A workgroup commissioned by the Alzheimer’s Association (AA) and the National Institute on Aging (NIA) recently published research criteria for preclinical Alzheimer’s disease (AD). We performed a preliminary assessment of these guidelines.
We employed Pittsburgh compound B positron emission tomography (PET) imaging as our biomarker of cerebral amyloidosis and 18fluorodeoxyglucose PET imaging and hippocampal volume as biomarkers of neurodegeneration. A group of 42 clinically diagnosed AD subjects was used to create imaging biomarker cut-points. A group of 450 cognitively normal (CN) subjects from a population based sample was used to develop cognitive cut-points and to assess population frequencies of the different preclinical AD stages using different cut-point criteria.
The new criteria subdivide the preclinical phase of AD into stages 1–3. To classify our CN subjects, two additional categories were needed. Stage 0 denotes subjects with normal AD biomarkers and no evidence of subtle cognitive impairment. Suspected Non-AD Pathophysiology (SNAP) denotes subjects with normal amyloid PET imaging, but abnormal neurodegeneration biomarker studies. At fixed cut-points corresponding to 90% sensitivity for diagnosing AD and the 10th percentile of CN cognitive scores, 43% of our sample was classified as stage 0; 16% stage 1; 12 % stage 2; 3% stage 3; and 23% SNAP.
This cross-sectional evaluation of the NIA-AA criteria for preclinical AD indicates that the 1–3 staging criteria coupled with stage 0 and SNAP categories classify 97% of CN subjects from a population-based sample, leaving just 3% unclassified. Future longitudinal validation of the criteria will be important.
Imaging biomarkers are useful outcome measures in treatment trials. We compared sample size estimates for future treatment trials performed over 6 or 12-months in progressive supranuclear palsy using both imaging and clinical measures. We recruited 16 probable progressive supranuclear palsy patients that underwent baseline, 6 and 12 month brain scans, and 16 age-matched controls with serial scans. Disease severity was measured at each time-point using the progressive supranuclear palsy rating scale. Rates of ventricular expansion and rates of atrophy of the whole brain, superior frontal lobe, thalamus, caudate and midbrain were calculated. Rates of atrophy and clinical decline were used to calculate sample sizes required to power placebo-controlled treatment trials over 6 and 12-months. Rates of whole brain, thalamus and midbrain atrophy, and ventricular expansion, were increased over 6 and 12-months in progressive supranuclear palsy compared to controls. The progressive supranuclear palsy rating scale increased by 9 points over 6-months, and 18 points over 12-months. The smallest sample size estimates for treatment trials over 6-months were achieved using rate of midbrain atrophy, followed by rate of whole brain atrophy and ventricular expansion. Sample size estimates were further reduced over 12-month intervals. Sample size estimates for the progressive supranuclear palsy rating scale were worse than imaging measures over 6-months, but comparable over 12-months. Atrophy and clinical decline can be detected over 6-months in progressive supranuclear palsy. Sample size estimates suggest that treatment trials could be performed over this interval, with rate of midbrain atrophy providing the best outcome measure.
Progressive supranuclear palsy; atrophy; midbrain; power calculations; short interval
To systematically compare two techniques for measuring brain atrophy rates from serial magnetic resonance imaging (MRI) studies.
Materials and Methods
Using the separation in atrophy rate between cohorts of cognitively normal elderly subjects and patients with Alzheimer's disease (AD) as the gold standard, we evaluated 1) different methods of computing volume change; 2) different methods for steps in image preprocessing - intensity normalization, alignment mask used, and bias field correction; 3) the effect of MRI acquisition hardware changes; and 4) the sensitivity of the method to variations in initial manual volume editing. For each of the preceding evaluations, measurements of whole-brain and ventricular atrophy rates were calculated.
In general, greater separation between the clinical groups was seen with ventricular rather than whole-brain measures. Surprisingly, neither the use of bias field correction nor a major hardware change between the scan pairs affected group separation.
