.
“Whether advanced diffusion models combined with data‐driven approaches can improve cell‐type–specific inference remains largely unexplored.”
Quote in the paperSubstantial claimClassified as probably not a substantial claim.
Study design and conductnone givenSignificantly upregulated and downregulated plaque-associated gene sets were analyzed separately for cell-type enrichment using reference marker gene sets.
“Significantly upregulated and downregulated plaque‐associated gene sets were analyzed separately for cell‐type enrichment using reference marker gene sets.”
Quote in the paperSubstantial claimMay combine two claims.
Interpretationown dataBy linking imaging features to cell-type-associated transcriptional patterns, this work advances the biological interpretability of dMRI and supports the development of noninvasive biomarkers for AD.
“By linking imaging features to cell‐type–associated transcriptional patterns, this work advances the biological interpretability of dMRI and supports the development of noninvasive biomarkers for AD.”
Quote in the paperSubstantial claim- Rests on an assumption: Findings from the 5xFAD mouse model generalize to human Alzheimer's disease.
May combine two claims.
Interpretationown datadMRI-derived microstructural measures capture spatial transcriptomic organization in both genotypes, with 5xFAD showing AD-associated neuronal and oligodendrocyte-related signatures implicated in intracellular organization and protein regulation, whereas B6 primarily showed oligodendrocyte-related MRI-gene correspondence functional in cell regulation.
“Collectively, these findings indicate that dMRI‐derived microstructural measures capture spatial transcriptomic organization in both genotypes, with 5xFAD mice showing AD‐associated neuronal and oligodendrocyte‐related signatures implicated in intracellular organization and protein regulation, whereas B6 mice primarily showed oligodendrocyte‐related MRI–gene correspondence functional in cell reg”
Quote in the paperSubstantial claim- Rests on an assumption: ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
Classified as probably not a substantial claim.
Study design and conductown dataThe accuracy of multimodal registration was evaluated using the Dice similarity coefficient based on 2D hemisphere brain masks.
“The accuracy of multimodal registration was evaluated using the Dice similarity coefficient based on 2D hemisphere brain masks.”
Quote in the paperSubstantial claimClassified as probably not a substantial claim. May combine two claims.
PremisecitationDeep learning models have demonstrated strong performance in AD classification and risk prediction and show growing potential for inferring cellular characteristics from imaging phenotypes.
“Deep learning models have demonstrated strong performance in AD classification and risk prediction and show growing potential for inferring cellular characteristics from imaging phenotypes.”
Quote in the paperSubstantial claim0 of 3 sources checked- cites [35] Khojaste‐Sarakhsi M, Haghighi SS, Ghomi SMTF, Marchiori E. Deep learning for Alzheimer's disease diagnosis: a survey. Artif Intell Med. 2022; 130: 102332. doi: 10.1016/j.artmed.2022.102332 35809971 source
- cites [36] Kaur A, Mittal M, Bhatti JS, Thareja S, Singh S. A systematic literature review on the significance of deep learning and machine learning in predicting Alzheimer's disease. Artif Intell Med. 2024; 154: 102928. doi: 10.1016/j.artmed.2024.102928 39029377 source
- cites [37] Ahmed HM, Elsharkawy ZF, Elkorany AS. Alzheimer disease diagnosis for magnetic resonance brain images using deep learning neural networks. Multimed Tools Appl. 2023; 82: 17963– 17977. doi: 10.1007/s11042-022-14203-1 source
May combine two claims.
Interpretationown dataGroup-level MRI alterations are most strongly supported in the CC, with additional cortical, hippocampal, and olfactory changes suggesting broader but more variable regional involvement.
“Together, these results indicate that group‐level MRI alterations are most strongly supported in the CC, with additional cortical, hippocampal, and olfactory changes suggesting broader but more variable regional involvement.”
Quote in the paperSubstantial claimMay combine two claims.
Interpretationown dataThis work identifies spatial correspondences between molecular and MRI features, contributes to translating genetic findings into interpretable neuroimaging markers, and supports future studies relating to genetic risk to brain phenotypes and behavioral and clinical outcomes across the lifespan.
“This work identifies spatial correspondences between molecular and MRI features, contributes to translating genetic findings into interpretable neuroimaging markers, and supports future studies relating to genetic risk to brain phenotypes and behavioral and clinical outcomes across the lifespan.”
Quote in the paperSubstantial claimMay combine two claims.
Study design and conductown dataAll these changes remained significant after FDR correction and showed raw p-values < 0.05 in the animal-level sensitivity analysis.
“All these changes remained significant after FDR correction (Table S1) and showed raw p‐values < 0.05 in the animal‐level sensitivity analysis (Table S2).”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningThese advances motivate frameworks combining dMRI with spatially resolved transcriptomics to better interpret AD-associated imaging phenotypes in terms of cellular and molecular organization.
“Together, these advances motivate frameworks combining dMRI with spatially resolved transcriptomics to better interpret AD‐associated imaging phenotypes in terms of cellular and molecular organization.”
Quote in the paperNot a substantial claimMay combine two claims.
