Jack, Clifford R.

AbstractBackgroundFluorodeoxyglucose positron emission tomography (FDG‐PET) is an established modality for differential diagnosis of dementia. Deriving phenotypic signatures from FDG‐PET via machine learning is challenging due to the high dimensional nature of brain images relative to the generally small number of labeled examples available for training, the class imbalance among those labels, and the cooccurrence of multiple pathologies. In this study, we developed a multi‐class, multi‐label framework to address these challenges.MethodA database of clinically acquired PET/CT images from 3,000 unique patients was used to develop a latent space model using matrix decomposition. This model was then applied to images from a separate cohort of Mayo Clinic Alzheimer’s Disease Research Center participants (n=1,745) labeled as cognitively unimpaired (CU) (n=1,436) or with the following potentially co‐occurring phenotypes: Alzheimer’s disease (AD) (n=165), Lewy body dementia (DLB) (n=92), behavioral variant frontotemporal dementia (bvFTD) (n=43), semantic (svPPA) (n=10) and logopenic (lvPPA) (n=13) variant PPA, and posterior cortical atrophy (PCA) (n=17). A k‐nearest neighbors classifier that is robust to these imbalanced and overlapping labels was then trained on these examples. The resulting classifier was evaluated by area under receiver‐operator characteristic curve (ROC‐AUC) via leave one out cross validation, using clinical diagnosis as the gold standard.ResultROC curves and AUC scores for each phenotype are illustrated in Fig. 1a. Because the classifier is based on a k‐nearest neighbors connectivity matrix, it has a convenient graphical representation, where images are nodes and edges are drawn between an image and its set of nearest neighbors in latent space. A self‐organizing force directed graph constructed in this way is illustrated in Fig 1b, highlighting the strong separation of CU and degenerative images, as well as the segregation of each phenotype within the neurodegenerative region of the graph.ConclusionIn this study, we developed a machine learning framework for classification of neurodegenerative disease based on k‐nearest neighbor analysis in a low dimensional latent space projection of FDG‐PET images. By leveraging low‐dimensional representations and k‐nearest neighbors analysis, this framework is robust in multi‐class, multi‐label tasks with strong class imbalance and provides a highly interpretable graphical representation.

AbstractBackgroundNeurite Orientation Dispersion and Density Imaging (NODDI) is becoming increasingly common as multiband acceleration becomes available for clinical scanning, but fitting the model can be extremely slow. Since the release of the original MATLAB implementation in 2012, three alternatives have become popular: Accelerated Microstructure Imaging via Convex Optimization (AMICO), the Microstructure Diffusion Toolbox (MDT), and the CUDA Diffusion Modelling Toolbox (cuDIMOT). They accelerate fitting through reformulation and/or using GPUs, but the implementation differences raise the question of which approach is best. To our knowledge all previous comparisons lacked knowledge of the true values and used the original implementation as the nominally correct one, with in vivo data. We remedy that using simulations with known NODDI values, in addition to in vivo scans.MethodWe created a software phantom with Neurite Density Index (NDI) values linearly increasing from 0.1 to 1.0 in x, Orientation Dispersion (ODI) from 0 to 1 in y, and Isotropic water Volume Fraction (ISOVF) from 0 to 0.9 in z. We simulated five scans using the ADNI3 advanced diffusion MRI protocol. The phantom had 20 voxels in each direction, and the fiber bundle principal direction randomly varied between the simulations. The simulations were fit by the original, AMICO, MDT, and cuDIMOT implementations, and root mean square errors were calculated as shown in Table 1. A discrepancy measure was also calculated using the brain voxels of three in vivo human scans acquired using the same protocol on a Siemens Prisma, substituting the original implementation’s values for the truth since the latter is unknown for in vivo scans.ResultcuDIMOT had the least error (for ODI < 0.8) and was fastest. The original implementation was a close second in accuracy but much slower. cuDIMOT’s ODI underestimation can be corrected with a transfer function in the noise‐free case, but that approach fails with actual scans.ConclusionThe original or cuDIMOT implementations should be used, with cuDIMOT the favorite when at least one CUDA‐capable GPU is available and the number of scans are large. Correcting cuDIMOT’s ODI underestimation using a different initialization or form of Dawson’s integral should be investigated.