Atrophy rate measurements from serial MRI are candidates for use as surrogate markers of disease progression in AD and other dementing neurodegenerative disorders. The final method has excellent precision and accurately captures the expected biology of AD - arguably the two most important features if this technique is to be used as a biomarker of disease progression.
serial MRI; Alzheimer's Disease; Brain Atrophy
Task-free functional magnetic resonance imaging (TF-fMRI) has great potential for advancing the understanding and treatment of neurologic illness. However, as with all measures of neural activity, variability is a hallmark of intrinsic connectivity networks (ICNs) identified by TF-fMRI. This variability has hampered efforts to define a robust metric of connectivity suitable as a biomarker for neurologic illness. We hypothesized that some of this variability rather than representing noise in the measurement process, is related to a fundamental feature of connectivity within ICNs, which is their non-stationary nature. To test this hypothesis, we used a large (n = 892) population-based sample of older subjects to construct a well characterized atlas of 68 functional regions, which were categorized based on independent component analysis network of origin, anatomical locations, and a functional meta-analysis. These regions were then used to construct dynamic graphical representations of brain connectivity within a sliding time window for each subject. This allowed us to demonstrate the non-stationary nature of the brain’s modular organization and assign each region to a “meta-modular” group. Using this grouping, we then compared dwell time in strong sub-network configurations of the default mode network (DMN) between 28 subjects with Alzheimer’s dementia and 56 cognitively normal elderly subjects matched 1∶2 on age, gender, and education. We found that differences in connectivity we and others have previously observed in Alzheimer’s disease can be explained by differences in dwell time in DMN sub-network configurations, rather than steady state connectivity magnitude. DMN dwell time in specific modular configurations may also underlie the TF-fMRI findings that have been described in mild cognitive impairment and cognitively normal subjects who are at risk for Alzheimer’s dementia.
The common neurodegenerative pathologies underlying dementia are Alzheimer’s disease (AD), Lewy body disease (LBD) and Frontotemporal lobar degeneration (FTLD). Our aim was to identify patterns of atrophy unique to each of these diseases using antemortem structural-MRI scans of pathologically-confirmed dementia cases and build an MRI-based differential diagnosis system. Our approach of creating atrophy maps using structural-MRI and applying them for classification of new incoming patients is labeled Differential-STAND (Differential-diagnosis based on STructural Abnormality in NeuroDegeneration). Pathologically-confirmed subjects with a single dementing pathologic diagnosis who had an MRI at the time of clinical diagnosis of dementia were identified: 48 AD, 20 LBD, 47 FTLD-TDP (pathology-confirmed FTLD with TDP-43). Gray matter density in 91 regions-of-interest was measured in each subject and adjusted for head-size and age using a database of 120 cognitively normal elderly. The atrophy patterns in each dementia type when compared to pathologically-confirmed controls mirrored known disease-specific anatomic patterns: AD-temporoparietal association cortices and medial temporal lobe; FTLD-TDP-frontal and temporal lobes and LBD-bilateral amygdalae, dorsal midbrain and inferior temporal lobes. Differential-STAND based classification of each case was done based on a mixture model generated using bisecting k-means clustering of the information from the MRI scans. Leave-one-out classification showed reasonable performance compared to the autopsy gold-standard and clinical diagnosis: AD (sensitivity:90.7%; specificity:84 %), LBD (sensitivity:78.6%; specificity:98.8%) and FTLD-TDP (sensitivity:84.4%; specificity:93.8%). The proposed approach establishes a direct a priori relationship between specific topographic patterns on MRI and “gold standard” of pathology which can then be used to predict underlying dementia pathology in new incoming patients.
MRI; Alzheimer’s disease; Lewy body disease; Frontotemporal lobar degeneration
A major recent discovery was the identification of an expansion of a non-coding GGGGCC hexanucleotide repeat in the C9ORF72 gene in patients with frontotemporal dementia and amyotrophic lateral sclerosis. Mutations in two other genes are known to account for familial frontotemporal dementia: microtubule-associated protein tau and progranulin. Although imaging features have been previously reported in subjects with mutations in tau and progranulin, no imaging features have been published in C9ORF72. Furthermore, it remains unknown whether there are differences in atrophy patterns across these mutations, and whether regional differences could help differentiate C9ORF72 from the other two mutations at the single-subject level. We aimed to determine the regional pattern of brain atrophy associated with the C9ORF72 gene mutation, and to determine which regions best differentiate C9ORF72 from subjects with mutations in tau and progranulin, and from sporadic frontotemporal dementia. A total of 76 subjects, including 56 with a clinical diagnosis of behavioural variant frontotemporal dementia and a mutation in one of these genes (19 with C9ORF72 mutations, 25 with tau mutations and 12 with progranulin mutations) and 20 sporadic subjects with behavioural variant frontotemporal dementia (including 50% with amyotrophic lateral sclerosis), with magnetic resonance imaging were included in this study. Voxel-based morphometry was used to assess and compare patterns of grey matter atrophy. Atlas-based parcellation was performed utilizing the automated anatomical labelling atlas and Statistical Parametric Mapping software to compute volumes of 37 regions of interest. Hemispheric asymmetry was calculated. Penalized multinomial logistic regression was utilized to create a prediction model to discriminate among groups using regional volumes and asymmetry score. Principal component analysis assessed for variance within groups. C9ORF72 was associated with symmetric atrophy predominantly involving dorsolateral, medial and orbitofrontal lobes, with additional loss in anterior temporal lobes, parietal lobes, occipital lobes and cerebellum. In contrast, striking anteromedial temporal atrophy was associated with tau mutations and temporoparietal atrophy was associated with progranulin mutations. The sporadic group was associated with frontal and anterior temporal atrophy. A conservative penalized multinomial logistic regression model identified 14 variables that could accurately classify subjects, including frontal, temporal, parietal, occipital and cerebellum volume. The principal component analysis revealed similar degrees of heterogeneity within all disease groups. Patterns of atrophy therefore differed across subjects with C9ORF72, tau and progranulin mutations and sporadic frontotemporal dementia. Our analysis suggested that imaging has the potential to be useful to help differentiate C9ORF72 from these other groups at the single-subject level.
