Jack, Clifford R.

AbstractBackgroundRates of cognitive and biomarker change in Autosomal Dominant Alzheimer disease (ADAD) vary substantially across individuals. Prior cross‐sectional work suggests that the location of the pathogenic variant within PSEN1, specifically whether the underlying variant affects transmembrane (TM) or cytoplasmic (CY) domains in PSEN1, may be a key determinant in these differential rates of progression. Here we use longitudinal data from the Dominantly Inherited Alzheimer Network observational study (DIAN‐Obs) to examine whether variants affecting TM versus CY domains in PSEN1 have differential rates of change in key cognitive and neurodegenerative markers, and whether these differences are relevant to ADAD clinical trials.MethodsUsing longitudinal clinical, cognitive, and MRI data from PSEN1 pathogenic variant carriers [TM group N=76 and CY group N=44; Table 1], we assessed rates of change in Mini‐Mental State Exam (MMSE), Clinical Dementia Rating® Sum of Boxes (CDR®‐SOB), and hippocampal volume (HV) using linear mixed effects models accounting for disease stage (estimated years to symptom onset [EYO]). We further assessed how PSEN1 mutation location (TM versus CY) impacts sample size and detectable effect size in a potential ADAD clinical trial (modeled as a 4‐year trial with annual assessments; 80% power; α = 0.05).ResultsPSEN1 TM and PSEN1 CY groups did not differ on baseline age, EYO, or CDR®. The PSEN1 TM group had significantly greater rates of change on MMSE (B[SE] = ‐0.42[0.1], p=0.002), CDR®‐SOB (B[SE] = 0.23[0.1], p=0.001), and HV atrophy (B[SE] = ‐58.93[14.3], p=0.0006 compared to the PSEN1 CY group (Fig.1). Consistent with these differential rates of change, power analyses indicated the required sample size to detect a 30% treatment effect on MMSE or HV would be reduced by 59.6% for MMSE and 91.0% for HV for a trial population comprised of PSEN1 TM versus CY carriers (Fig.2).ConclusionsIndividuals who had a variant affecting the transmembrane domains of PSEN1 had greater rates of cognitive decline and neurodegeneration compared to those with variants affecting cytoplasmic domains. In addition to having implications for ADAD pathophysiology, these results suggest that incorporating information regarding the location of PSEN1 variants may be beneficial in analyzing and designing stratification approaches for ADAD trials.

AbstractBackgroundRates of cognitive and biomarker change in Autosomal Dominant Alzheimer disease (ADAD) vary substantially across individuals. Prior cross‐sectional work suggests that the location of the pathogenic variant within PSEN1, specifically whether the underlying variant affects transmembrane (TM) or cytoplasmic (CY) domains in PSEN1, may be a key determinant in these differential rates of progression. Here we use longitudinal data from the Dominantly Inherited Alzheimer Network observational study (DIAN‐Obs) to examine whether variants affecting TM versus CY domains in PSEN1 have differential rates of change in key cognitive and neurodegenerative markers, and whether these differences are relevant to ADAD clinical trials.MethodsUsing longitudinal clinical, cognitive, and MRI data from PSEN1 pathogenic variant carriers [TM group N=76 and CY group N=44; Table 1], we assessed rates of change in Mini‐Mental State Exam (MMSE), Clinical Dementia Rating® Sum of Boxes (CDR®‐SOB), and hippocampal volume (HV) using linear mixed effects models accounting for disease stage (estimated years to symptom onset [EYO]). We further assessed how PSEN1 mutation location (TM versus CY) impacts sample size and detectable effect size in a potential ADAD clinical trial (modeled as a 4‐year trial with annual assessments; 80% power; α = 0.05).Results PSEN1 TM and PSEN1 CY groups did not differ on baseline age, EYO, or CDR®. The PSEN1 TM group had significantly greater rates of change on MMSE (B[SE] = ‐0.42[0.1], p=0.002), CDR®‐SOB (B[SE] = 0.23[0.1], p=0.001), and HV atrophy (B[SE] = ‐58.93[14.3], p=0.0006 compared to the PSEN1 CY group (Fig.1). Consistent with these differential rates of change, power analyses indicated the required sample size to detect a 30% treatment effect on MMSE or HV would be reduced by 59.6% for MMSE and 91.0% for HV for a trial population comprised of PSEN1 TM versus CY carriers (Fig.2).ConclusionsIndividuals who had a variant affecting the transmembrane domains of PSEN1 had greater rates of cognitive decline and neurodegeneration compared to those with variants affecting cytoplasmic domains. In addition to having implications for ADAD pathophysiology, these results suggest that incorporating information regarding the location of PSEN1 variants may be beneficial in analyzing and designing stratification approaches for ADAD trials.