Study design and conductreasoningPerformance depends on annotation accuracy, sensitivity, and spatial coverage.
“Performance depends on annotation accuracy, sensitivity, and spatial coverage, and substantial differences remained between predicted and ST‐derived maps.”
Quote in the paperSubstantial claimMay combine two claims.
InterpretationreasoningIn plaque-rich regions, microglial activation, neuroinflammation, myelin disruption, and plaque-related susceptibility sources may jointly contribute to reduced MD and increased NDI.
“In plaque‐rich regions, microglial activation, neuroinflammation, myelin disruption, and plaque‐related susceptibility sources may jointly contribute to reduced MD and increased NDI.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Study design and conductnone givenFindings from a transgenic mouse model may not generalize to human AD.
“Second, findings from a transgenic mouse model may not generalize to human AD.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningThe integrative MRI-ST framework should be considered a proof of concept.
“The integrative MRI–ST framework should therefore be considered a proof of concept.”
Quote in the paperNot a substantial claimMay combine two claims.
InterpretationreasoningOligodendrocyte-associated transcriptional patterns were consistently linked to dMRI, whereas associations with other cell types varied by genotype.
“Oligodendrocyte‐associated transcriptional patterns were therefore consistently linked to dMRI, whereas associations with other cell types varied by genotype.”
Quote in the paperSubstantial claimMay combine two claims.
Study design and conductnone givenAβ plaques were identified using high-resolution QSM and mapped onto dMRI metrics; partial-volume effects may reduce plaque classification accuracy.
“Third, Aβ plaques were identified using high‐resolution QSM and mapped onto dMRI metrics; partial‐volume effects may reduce plaque classification accuracy.”
Quote in the paperNot a substantial claimMay combine two claims.
Interpretationown datadMRI is sensitive to widespread microstructural abnormalities and has greater sensitivity in white matter.
“Decreased FA and increased MD across white and gray matter support sensitivity of dMRI to widespread microstructural abnormalities while reinforcing its greater sensitivity in white matter.”
Quote in the paperSubstantial claimMay combine two claims.
Interpretationown dataThese findings provide preliminary evidence that advanced diffusion metrics contain information related to AD-associated cell-type signatures and support their predictive value, rather than establishing definitive cell-type maps or superiority of a specific architecture.
“Thus, these findings provide preliminary evidence that advanced diffusion metrics contain information related to AD‐associated cell‐type signatures and support their predictive value, rather than establishing definitive cell‐type maps or superiority of a specific architecture.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
Study design and conductnone givenThe analyses are spatially associative and do not establish causal relationships between MRI contrasts and specific cellular or molecular mechanisms.
“First, the analyses are spatially associative and do not establish causal relationships between MRI contrasts and specific cellular or molecular mechanisms.”
Quote in the paperNot a substantial claimMay combine two claims.
InterpretationreasoningThe pattern of increased diffusion restriction in plaque regions is consistent with reduced extracellular water, increased microstructural barriers, and greater orientation dispersion, indicating a compact and disorganized local microenvironment.
“This pattern is consistent with reduced extracellular water, increased microstructural barriers, and greater orientation dispersion, indicating a compact and disorganized local microenvironment.”
Quote in the paperSubstantial claimMay combine two claims.
InterpretationreasoningIn 5xFAD mice, reduced oligodendrocyte-associated transcriptional signatures may reflect myelin injury or altered membrane organization, affecting DTI- and NODDI-derived measures of diffusivity and neurite density.
“In 5xFAD mice, reduced oligodendrocyte‐associated transcriptional signatures may reflect myelin injury or altered membrane organization, affecting DTI‐ and NODDI‐derived measures of diffusivity and neurite density.”
Quote in the paperSubstantial claimClassified as probably not a substantial claim. May combine two claims.
Study design and conductnone givenThe limited number of biological samples, particularly for spatial transcriptomics, may reduce power, limit detection of smaller effects, and increase uncertainty about generalizability.
“Finally, the limited number of biological samples, particularly for spatial transcriptomics, may reduce power, limit detection of smaller effects, and increase uncertainty about generalizability.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
InterpretationreasoningSANDI compartments may receive contributions from dendrites, axons, glial processes, myelinated and unmyelinated fibers, cell bodies, extracellular space, and disease-related remodeling.
“SANDI compartments may also receive contributions from dendrites, axons, glial processes, myelinated and unmyelinated fibers, cell bodies, extracellular space, and disease‐related remodeling.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Study design and conductnone givenLong ex vivo dMRI scans may be affected by temporal signal drift.
“Fourth, long ex vivo dMRI scans may be affected by temporal signal drift.”
Quote in the paperNot a substantial claimMay combine two claims.
Interpretationown dataAD pathology appears to alter both the strength and composition of MRI-gene associations, suggesting that tissue remodeling modifies spatial coupling between microstructural measures and transcriptional organization.
“AD pathology appears to alter both the strength and composition of MRI–gene associations, suggesting that tissue remodeling modifies spatial coupling between microstructural measures and transcriptional organization.”
Quote in the paperSubstantial claimMay combine two claims.