frontotemporal dementia; magnetic resonance imaging; C9ORF72; tau; progranulin
Functions of the ADNI MRI core fall into three categories: (1) those of the central MRI core lab at Mayo Clinic, Rochester, Minnesota, needed to generate high quality MRI data in all subjects at each time point; (2) those of the funded ADNI MRI core imaging analysis groups responsible for analyzing the MRI data, and (3) the joint function of the entire MRI core in designing and problem solving MR image acquisition, pre-processing and analyses methods. The primary objective of ADNI was and continues to be improving methods for clinical trials in Alzheimer's disease. Our approach to the present (“ADNI-GO”) and future (“ADNI-2”, if funded) MRI protocol will be to maintain MRI methodological consistency in previously enrolled “ADNI-1” subjects who are followed longitudinally in ADNI-GO and ADNI-2. We will modernize and expand the MRI protocol for all newly enrolled ADNI-GO and ADNI-2 subjects. All newly enrolled subjects will be scanned at 3T with a core set of three sequence types: 3D T1-weighted volume, FLAIR, and a long TE gradient echo volumetric acquisition for micro hemorrhage detection. In addition to this core ADNI-GO and ADNI-2 protocol, we will perform vendor specific pilot sub-studies of arterial spin labeling perfusion, resting state functional connectivity and diffusion tensor imaging. One each of these sequences will be added to the core protocol on systems from each MRI vendor. These experimental sub-studies are designed to demonstrate the feasibility of acquiring useful data in a multi-center (but single vendor) setting for these three emerging MRI applications.
The behavioural variant of frontotemporal dementia is a progressive neurodegenerative syndrome characterized by changes in personality and behaviour. It is typically associated with frontal lobe atrophy, although patterns of atrophy are heterogeneous. The objective of this study was to examine case-by-case variability in patterns of grey matter atrophy in subjects with the behavioural variant of frontotemporal dementia and to investigate whether behavioural variant of frontotemporal dementia can be divided into distinct anatomical subtypes. Sixty-six subjects that fulfilled clinical criteria for a diagnosis of the behavioural variant of frontotemporal dementia with a volumetric magnetic resonance imaging scan were identified. Grey matter volumes were obtained for 26 regions of interest, covering frontal, temporal and parietal lobes, striatum, insula and supplemental motor area, using the automated anatomical labelling atlas. Regional volumes were divided by total grey matter volume. A hierarchical agglomerative cluster analysis using Ward's clustering linkage method was performed to cluster the behavioural variant of frontotemporal dementia subjects into different anatomical clusters. Voxel-based morphometry was used to assess patterns of grey matter loss in each identified cluster of subjects compared to an age and gender-matched control group at P < 0.05 (family-wise error corrected). We identified four potentially useful clusters with distinct patterns of grey matter loss, which we posit represent anatomical subtypes of the behavioural variant of frontotemporal dementia. Two of these subtypes were associated with temporal lobe volume loss, with one subtype showing loss restricted to temporal lobe regions (temporal-dominant subtype) and the other showing grey matter loss in the temporal lobes as well as frontal and parietal lobes (temporofrontoparietal subtype). Another two subtypes were characterized by a large amount of frontal lobe volume loss, with one subtype showing grey matter loss in the frontal lobes as well as loss of the temporal lobes (frontotemporal subtype) and the other subtype showing loss relatively restricted to the frontal lobes (frontal-dominant subtype). These four subtypes differed on clinical measures of executive function, episodic memory and confrontation naming. There were also associations between the four subtypes and genetic or pathological diagnoses which were obtained in 48% of the cohort. The clusters did not differ in behavioural severity as measured by the Neuropsychiatric Inventory; supporting the original classification of the behavioural variant of frontotemporal dementia in these subjects. Our findings suggest behavioural variant of frontotemporal dementia can therefore be subdivided into four different anatomical subtypes.