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.

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.

AbstractBackgroundThe U.S. Study to Protect Brain Health Through Lifestyle Intervention to Reduce Risk (U.S. POINTER) is a 2‐year randomized controlled trial to evaluate the effect of lifestyle interventions in older adults (60‐79 years). The POINTER Imaging ancillary study is collecting Arterial Spin Labeling (ASL) MRI data to map cerebral blood flow (CBF) and evaluate the effect of interventions in the perfusion imaging biomarker. The optimization of timing parameters is essential for robust and efficient acquisition of ASL MRI.MethodA subset (n=83; F/M=51/32; age=69.0±5.5) of POINTER Imaging participants at two imaging sites underwent two ASL MRI scans during the same baseline session. Two 3D Pulsed ASL were obtained with TI1 of 790ms and TI2 values of 2000ms or 2750ms, which provide the delay times (between labeling and imaging) of 1210ms or 1960ms. ASL images were quantified into a physiological unit of CBF, ml/100g/min, using a single compartment kinetic model. CBF maps from two scans were normalized into a standard space and a voxel‐wise paired t‐test was performed after site variations were harmonized and covariates (age and sex) were adjusted. The age and sex effects in the difference of CBF values between two scans were also examined using linear regression. To control false positives, clusters smaller than 41ml were removed based on a simulation with α=0.05.ResultThe CBF values in ASL MRI rely on the timing of image acquisition: only a fraction of label bolus arrives in the tissue with a short delay while images suffer from low signal due to the magnetization decay with a long delay. Figure 1 shows the regions that require a long delay time (red) or a short delay (blue). The difference in CBF is proportional to arterial transit time (ATT) and the analysis shows females have shorter ATT mainly in the peri‐Sylvian region (Figure 2). No cluster survived correction in the analysis of the aging effect.ConclusionThe acquisition of two CBF maps not only offers optimization of acquisition timing but provides an additional perfusion parameter, arterial transit time, which may be a potential biomarker of cerebrovascular pathophysiology.

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.

AbstractBackgroundWhite matter hyperintensities (WMH) are markers of cerebral small vessel disease and have been associated with higher risk of typical amnestic Alzheimer’s disease (AD), however, little is known about the role of WMHs in atypical variants of AD. This study aimed to determine the frequency and distribution of WMH in two atypical variants of AD: logopenic progressive aphasia (LPA) and posterior cortical atrophy (PCA), and determine whether WMH are associated with demographic and clinical outcomes in these cohorts.MethodSeventy‐five LPA and 39 PCA patients were recruited by the Neurodegenerative Research Group and underwent two‐ dimensional fluid‐attenuated inversion recovery (FLAIR) MRI and Pittsburgh Compound B (PiB) PET. All patients were PiB‐positive. All FLAIR images were segmented using an in‐house software, then were manually edited by a trained imaging analyst. WMH volume was scaled by total intracranial volume to correct for differences in head size. Adjusted WMH burdens were compared between LPA and PCA, and associations were assessed between WMH burden and age, global PiB standard uptake value ratios (SUVRs), and clinical metrics, including tests of general cognitive impairment, executive function, naming, repetition, visuospatial and visuoperceptual function, and parkinsonism. Associations between WMHs and clinical scores were adjusted for age and brain volume to correct for atrophy.ResultTotal WMH burden was similar in PCA and LPA, however, regional differences were identified with PCA having greater subcortical WMH burden in right occipital, parietal, and temporal lobes compared to LPA, and LPA having greater WMH burden in left deep grey and white matter and greater subcortical burden in left parietal lobe compared to PCA. Total WMH burden was greater with older age in both groups but was not related to global PiB SUVR either globally or locally. Greater total WMH was associated with visuoperceptual performance in both cohorts after correcting for atrophy. When assessed regionally, visuoperceptual performance was associated with right subcortical occipital WMH across the cohort.ConclusionRegional distribution of WMHs differ across clinical variants of atypical AD and WMHs may impact clinical outcomes, although further work is needed to determine whether they reflect cerebrovascular disease or regionally specific neurodegenerative changes.