AbstractBackgroundIn 2018 we presented the Computational DESH (CDESH) metric (DOI:10.1016/j.nicl.2018.11.015), a machine learning measure of abnormal CSF space patterns commonly observed on MRI in Disproportionately Enlarged Subarachnoid‐space Hydrocephalus (DESH‐type) NPH. We have shown that CDESH positive individuals often have discordant CSF and PET amyloid measures (DOI:10.1016/j.jalz.2019.06.4921). Positive CDESH scores are predictive of cognitive decline (DOI:10.1016/j.neuroimage.2021.117899, DOI: 10.1212/WNL.0000000000008616). The score is calculated from regional CSF volumes in a sulcal atlas and within the ventricles using a support vector machine (SVM). The Mayo Clinic Study of Aging (MCSA) changed scanner platforms (GE to Siemens) and T1‐weighted image resolution (1.2mm x 1.05mm x 1.05mm to 0.8mm isotropic) in 2017. We refer to these as “isotropic” or “anisotropic” scans. Our objective was to assess the dependence of a machine learning method (specifically CDESH) to image input variation and, if necessary, determine what needed corrections.MethodAll data considered is from the MCSA or Mayo Clinic ADRC. The SVM was trained on anisotropic images with laboriously labelled ground truth. We resampled the isotropic data from a crossover study (N=112, typically scanned same‐day with both platforms/resolutions) to match the anisotropic resolution. Two approaches to correction were evaluated (Figure 1). In Figure 2 we compare of the score with and without correction cross‐sectionally in 998 MCSA participants. In Figure 3 we present longitudinal series from 23 participants with multiple scans using the isotropic resolution protocol and at least one prior anisotropic image.ResultsComparisons of data from the crossover study are presented in Figure 1. Matching voxel resolution to the data used to train the SVM is critical and a subsequent mild linear score adjustment may be warranted. Cross‐sectional comparisons (Figure 2) confirm improvement matching voxel resolution with a possible mild linear score adjustment. After matching voxel resolution CDESH score trajectories (Figure 3) are markedly more stable across this scanner/protocol change.ConclusionThe SVM‐based CDESH score is sensitive to voxel resolution. Corrections are needed in image processing and possibly in the scalar output score to achieve stability across protocol and scanner changes. The resulting correction supports coherent long timeseries data and longitudinal extension of previous work.

AbstractBackgroundDiffusion MRI (dMRI) based quantification of microstructural changes in white matter (WM) are increasingly proposed to measure cerebrovascular disease (CVD) related changes. Our objective was to compare the proposed dMRI markers using postmortem neuropathologic CVD data and then evaluate the predictors of longitudinal dMRI markers using serial scans.MethodWe identified n=51 participants (mean age: 83.8 years, 63% males) with ante‐mortem dMRI and postmortem CVD evaluation and n=718 participants (mean age: 71.1 years, 56% males) with at least two dMRI scans from the population‐based sample of Mayo Clinic Study of Aging. We computed dMRI measures proposed for measuring CVD: Free Water (FW), Fractional Anisotropy (FA) adjusted for FW (FAadj), Peak width Skeletonized Mean Diffusivity (PSMD), and FA of the genu of corpus callosum (Genu‐FA). We used the FW and PSMD (PSMD release 1.8.1) kits from the MarkVCID consortium.Using weighted linear regression models with adjustments for MRI scan time to death, we evaluated associations between the baseline dMRI and two pathology CVD scores: Strozyk (represents the presence and number of macroscopic lesions) and Kalaria (represents the summary score of vessel wall modifications) scales. We ran linear mixed effect models with all available serial dMRI measures as outcomes and vascular risk (measured by the number of cardiovascular and metabolic conditions‐CMC), baseline imaging measures (amyloid and white matter hyperintensities (WMH)) and their interaction with time as key predictors.ResultMost dMRI markers were predictors of postmortem CVD pathology (Table 1). In longitudinal dMRI models, terms associated with amyloid explained little variability in dMRI (<1%) (Table 2). WMH was a significant predictor of baseline and decline in dMRI measures with WMH explaining a considerable percentage of variability in FW (34%) and PSMD (28%). Vascular risk measures were significant predictors of three out of the four dMRI measures.ConclusionThe diffusion measures proposed as surrogate markers of CVD map reasonably well to CVD pathological scales and do not change substantially as a function of baseline amyloidosis. Our findings shed light on variability in the proposed diffusion markers using postmortem data and longitudinal imaging data. Further work is needed to evaluate their clinical utility.