behavioural variant frontotemporal dementia; atrophy; cluster analysis; voxel-based morphometry
Biomarkers of brain Aβ amyloid deposition can be measured either by cerebrospinal fluid Aβ42 or Pittsburgh compound B positron emission tomography imaging. Our objective was to evaluate the ability of Aβ load and neurodegenerative atrophy on magnetic resonance imaging to predict shorter time-to-progression from mild cognitive impairment to Alzheimer’s dementia and to characterize the effect of these biomarkers on the risk of progression as they become increasingly abnormal. A total of 218 subjects with mild cognitive impairment were identified from the Alzheimer’s Disease Neuroimaging Initiative. The primary outcome was time-to-progression to Alzheimer’s dementia. Hippocampal volumes were measured and adjusted for intracranial volume. We used a new method of pooling cerebrospinal fluid Aβ42 and Pittsburgh compound B positron emission tomography measures to produce equivalent measures of brain Aβ load from either source and analysed the results using multiple imputation methods. We performed our analyses in two phases. First, we grouped our subjects into those who were ‘amyloid positive’ (n = 165, with the assumption that Alzheimer's pathology is dominant in this group) and those who were ‘amyloid negative’ (n = 53). In the second phase, we included all 218 subjects with mild cognitive impairment to evaluate the biomarkers in a sample that we assumed to contain a full spectrum of expected pathologies. In a Kaplan–Meier analysis, amyloid positive subjects with mild cognitive impairment were much more likely to progress to dementia within 2 years than amyloid negative subjects with mild cognitive impairment (50 versus 19%). Among amyloid positive subjects with mild cognitive impairment only, hippocampal atrophy predicted shorter time-to-progression (P < 0.001) while Aβ load did not (P = 0.44). In contrast, when all 218 subjects with mild cognitive impairment were combined (amyloid positive and negative), hippocampal atrophy and Aβ load predicted shorter time-to-progression with comparable power (hazard ratio for an inter-quartile difference of 2.6 for both); however, the risk profile was linear throughout the range of hippocampal atrophy values but reached a ceiling at higher values of brain Aβ load. Our results are consistent with a model of Alzheimer’s disease in which Aβ deposition initiates the pathological cascade but is not the direct cause of cognitive impairment as evidenced by the fact that Aβ load severity is decoupled from risk of progression at high levels. In contrast, hippocampal atrophy indicates how far along the neurodegenerative path one is, and hence how close to progressing to dementia. Possible explanations for our finding that many subjects with mild cognitive impairment have intermediate levels of Aβ load include: (i) individual subjects may reach an Aβ load plateau at varying absolute levels; (ii) some subjects may be more biologically susceptible to Aβ than others; and (iii) subjects with mild cognitive impairment with intermediate levels of Aβ may represent individuals with Alzheimer’s disease co-existent with other pathologies.
mild cognitive impairment; amyloid imaging; magnetic resonance imaging; cerebrospinal fluid; Alzheimer’s disease biomarkers
MRI has evolved into an important diagnostic technique in medical imaging. However, reliability of the derived diagnosis can be degraded by artifacts, which challenge both radiologists and automatic computer-aided diagnosis. This paper proposes a fully automatic method for measuring image quality of 3D structural MRI. Quality measures are derived by analyzing the air background of magnitude images and are capable of detecting image degradation from several sources, including bulk motion, residual magnetization from incomplete spoiling, blurring, ghosting, etc. The method has been validated on 749 3D T1-weighted 1.5 T and 3 T head scans acquired at 36 Alzheimer's Disease Neuroimaging Initiative (ADNI) study sites operating with various software and hardware combinations. Results are compared against qualitative grades assigned by the ADNI quality control center (taken as the reference standard). The derived quality indices are independent of the MRI system used and agree with the reference standard quality ratings with high sensitivity and specificity (>85%). The proposed procedures for quality assessment could be of great value for both research and routine clinical imaging. It could greatly improve workflow through its ability to rule-out the need for a repeat scan while the patient is still in the magnet bore.