Interpretationown dataThese findings provide a framework for biologically informed interpretation of MRI contrasts in neurodegenerative disease and support integrative imaging-transcriptomic approaches for probing tissue organization in AD.
“Together, these findings provide a framework for biologically informed interpretation of MRI contrasts in neurodegenerative disease and support integrative imaging–transcriptomic approaches for probing tissue organization in AD.”
Quote in the paperNot a substantial claim
The paper relies on this without support.
Findings from the 5xFAD mouse model generalize to human Alzheimer's disease.
“Future studies should evaluate translation to human AD and across disease stages to support development of noninvasive biomarkers for early detection, monitoring, and therapeutic assessment.”
The paper relies on this without support.
ST-derived cell-type annotations accurately represent the presence of the corresponding cell types.
“The target labels represent ST‐derived cell‐type annotations based on marker‐gene expression at spatially matched Visium spots/pixels after multimodal registration, rather than direct histological measurements of cell density.”
The paper relies on this without support.
QSM-defined plaque regions correspond to amyloid plaques.
“Plaque regions were identified using the QSM‐based plaque detector (Figure 6A), based on the more diamagnetic susceptibility of amyloid plaques relative to surrounding non‐plaque tissue (Figure 6C).”
Cites
Spin Diffusion Measurements: Spin Echoes in the Presence of a Time-Dependent Field Gradient
Stejskal, E. O., & Tanner, J. E. (1965) · The Journal of Chemical Physics · doi:10.1063/1.1695690 · Cited by 8,239 · see citing works
A Unique Microglia Type Associated with Restricting Development of Alzheimer’s Disease
Keren-Shaul, H., Spinrad, A., Weiner, A., Matcovitch-Natan, O., Dvir-Szternfeld, R., Ulland, T. K., David, E., Baruch, K., Lara-Astaiso, D., Toth, B., et al. (2017) · Cell · doi:10.1016/j.cell.2017.05.018 · Cited by 5,493 · see citing works
ShinyGO: a graphical gene-set enrichment tool for animals and plants
Ge, S. X., Jung, D., & Yao, R. (2020) · Bioinformatics · doi:10.1093/bioinformatics/btz931 · Cited by 5,270 · see citing works
A reproducible evaluation of ANTs similarity metric performance in brain image registration
Avants, B. B., Tustison, N. J., Song, G., Cook, P. A., Klein, A., & Gee, J. C. (2011) · NeuroImage · doi:10.1016/j.neuroimage.2010.09.025 · Cited by 4,796 · see citing works
The basis of anisotropic water diffusion in the nervous system—A technical review.
Beaulieu, C. (2002) · NMR in Biomedicine · doi:10.1002/nbm.782 · Cited by 4,653 · see citing works
Visualization and analysis of gene expression in tissue sections by spatial transcriptomics
Ståhl, P. L., Salmén, F., Vickovic, S., Lundmark, A., Navarro, J. F., Magnusson, J., Giacomello, S., Asp, M., Westholm, J. O., Huss, M., et al. (2016) · Science · doi:10.1126/science.aaf2403 · Cited by 3,834 · see citing works
NODDI: Practical in vivo neurite orientation dispersion and density imaging of the human brain
Zhang, H., Schneider, T., Wheeler-Kingshott, C. A., & Alexander, D. C. (2012) · NeuroImage · doi:10.1016/j.neuroimage.2012.03.072 · Cited by 3,288 · see citing works
Diffusion tensor imaging of the brain
Alexander, A. L., Lee, J. E., Lazar, M., & Field, A. S. (2007) · Neurotherapeutics · doi:10.1016/j.nurt.2007.05.011 · Cited by 2,747 · see citing works
Single-cell transcriptomic analysis of Alzheimer’s disease
Mathys, H., Davila-Velderrain, J., Peng, Z., Gao, F., Mohammadi, S., Young, J. Z., Menon, M., He, L., Abdurrob, F., Jiang, X., et al. (2019) · Nature · doi:10.1038/s41586-019-1195-2 · Cited by 2,568 · see citing works
Jensen, J. H., Helpern, J. A., Ramani, A., Lu, H., & Kaczynski, K. (2005) · Magnetic Resonance in Medicine · doi:10.1002/mrm.20508 · Cited by 2,488 · see citing works
White matter integrity, fiber count, and other fallacies: The do's and don'ts of diffusion MRI
Jones, D. K., Knösche, T. R., & Turner, R. (2013) · NeuroImage · doi:10.1016/j.neuroimage.2012.06.081 · Cited by 2,479 · cited 2× · see citing works
“16 Brain tissue contains multiple barriers and compartments, including cell membranes, neurites, myelin, extracellular space, and pathological deposits, which cause diffusion to deviate from Gaussian behavior.”