AbstractBackgroundCerebral Microbleeds (CMBs) are tracked in anti‐amyloid drug safety monitoring. T2*GRE or SWI MRI images are visually assessed for the presence of CMBs. Automated methods tuned for sensitivity often have unacceptably high false positive rates requiring extensive human cleanup.MethodAssume contemporaneous T2*GRE and T1‐weighted images. CMB detection (Figures 1,2) has three steps: 1) Segmentation outputs from the T1‐weighted image are resampled into the T2*GRE image voxel raster; 2) Normalized cross‐correlation of the T2*GRE with a CMB template identifies “candidate CMBs”; 3) a neural network using voxel patches centered on each candidate CMB rejects false positives. Patches presented to the network are 4D and include five contrasts drawn from the T2*GRE, correlation, T1‐weighted, WM probability and CSF probability images. The simple network architecture consists of convolutional blocks followed by dense blocks. The system is trained using images from ADNI, ARIC, MCSA, and the Mayo Clinic ADRC that have been previously visually graded for CMBs (Table in Figure 3). Importantly, we include participants with and without CMBs in our data reflecting the fact that the majority of people (even the elderly) have no CMBs. Data is split 70:10:20 by subject for training, validation, and testing. Various network configurations were evaluated.ResultHigh candidate detection sensitivity can be achieved at the expense of presenting hundreds to thousands of false positive candidates per input image to the rejection network. Our best performing network configuration to date has AUC 0.982 (95% bootstrap CI 0.971 to 0.990 based on the 10% validation subset). Choosing an operating point with 95% sensitivity in the validation set and admits, on average, 0.6 false positive CMB candidates per input T2*GRE image. Sensitivity is hindered by a small number of obvious errors in the ground truth visual assessment. Validation loss tracks training loss with no indication of over‐fitting. Example patches are shown in Figure 4.ConclusionExcessive false positive detection requires significant human correction and effort, preventing automated CMB detection from being widely adopted. We have an automated detection system that is sensitive (95%) and provides excellent false positive rejection (<1 false positive CMB/input image) reducing needed human effort.

AbstractBackgroundAlzheimer’s disease (AD) pathologic changes (e.g. amyloid and tau aggregation), and cerebrovascular disease often coexist, and it remains unclear whether these are mechanistically linked processes (Graff‐Radford, Brain 2019; Provenzano, JAMA Neurol 2013). Most studies evaluating these associations are cross‐sectional; however, establishing temporal relationships may provide additional insight into potential mechanistic associations. One method to assess the temporal evolution of a biomarker is the Accelerated Failure Time (AFT) model, which assumes a common pattern of progression for a given biomarker that is shifted earlier or later in time for each individual (Therneau, Neuroimage 2021). In this work we investigate the relationship of disease trajectories of cerebrovascular disease (via white matter hyperintensity, WMH, volume) and AD biomarkers (via amyloid and tau PET).MethodsWe included Mayo Clinic Study of Aging (MCSA) and Mayo Alzheimer’s Disease Research Center (ADRC) participants with amyloid PET, tau PET, and MRI, age > 50 years, and a clinical diagnosis of cognitively unimpaired (CU), mild cognitive impairment (MCI) or Alzheimer’s disease dementia. The AFT model was fit with amyloid PET (PiB) global meta‐ROI SUVR, tau PET (flortaucipir) temporal meta‐ROI SUVR, and WMH volume as the tri‐variate endpoints, with APOE genotype, sex, education, total intracranial volume, and referral/population subject as covariates. A per subject/endpoint random effect represents individual adjustments for each biomarker to indicate if the subject's biomarker became abnormal earlier or later relative to their demographic peers, accounting for the covariates.ResultsThe study included 1144 participants (Table 1). There was a strong correlation between earlier onset of amyloid and earlier onset of tau (r=0.64, CI=0.60‐0.67) (Figure 1, Table 2); early/late amyloid timing explained nearly half (0.642 = 41%) of the variation in tau onset. There were only weak associations between an individual’s WMH and amyloid PET onset adjustments (r=0.10, 1.0%) and WMH and tau PET onset adjustments (r=‐0.07, 0.5%).ConclusionThis analysis supports a strong mechanistic relationship between amyloid and tau aggregation, as reported by others. However, our findings do not support a mechanistic association between WMH and amyloid or tau PET.