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.

AbstractBackgroundStudies of aging and dementia often acquire brain MRI and PET images, then de‐identify and share these widely with other researchers. It is known that research participants can potentially be re‐identified from brain MRI and CT using face recognition, but this has not been tested with PET imaging.MethodWe generated face reconstruction images of 182 volunteers using their existing amyloid (PiB), tau (FTP), and FDG PET scans from clinical GE PET/CT scanners, and we measured how often commercial automatic face recognition (Microsoft Azure) correctly matched these reconstructions with photographs of their faces. We additionally compared these match rates with face reconstructions from CT (from PET/CT) and MRI. When testing face recognition from PET scans, CT was used only during PET reconstruction, not for face recognition. We also validated our updated mri_reface software, originally developed to de‐identify (“de‐face”) faces in MRI, with new support for PET images. We measured face recognition rates before and after applying mri_reface to PET, CT, and MRI. In addition, we measured effects of de‐facing on biomarker measurements (global and regional SUVR) using 244 ADNI Florbetapir (FBP) amyloid PET scans with two automated pipelines: FreeSurfer 6.0 (PETSurfer) and an in‐house SPM12‐based pipeline.ResultRates of successful face recognition with PET ranged from 32‐41% (Table 1) across tracers. These were lower than MRI (98%) and CT (78%) but still concerning for participant privacy. Using mri_reface reduced face recognition to 0‐3.5% for PET, 5% for CT, and 8% for MRI. mri_reface had little effect on global amyloid SUVR measurements from ADNI PET images (Figure 1). ICC values between original and de‐faced images were 1.00 with biases and median absolute differences < 0.5%, across both pipelines. Effects on individual regions all had ICC values > 0.98 with biases and median absolute differences < 2%.ConclusionFace reconstructions from PET images have sufficient quality for potential re‐identification using publicly available face recognition technology. Research studies should consider using face de‐identification (de‐facing) software on PET images in addition to CT and structural MRI. Our free, automatic mri_reface software greatly reduced the potential for face recognition without substantially affecting SUVR measurements.

AbstractBackgroundResting state functional connectivity (rsfMRI) studies have shown functional network disruptions in logopenic progressive aphasia (LPA) (reduced connectivity within the left temporal language network) and posterior cortical atrophy (PCA) (reduced connectivity within ventral visual network). However, little is known about how these networks compare and how connectivity differs within and between networks in these atypical Alzheimer’s Disease (AD) phenotypes.MethodFifty‐nine amyloid‐positive PCA patients and eighty‐five amyloid‐positive LPA patients were recruited by the Neurodegenerative Research Group at Mayo Clinic, Rochester, MN, and underwent structural and resting state functional MRI. Spatially preprocessed data was analyzed using the CONN functional connectivity toolbox to explore the default mode network (DMN), salience, sensorimotor, left language, visual and memory networks, which were generated by averaging multiple regions‐of‐interest (ROI) within each network. The data was analyzed at both the voxel‐level and network‐level (ROI‐ROI).ResultVoxel‐level differences in functional connectivity between PCA and LPA were observed in all networks (p<0.05; cluster size, FWE=0.05). PCA showed reduced connectivity in the occipital lobe in visual and memory networks, parietal lobe in language and DMN, and sensorimotor strip in sensorimotor network, particularly on the right, compared to LPA. Conversely, LPA showed reduced connectivity in the frontal cortex in salience, memory, visual and sensorimotor networks, inferior parietal lobe in visual network, and anterior temporal pole and sensorimotor cortex in the language network, compared to PCA. Between network differences (p<0.05; cluster size, FDR=0.05) were observed between groups, with PCA showing reduced connectivity between the visual and sensorimotor networks, while LPA showed reduced connectivity between the language and sensorimotor networks. The DMN was anti‐correlated to the language network in PCA, while it was positively correlated to the language network in LPA. In both phenotypes, the DMN was anticorrelated with the salience network. The visual network in PCA showed connections to all networks, while in LPA, it was only connected to the salience, language and DMN.ConclusionThe atypical AD phenotypes were associated with multiple disruptions in connectivity, both within and between brain networks. Phenotype specific differences in connectivity patterns were noted in visual network for PCA and language network for LPA.