The Cellular Phase of Alzheimer’s Disease
De Strooper, B., & Karran, E. (2016) · Cell · doi:10.1016/j.cell.2015.12.056 · Cited by 1,990 · see citing works
RNA sequencing: the teenage years
Stark, R., Grzelak, M., & Hadfield, J. (2019) · Nature Reviews Genetics · doi:10.1038/s41576-019-0150-2 · Cited by 1,906 · see citing works
Shared and distinct transcriptomic cell types across neocortical areas
Tasic, B., Yao, Z., Graybuck, L. T., Smith, K. A., Nguyen, T. N., Bertagnolli, D., Goldy, J., Garren, E., Economo, M. N., Viswanathan, S., et al. (2018) · Nature · doi:10.1038/s41586-018-0654-5 · Cited by 1,746 · see citing works
Looking into the functional architecture of the brain with diffusion MRI.
Le Bihan, D. (2003) · Nature Reviews Neuroscience · doi:10.1038/nrn1119 · Cited by 1,571 · see citing works
PanglaoDB: a web server for exploration of mouse and human single-cell RNA sequencing data
Franzén, O., Gan, L.-M., & Björkegren, J. L. M. (2019) · Database · doi:10.1093/database/baz046 · Cited by 1,551 · see citing works
The Allen Mouse Brain Common Coordinate Framework: A 3D Reference Atlas
Wang, Q., Ding, S., Li, Y., Royall, J. J., Feng, D. H., Lesnar, P., Graddis, N., Naeemi, M., Facer, B. A. C., Ho, A., et al. (2020) · Cell · doi:10.1016/j.cell.2020.04.007 · Cited by 1,410 · see citing works
<p>Alzheimer’s disease: pathogenesis, diagnostics, and therapeutics</p>
Tiwari, S., Atluri, V., Kaushik, A., Yndart, A., & Nair, M. (2019) · International Journal of Nanomedicine · doi:10.2147/ijn.s200490 · Cited by 1,308 · see citing works
Hu, C., Li, T., Xu, Y., Zhang, X., Li, F., Bai, J., Chen, J., Jiang, W., Yang, K., Ou, Q., et al. (2022) · Nucleic Acids Research · doi:10.1093/nar/gkac947 · Cited by 1,284 · see citing works
Self-citationA high-resolution transcriptomic and spatial atlas of cell types in the whole mouse brain
Yao, Z., van Velthoven, C. T. J., Kunst, M., Zhang, M., McMillen, D., Lee, C., Jung, W., Goldy, J., Abdelhak, A., Aitken, M., et al. (2023) · Nature · doi:10.1038/s41586-023-06812-z · Cited by 1,194 · see citing works
Generalized ${ q}$-Sampling Imaging
Yeh, F.-C. F., Wedeen, V. J., & Tseng, W. I. (2010) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2010.2045126 · Cited by 1,030 · see citing works
Molecular and cellular mechanisms underlying the pathogenesis of Alzheimer’s disease
Guo, T., Zhang, D., Zeng, Y., Huang, T. Y., Xu, H., & Zhao, Y. (2020) · Molecular Neurodegeneration · doi:10.1186/s13024-020-00391-7 · Cited by 994 · see citing works
Self-citationGrubman, A., Chew, G., Ouyang, J. F., Sun, G., Choo, X. Y., McLean, C., Simmons, R. K., Buckberry, S., Vargas-Landin, D. B., Poppe, D., et al. (2019) · Nature Neuroscience · doi:10.1038/s41593-019-0539-4 · Cited by 956 · see citing works
Spatial Transcriptomics and In Situ Sequencing to Study Alzheimer’s Disease
Chen, W.-T., Lu, A., Craessaerts, K., Pavie, B., Sala Frigerio, C., Corthout, N., Qian, X., Laláková, J., Kühnemund, M., Voytyuk, I., et al. (2020) · Cell · doi:10.1016/j.cell.2020.06.038 · Cited by 949 · see citing works
Schweser, F., Deistung, A., Lehr, B. W., & Reichenbach, J. R. (2010) · NeuroImage · doi:10.1016/j.neuroimage.2010.10.070 · Cited by 713 · see citing works
Spatially resolved transcriptomes—next generation tools for tissue exploration
Asp, M., Bergenstråhle, J., & Lundeberg, J. (2020) · BioEssays · doi:10.1002/bies.201900221 · Cited by 666 · see citing works
Burt, J. B., Demirtaş, M., Eckner, W. J., Navejar, N. M., Ji, J. L., Martin, W. J., Bernacchia, A., Anticevic, A., & Murray, J. D. (2018) · Nature Neuroscience · doi:10.1038/s41593-018-0195-0 · Cited by 657 · see citing works
Correlated gene expression supports synchronous activity in brain networks
Richiardi, J., Altmann, A., Milazzo, A.-C., Chang, C., Chakravarty, M. M., Banaschewski, T., Barker, G. J., Bokde, A. L. W., Bromberg, U., Büchel, C., et al. (2015) · Science · doi:10.1126/science.1255905 · Cited by 625 · see citing works
Quantitative susceptibility mapping of human brain reflects spatial variation in tissue composition
Li, W., Wu, B. H., & Liu, C. (2011) · NeuroImage · doi:10.1016/j.neuroimage.2010.11.088 · Cited by 625 · see citing works
White matter characterization with diffusional kurtosis imaging
Fieremans, E., Jensen, J. H., & Helpern, J. A. (2011) · NeuroImage · doi:10.1016/j.neuroimage.2011.06.006 · Cited by 561 · see citing works
Spatial maps of prostate cancer transcriptomes reveal an unexplored landscape of heterogeneity
Berglund, E., Maaskola, J., Schultz, N., Friedrich, S., Marklund, M., Bergenstråhle, J., Tarish, F., Tanoglidi, A., Vicković, S., Larsson, L., et al. (2018) · Nature Communications · doi:10.1038/s41467-018-04724-5 · Cited by 550 · see citing works
Classification of white matter lesions on magnetic resonance imaging in elderly persons
Kim, K. W., MacFall, J. R., & Payne, M. E. (2008) · Biological Psychiatry · doi:10.1016/j.biopsych.2008.03.024 · Cited by 539 · see citing works
Generative modeling of brain maps with spatial autocorrelation
Burt, J. B., Helmer, M., Shinn, M., Anticevic, A., & Murray, J. D. (2020) · NeuroImage · doi:10.1016/j.neuroimage.2020.117038 · Cited by 528 · see citing works
Quantitative susceptibility mapping: current status and future directions
Haacke, E. M., Liu, S., Buch, S., Zheng, W., Wu, D., & Ye, Y. (2015) · Magnetic Resonance Imaging · doi:10.1016/j.mri.2014.09.004 · Cited by 511 · see citing works
Systematic phenotyping and characterization of the 5xFAD mouse model of Alzheimer’s disease.