AbstractBackgroundMulti‐compartment modelling of white matter (WM) microstructure using Neurite Orientation Dispersion and Density Imaging (NODDI) can provide information on WM health through the neurite (axons and dendrites) density index and free water measures. We hypothesized that cerebrovascular disease (CVD), Alzheimer’s disease (AD), and an FDG pattern suggestive of TDP‐43 proteinopathy would be associated with distinct NODDI readouts of WM damage which would be informative for identifying the substrate for cognitive impairment.MethodWe identified two independent cohorts with multi‐shell diffusion MRI, amyloid and tau PET, and cognitive assessments. One, were participants from the Mayo Clinic Study of Aging, a population‐based cohort of 347 elderly randomly sampled from the Olmsted county, Minnesota. The second were 61 amyloid positive AD participants from the Mayo AD Research Center.ResultWe observed an increase in free water and decrease in neurite density in the genu of the corpus callosum associated with vascular risk factors, which we refer to as vascular WM component (Figure 1). Tau PET signal reflective of 3R/4R tau deposition was associated with worsening neurite density index in the temporal WM where we measured parahippocampal cingulum and inferior temporal WM bundles. Worsening temporal neurite density was associated with (antemortem confirmed) FDG TDP‐43 signature. Post‐mortem neuropathologic data on a small subset (n=9) of this sample lend support to our findings. In the population‐based cohort where vascular disease was more prevalent, the vascular WM component explained variability in global cognition (partial R2 of free water and neurite density = 8.3%) and MMSE performance (8.2%) which was comparable to amyloid PET (7.4% for global cognition and 6.6% for memory) (Table 1). In the AD dementia clinic‐based cohort, tau deposition was the greatest contributor to cognitive performance (9.6%), but there was also a non‐trivial contribution of the temporal WM component (8.5%) to cognitive performance. The differences observed between the two cohorts were reflective of their distinct clinical composition.ConclusionWhite matter microstructural damage assessed using advanced diffusion models may add significant value for distinguishing the underlying substrate (whether CVD versus neurodegenerative disease caused by tau deposition or TDP‐43 pathology) for cognitive impairment in older individuals.

AbstractBackgroundIn vivo tau‐positron emission tomography (PET) is an attractive biomarker for Alzheimer’s disease (AD) diagnosis and treatment. However, tau‐PET is less widely available than other modalities. In this study, we tested cross‐modality synthesis of tau‐PET brain images from fluorodeoxyglucose F‐18 (FDG)‐PET using a deep convolutional neural network (CNN).MethodParticipants (n=1,192) who had brain FDG‐PET with 18F‐FDG and tau‐PET with Flortaucipir (F‐18‐AV‐1451) were included for training and testing. This cohort spanned normal aging (ages 26‐98), pre‐clinical, and clinical AD and related disorders including the FTD and DLB spectrum. External validation was done using ADNI (n=288). The PET scans were co‐registered to the corresponding MRI and subsequently warped to Mayo Clinic Adult Lifespan Template (MCALT) space. Tau‐PET images were SUVR‐normalized to the cerebellar crus, and FDG to the pons. A 3D dense‐U‐net model was utilized as an architecture. Cross‐validation experiments were conducted using 5‐fold validations (60% training set, 20% validation set, and 20% test set) with mean squared error as the loss function.ResultOur dense‐U‐net model successfully synthesized tau‐PET from metabolic images with good correlation and low prediction error for regional SUVRs (Figure 1A‐C). The model showed a robust prediction ability, performing accurately in an independent, external ADNI cohort (Figure 1D‐F). The model‐imputed tau‐PET significantly improved performance in classifying tau positivity (mean AUROC(±SD)=0.78±0.04 and 0.85±0.03 for FDG‐PET and synthesized tau‐PET, respectively) and diagnostic groups (cognitively unimpaired with abnormal amyloid‐PET vs. cognitively impaired with abnormal amyloid‐PET) compared to the original input FDG data (mean AUROC(±SD)=0.89±0.04, 0.85±0.05 0.91±0.04 for actual tau‐PET, FDG‐PET and synthesized tau‐PET, respectively) (Figure 2), suggesting enhanced clinical utility for metabolic images. The ADNI cohort also showed similar results (for tau positivity: AUROC=0.66 and 0.78 for FDG‐PET and synthesized tau‐PET, respectively; for CU A+ vs. CI A+: AUROC=0.86, 0.62, and 0.73 for actual tau‐PET, FDG‐PET, and synthesized tau‐PET; Figure 3).ConclusionWe showed that using a CNN model to predict tau‐PET from FDG‐PET is feasible. The synthesized tau‐PET can augment the value of FDG‐PET, facilitating the multi‐modal diagnosis of AD.