Forner, S., Kawauchi, S., Balderrama-Gutierrez, G., Kramár, E. A., Matheos, D. P., Phan, J., Javonillo, D. I., Tran, K. M., Hingco, E., da Cunha, C., et al. (2021) · Scientific Data · doi:10.1038/s41597-021-01054-y · Cited by 501 · see citing works
Olabarria, M., Noristani, H. N., Verkhratsky, A., & Rodríguez, J. J. (2010) · Glia · doi:10.1002/glia.20967 · Cited by 499 · see citing works
Eimer, W. A., & Vassar, R. (2013) · Molecular Neurodegeneration · doi:10.1186/1750-1326-8-2 · Cited by 425 · see citing works
SANDI: a compartment‐based model for non‐invasive apparent soma and neurite imaging by diffusion MRI
Palombo, M., Ianus, A., Guerreri, M., Nunes, D., Alexander, D. C., Shemesh, N., & Zhang, H. (2020) · NeuroImage · doi:10.1016/j.neuroimage.2020.116835 · Cited by 349 · see citing works
A method for estimating and removing streaking artifacts in quantitative susceptibility mapping
Li, W., Wang, N., Yu, F. F., Han, H., Cao, W., Romero, R. S., Tantiwongkosi, B., Duong, T. Q., & Liu, C. (2014) · NeuroImage · doi:10.1016/j.neuroimage.2014.12.043 · Cited by 339 · see citing works
Self-citationOverview of quantitative susceptibility mapping
Deistung, A., Schweser, F., & Reichenbach, J. R. (2017) · NMR in Biomedicine · doi:10.1002/nbm.3569 · Cited by 338 · see citing works
Diffusion tensor imaging in mild cognitive impairment and Alzheimerʼs disease: a review
Chua, T. C., Wen, W., Slavin, M. J., & Sachdev, P. S. (2008) · Current Opinion in Neurology · doi:10.1097/wco.0b013e3282f4594b · Cited by 273 · see citing works
Role of structural MRI in Alzheimer's disease
Vemuri, P., & Jack, C. R. (2010) · Alzheimer s Research & Therapy · doi:10.1186/alzrt47 · Cited by 246 · see citing works
Desai, M. K., Mastrangelo, M. A., Ryan, D. A., Sudol, K. L., Narrow, W. C., & Bowers, W. J. (2010) · The American Journal of Pathology · doi:10.2353/ajpath.2010.100087 · Cited by 232 · see citing works
Beta‐amyloid deposition and the aging brain
Rodrigue, K. M., Kennedy, K. M., & Park, D. C. (2009) · Neuropsychology Review · doi:10.1007/s11065-009-9118-x · Cited by 230 · see citing works
A diffusion MRI tractography connectome of the mouse brain and comparison with neuronal tracer data
Calabrese, E., Badea, A., Cofer, G., Qi, Y., & Johnson, G. A. (2015) · Cerebral Cortex · doi:10.1093/cercor/bhv121 · Cited by 221 · see citing works
Determinants of costs of care for patients with Alzheimer's disease
Jönsson, L., Eriksdotter, M., Kilander, L., Soininen, H. S., Hallikainen, M., Waldemar, G., Nygaard, H. A., Andreasen, N., Winblad, B. G., & Wimo, A. B. (2006) · International Journal of Geriatric Psychiatry · doi:10.1002/gps.1489 · Cited by 220 · see citing works
The importance of correcting for signal drift in diffusion MRI
Vos, S. B., Tax, C. M. W., Luijten, P. R., Sebastien Ourselin, A., Leemans, A., & Froeling, M. (2016) · Magnetic Resonance in Medicine · doi:10.1002/mrm.26124 · Cited by 217 · see citing works
Deep learning for Alzheimer's disease diagnosis: A survey
Khojaste-Sarakhsi, M., Haghighi, S. S., Ghomi, S. M. T. F., & Marchiori, E. (2022) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2022.102332 · Cited by 194 · see citing works
The potential of functional MRI as a biomarker in early Alzheimer's disease
Sperling, R. (2011) · Neurobiology of Aging · doi:10.1016/j.neurobiolaging.2011.09.009 · Cited by 187 · see citing works
Jelescu, I. O., de Skowronski, A., Geffroy, F., Palombo, M., & Novikov, D. S. (2022) · NeuroImage · doi:10.1016/j.neuroimage.2022.119277 · Cited by 177 · see citing works
Fulcher, B., Arnatkevičiūtė, A., & Fornito, A. (2021) · Nature Communications · doi:10.1038/s41467-021-22862-1 · Cited by 174 · see citing works
Cortical iron reflects severity of Alzheimer's disease
van Duijn, S., Bulk, M., van Duinen, S. G., Nabuurs, R. J. A., van Buchem, M. A., van der Weerd, L., & Natté, R. (2017) · Journal of Alzheimer s Disease · doi:10.3233/jad-161143 · Cited by 172 · see citing works
Gong, N.-J., Dibb, R., Bulk, M., van der Weerd, L., & Liu, C. (2019) · NeuroImage · doi:10.1016/j.neuroimage.2019.02.019 · Cited by 162 · see citing works
Kulijewicz‐Nawrot, M., Verkhratsky, A., Chvátal, A., Syková, E., & Rodríguez, J. J. (2012) · Journal of Anatomy · doi:10.1111/j.1469-7580.2012.01536.x · Cited by 160 · see citing works
Praet, J., Manyakov, N. V., Muchene, L., Mai, Z., Terzopoulos, V., de Backer, S., Torremans, A., Guns, P.-J., Van De Casteele, T., Bottelbergs, A., et al. (2018) · Alzheimer's Research & Therapy · doi:10.1186/s13195-017-0329-8 · Cited by 133 · see citing works
Diffusion time dependence, power‐law scaling, and exchange in gray matter
Olesen, J. L., Østergaard, L., Shemesh, N., & Jespersen, S. N. (2022) · NeuroImage · doi:10.1016/j.neuroimage.2022.118976 · Cited by 125 · see citing works
Imaging transcriptomics of brain disorders
Arnatkeviciute, A., Fulcher, B. D., Bellgrove, M. A., & Fornito, A. (2022) · Biological Psychiatry Global Open Science · doi:10.1016/j.bpsgos.2021.10.002 · Cited by 113 · see citing works
What’s new and what’s next in diffusion MRI preprocessing
Tax, C., Bastiani, M., Veraart, J., Garyfallidis, E., & Irfanoglu, M. (2021) · NeuroImage · doi:10.1016/j.neuroimage.2021.118830 · Cited by 107 · see citing works
Falangola, M. F., Guilfoyle, D. N., Tabesh, A., Hui, E. S., Nie, X., Jensen, J. H., Gerum, S. V., Hu, C., LaFrancois, J., Collins, H. R., et al. (2014) · NMR in Biomedicine · doi:10.1002/nbm.3140 · Cited by 96 · see citing works
Biogenic metallic elements in the human brain?
Everett, J., Lermyte, F., Brooks, J., Tjendana-Tjhin, V., Plascencia-Villa, G., Hands-Portman, I., Donnelly, J. M., Billimoria, K., Perry, G., Zhu, X., et al. (2021) · Science Advances · doi:10.1126/sciadv.abf6707 · Cited by 94 · see citing works
Ota, M., Sato, N., Nakaya, M., Shigemoto, Y., Kimura, Y., Chiba, E., Yokoi, Y., Tsukamoto, T., & Matsuda, H. (2022) · Journal of Alzheimer s Disease · doi:10.3233/jad-220534 · Cited by 79 · see citing works
Patch2Self: Denoising Diffusion MRI with Self-Supervised LearningPreprint
Fadnavis, S., Batson, J., & Garyfallidis, E. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.01355 · Cited by 72 · see citing works
Neurite orientation dispersion and density imaging of mouse brain microstructure
Wang, N., Zhang, J., Cofer, G., Qi, Y., Anderson, R. J., White, L. E., & Allan Johnson, G. (2019) · Brain Structure and Function · doi:10.1007/s00429-019-01877-x · Cited by 69 · cited 1× · see citing works
“…an in vivo study to achieve higher anatomic resolution and better tissue contrast, which improves the detection of microstructural changes, 45,46 No significant signal drift was observed as described in the previous study 49 ; therefore, no additional retrospective drift correction was applied.”