AbstractBackgroundDisruption of the default mode network (DMN) is thought to be a common feature of aging and across Alzheimer’s disease (AD) clinical phenotypes, while syndromic variability could be reflected in selective degeneration of macro‐scale networks optimally supporting specific mental abilities (Fig 1). As part of a collaborative effort between Mayo Clinic and the Human Connectome Project‐Aging (HCP‐A), we aimed to delineate a biomarker of global network failure across aging and the dysexecutive (dAD) and amnestic phenotypes of AD, and biomarkers of focal degeneration specific to each AD phenotype.MethodWe used a regression framework to assess the relationships between a marker of DMN dysfunction, the network failure quotient (NFQ), and factors relevant to AD such as age, cortical thickness (global cortical thickness, inferior parietal lobe), hippocampal volume, and cognition (episodic memory, cognitive flexibility) in the HCP‐A cohort, which consists of 725 non‐demented older adults. We then fit centile curves to the HCP‐A cohort to assess whether the NFQ and focal degeneration markers could distinguish dAD (n=10) and amnestic AD (n=8) patients from the HCP‐A cohort at the group‐ and individual‐level. Participants were scanned using the HCP‐A protocol, allowing for the acquisition of high‐resolution multi‐echo structural images and longer resting‐state connectivity acquisition time compared to traditional protocols.ResultThe NFQ related to age, cortical thickness, hippocampal volume, and cognition in the HCP‐A cohort. It also strongly discriminated dAD and amnestic AD from the HCP‐A cohort but did not discriminate between AD phenotypes. In contrast, markers of focal degeneration were phenotype‐specific; parieto‐frontal areas yielded largest effect sizes for dAD patients, and hippocampal and temporal areas best discriminated patients with amnestic AD (Fig 2).ConclusionThe NFQ may reflect shared system‐level pathophysiological mechanisms across aging and the dysexecutive and amnestic phenotypes of AD whereas focal degeneration markers highlight distinct pathognomonic processes across AD phenotypes. These findings suggest that inter‐individual variability in cognitive impairment in AD would be closely tied to modular, phenotype‐specific network degeneration combined with a common global functional disruption of the DMN. Such information has crucial implications for clinical biomarkers and to provide a deeper understanding of heterogeneity in AD.

AbstractBackgroundDysexecutive Alzheimer's disease (dAD) is a relatively early‐onset variant of AD primarily degrading core executive functions in the absence of predominant behavioral symptoms. Clinical observations suggest substantial variability in clinical, cognitive and neuroimaging profiles within this syndrome. In this presentation, we will discuss our ongoing work disentangling this heterogeneity using unsupervised machine learning techniques.MethodWe collected clinical, neuropsychological and multimodal imaging (MRI, FDG‐PET, amyloid‐PET, tau‐PET) data on 52 dAD patients assessed in our behavioral neurology clinic at Mayo Clinic Rochester. We first performed a spectral clustering analysis based on FDG‐PET to delineate latent patterns of network degeneration (referred as “eigenbrains”) and assessed the relationships between these eigenbrains and clinical and cognitive symptomatology. We then performed a hierarchical clustering on these eigenbrains to derive data‐driven subtypes of dAD. We compared clinical and cognitive data between subtypes, and then compared the imaging profiles to 52 amyloid‐negative age‐ and sex‐matched controls.ResultSix eigenbrains explained approximatively 48% of the variance in FDG‐PET patterns and primarily reflected heteromodal to primary motor/sensory, left‐right hemispheric asymmetry, and anterior‐to‐posterior gradients of macro‐scale cortical organization (Fig 1). These eigenbrains differentially related to reported age at symptom onset, degree of clinical impairment, and performance on a wide range of cognitive domains (executive functions, episodic memory, and visuospatial). Hierarchical clustering revealed four dAD subtypes (diffuse, biparietal, left‐hemisphere, right‐hemisphere). These subtypes differed in reported age at symptom onset and cognitive profile, where the heteromodal‐diffuse subtype exhibited an overall worse clinical picture and the biparietal had a milder profile compared to other subtypes, which was not explained by disease duration. Additionally, spatial patterns of tau distribution and neurodegeneration overlapped with patterns of FDG‐PET hypometabolism in each subtype, whereas patterns of amyloid deposition were similar across subtypes (Figs 2 & 3).ConclusionAlmost half of the variance in macro‐scale patterns of hypometabolism in this dAD cohort was accounted for by six eigenbrains. These eigenbrains captured the inter‐individual variability in age at symptom onset and cognitive impairment. Four dAD subtypes derived from these eigenbrains revealed meaningful differences in clinical, cognitive, and imaging profiles. Recognizing this heterogeneity has significant clinical implications for diagnosis, prognosis, and symptom monitoring.