Self-citationWhite matter degeneration in subjective cognitive decline: a diffusion tensor imaging study
Li, X.-Y., Tang, Z.-C., Sun, Y., Tian, J., Liu, Z.-Y., & Han, Y. (2016) · Oncotarget · doi:10.18632/oncotarget.10091 · Cited by 65 · see citing works
Self-citationHolmes, H. E., Powell, N. M., Ma, D., Ismail, O., Harrison, I. F., Wells, J. A., Colgan, N., O'Callaghan, J. M., Johnson, R. A., Murray, T. K., et al. (2017) · Frontiers in Neuroinformatics · doi:10.3389/fninf.2017.00020 · Cited by 55 · see citing works
Kaur, A., Mittal, M., Bhatti, J. S., Thareja, S., & Singh, S. (2024) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2024.102928 · Cited by 54 · see citing works
The DELCODE Study Group, Brueggen, K., Dyrba, M., Cardenas-Blanco, A., Schneider, A., Fliessbach, K., Buerger, K., Janowitz, D., Peters, O., Menne, F., et al. (2019) · Journal of Neurology · doi:10.1007/s00415-019-09429-3 · Cited by 51 · see citing works
Zhou, C., Sun, X., Hu, Y., Song, J., Dong, S., Kong, D., Wang, Y., Hua, X., Han, J., Zhou, Y., et al. (2019) · Aging · doi:10.18632/aging.102260 · Cited by 50 · see citing works
Adewale, Q., Khan, A. F., Carbonell, F., Iturria-Medina, Y., & Initiative, A. D. N. (2021) · eLife · doi:10.7554/elife.62589 · Cited by 44 · see citing works
Qin, Y.-Y., Li, M.-W., Zhang, S., Zhang, Y., Zhao, L.-Y., Lei, H., Oishi, K., & Zhu, W.-Z. (2013) · Neuroradiology · doi:10.1007/s00234-013-1195-0 · Cited by 42 · see citing works
Probing demyelination and remyelination of the cuprizone mouse model using multimodality MRI
Wang, N., Zhuang, J., Wei, H., Dibb, R., Qi, Y. P., & Liu, C. (2019) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.26758 · Cited by 41 · see citing works
Self-citationColon-Perez, L. M., Ibanez, K. R., Suarez, M., Torroella, K., Acuna, K., Ofori, E., Levites, Y., Vaillancourt, D. E., Golde, T. E., Chakrabarty, P., et al. (2019) · NeuroImage · doi:10.1016/j.neuroimage.2019.116138 · Cited by 41 · see citing works
Ianuş, A., Carvalho, J., Fernandes, F. F., Cruz, R., Chavarrias, C., Palombo, M., & Shemesh, N. (2022) · NeuroImage · doi:10.1016/j.neuroimage.2022.119135 · Cited by 40 · see citing works
Lampinen, B., Szczepankiewicz, F., Lätt, J., Knutsson, L., Mårtensson, J., Björkman-Burtscher, I. M., van Westen, D., Sundgren, P. C., Ståhlberg, F., & Nilsson, M. (2023) · NeuroImage · doi:10.1016/j.neuroimage.2023.120338 · Cited by 38 · see citing works
Habuza, T., Zaki, N., Mohamed, E. A., & Statsenko, Y. A. (2022) · IEEE Access · doi:10.1109/access.2022.3174601 · Cited by 36 · see citing works
In vivo diffusion tensor imaging of amyloid‐β‐induced white matter damage in mice
Sun, S.-W., Liang, H.-F., Mei, J., Xu, D., & Shi, W.-X. (2013) · Journal of Alzheimer s Disease · doi:10.3233/jad-130236 · Cited by 27 · see citing works
Resolution and b value dependent structural connectome in ex vivo mouse brain
Crater, S., Maharjan, S., Qi, Y., Zhao, Q., Cofer, G., Cook, J. J., Johnson, G., & Wang, N. (2022) · bioRxiv · doi:10.1016/j.neuroimage.2022.119199 · Cited by 27 · see citing works
Self-citationFalangola, M. F., Nie, X., Ward, R., Dhiman, S., Voltin, J., Nietert, P. J., & Jensen, J. H. (2021) · Magnetic Resonance Imaging · doi:10.1016/j.mri.2021.06.022 · Cited by 25 · see citing works
Accelerating quantitative susceptibility imaging acquisition using compressed sensing
Wang, N., Cofer, G. P., Anderson, R. J., Qi, Y. P., Liu, C., & Johnson, G. A. (2018) · Physics in Medicine and Biology · doi:10.1088/1361-6560/aaf15d · Cited by 24 · see citing works
Self-citationOlfactory Evaluation in Alzheimer’s Disease Model Mice
Zhang, J., Zhao, Z., Sun, S., Li, J., Wang, Y., Dong, J., Yang, S. C., Lou, Y., Yang, J., Li, W., et al. (2022) · Brain Sciences · doi:10.3390/brainsci12050607 · Cited by 20 · see citing works
Pathway and annotation database for iDEP
Ge, S. (2018) · Zenodo (CERN European Organization for Nuclear Research) · doi:10.5281/zenodo.1451846 · Cited by 20 · see citing works
Characterization and correlation of signal drift in diffusion weighted MRI
Hansen, C. B., Nath, V., Hainline, A. E., Schilling, K. G., Parvathaneni, P., Bayrak, R. G., Blaber, J. A., Irfanoglu, O., Pierpaoli, C. M., Anderson, A. W., et al. (2018) · Magnetic Resonance Imaging · doi:10.1016/j.mri.2018.11.009 · Cited by 16 · see citing works
In vivo cortical diffusion imaging relates to Alzheimer's disease neuropathology
Torso, M., Ridgway, G. R., Valotti, M., Hardingham, I., Chance, S. A., for the National Alzheimer’s Coordinating Center, Brewer, J., Lopez, O., Hyman, B., Grabowski, T., et al. (2023) · Alzheimer's Research & Therapy · doi:10.1186/s13195-023-01309-3 · Cited by 16 · see citing works
Han, X., Maharjan, S., Chen, J., Zhao, Y., Qi, Y., White, L. E., Johnson, G. A., & Wang, N. (2024) · NeuroImage · doi:10.1016/j.neuroimage.2024.120734 · Cited by 15 · see citing works
Self-citationAlzheimer disease diagnosis for magnetic resonance brain images using deep learning neural networks
Ahmed, H. M., Elsharkawy, Z. F., & Elkorany, A. S. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-022-14203-1 · Cited by 14 · see citing works
Yoo, C.-H., Kim, J., Baek, H.-M., Chang, K.-A., & Choe, B.-Y. (2023) · International Journal of Molecular Sciences · doi:10.3390/ijms24065073 · Cited by 13 · see citing works
Whittaker, H. T., Zhu, S., Di Curzio, D. L., Buist, R., Li, X.-M., Noy, S., Wiseman, F. K., Thiessen, J. D., & Martin, M. (2018) · Magnetic Resonance Imaging · doi:10.1016/j.mri.2018.03.010 · Cited by 12 · see citing works
Müller, H.-P., Roselli, F., Rasche, V., & Kassubek, J. (2020) · Frontiers in Neuroscience · doi:10.3389/fnins.2020.00734 · Cited by 12 · see citing works
Age-dependent microstructure alterations in 5xFAD mice by high-resolution diffusion tensor imaging
Maharjan, S., Tsai, A. P., Lin, P. B., Ingraham, C., Jewett, M. R., Landreth, G. E., Oblak, A. L., & Wang, N. (2022) · Frontiers in Neuroscience · doi:10.3389/fnins.2022.964654 · Cited by 11 · see citing works
Self-citationShen, Y., Shen, Y., Wang, M., Jin, K., Yang, P., Cao, Z., Zhu, Q., Zhao, Z., Li, H., Han, L., et al. (2025) · Cell Reports · doi:10.1016/j.celrep.2025.116073 · Cited by 10 · see citing works
A cross‐comparative analysis of in vivo versus ex vivo MRI indices in a mouse model of concussion
Vinh To, X., Kurniawan, N. D., Cumming, P., & Nasrallah, F. A. (2023) · Brain Research · doi:10.1016/j.brainres.2023.148562 · Cited by 5 · see citing works
Chen, J., Han, X., Liu, Z., Zhou, C., Hu, R., Tabassam, S., Wyatt-Johnson, S. K., Oblak, A. L., Brutkiewicz, R. R., Lin, M., et al. (2026) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2025.3614233 · Cited by 2 · see citing works
Self-citationHigh-resolution mri guided whole mouse brain neuronal cell type atlas using deep learning
Han, X., Hu, R., Liu, Z., Chen, J., Jafry, M., Song, H., Zhao, Y., Lin, M., White, L. E., Johnson, G. A., et al. (2026) · Communications Biology · doi:10.1038/s42003-026-10608-y
Self-citationPathway and annotation database for iDEP
Ge, S. (2018) · Zenodo · doi:10.5281/zenodo.1451847
Flagged first, then by citations
- Authors
- Han, X., Hu, R., Ping, X., Chen, J., Liu, J., Liu, Z., Jin, X., Gao, X., Lin, M., Zhao, Y., & Wang, N.
- Affiliations
- Advanced Imaging Research (United States) (US)
- Indiana University – Purdue University Indianapolis (US)
- University of Minnesota (US)
- Indiana University Health (US)
- Southwestern Medical Center (US)
- Corresponding author
- Wang, Nian
- Venue
- Alzheimer's & Dementia
- Publisher
- Wiley
- Published
- 2026-10
- Volume
- 22
- Issue
- 10
- Pages
- e71908-e71908
- ISSN
- 1552-5260 (ISSN-L)
- 1552-5279
- PMC
- PMC13650524
- OpenAlex
- W7221151605
- Semantic Scholar
- 6aca16f3fc7eb181b3eadf7a52ac5da2a6f85b06
- Cited by
- 0
- References
- 85
- FWCI
- 0.00
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Alzheimer s & Dementia (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- Funders
- Cancer Prevention and Research Institute of Texas (RP210099)
- Ralph W. and Grace M. Showalter Research Trust Fund
- Foundation for the National Institutes of Health (R01 NS125020)
- 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
- MeSH terms
- Spatial Transcriptomics, Alzheimer Disease, Animals, Brain, Disease Models, Animal, Mice, Transgenic, Neurons, Oligodendroglia, Neural Networks, Computer, Microglia, Diffusion Magnetic Resonance Imaging, Mice, Plaque, Amyloid, Transcriptome
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