Magnetic resonance imaging; automatic quality assessment; image quality; artifacts detection
Rates of brain atrophy derived from serial magnetic resonance (MR) studies may be used to assess therapies for Alzheimer's disease (AD). These measures may be confounded by changes in scanner voxel sizes. For this reason, the Alzheimer's Disease Neuroimaging Initiative (ADNI) included the imaging of a geometric phantom with every scan. This study compares voxel scaling correction using a phantom with correction using a 9 degrees of freedom (9DOF) registration algorithm. We took 129 pairs of baseline and 1-year repeat scans, and calculated the volume scaling correction, previously measured using the phantom. We used the registration algorithm to quantify any residual scaling errors, and found the algorithm to be unbiased, with no significant (p = 0.97) difference between control (n = 79) and AD subjects (n = 50), but with a mean (SD) absolute volume change of 0.20 (0.20) % due to linear scalings. 9DOF registration was shown to be comparable to geometric phantom correction in terms of the effect on atrophy measurement and unbiased with respect to disease status. These results suggest that the additional expense and logistic effort of scanning a phantom with every patient scan can be avoided by registration-based scaling correction. Furthermore, based upon the atrophy rates in the AD subjects in this study, sample size requirements would be approximately 10–12% lower with (either) correction for voxel scaling than if no correction was used.
Scanner drift; Registration; Brain atrophy; Boundary shift integral; Alzheimer's disease
Voxel-based morphometry (VBM) is a popular method for probing inter-group differences in brain morphology. Variation in the detailed implementation of the algorithm, however, will affect the apparent results of VBM analyses and in turn the inferences drawn about the anatomic expression of specific disease states. We qualitatively assessed group comparisons of 43 normal elderly control subjects and 51 patients with probable Alzheimer's disease, using five different VBM variations. Based on the known pathologic expression of the disease, we evaluated the biological plausibility of each. The use of a custom template and custom tissue class prior probability images (priors) produced inter-group comparison maps with greater biological plausibility than the use of the Montreal Neurological Institute (MNI) template and priors. We present a method for initializing the normalization to a custom template, and conclude that, when incorporated into the VBM processing chain, it yields the most biologically plausible inter-group differences of the five methods presented.
We tested if rates of brain atrophy accelerate in individuals with amnestic mild cognitive impairment (aMCI) as they progress to typical late onset Alzheimer's Disease (AD). We included comparisons to aMCI subjects who did not progress (labeled aMCI-S) and also to cognitively normal elderly subjects (CN).
We studied 46 aMCI subjects who progressed to AD (labeled aMCI-P), 46 CN, and 23 aMCI-S. All subjects must have had three or more serial MRI scans. Rates of brain shrinkage and ventricular expansion were measured across all available serial MRI scans in each subject. Change in volumes relative to the point at which subjects progressed to a clinical diagnosis of AD (the index date) was modeled in aMCI-P. Change in volumes relative to age was modeled in all three clinical groups.
In aMCI-P the change in pre to post index rate (i.e. acceleration) of ventricular expansion was 1.7 cm3/yr, and acceleration in brain shrinkage was 5.3 cm3/yr. Brain volume declined and ventricular volume increased in all three groups with age. Volume changes decelerated with increasing age in aMCI-P, and to a lesser extent aMCI-S, but were linear in the matched CN. Among all aMCI subjects, rates of atrophy were greater in apolipoprotein E ε4 carriers than non-carriers.
Rates of atrophy accelerate as individuals progress from aMCI to typical late onset AD. Rates of atrophy are greater in younger than older aMCI-P and aMCI-S subjects. We did not find that atrophy rates varied with age in 70 – 90 year old CN subjects.
Tensor-based morphometry can recover three-dimensional longitudinal brain changes over time by nonlinearly registering baseline to follow-up MRI scans of the same subject. Here, we compared the anatomical distribution of longitudinal brain structural changes, over 12 months, using a subset of the ADNI dataset consisting of 20 patients with Alzheimer’s disease (AD), 40 healthy elderly controls, and 40 individuals with mild cognitive impairment (MCI). Each individual longitudinal change map (Jacobian map) was created using an unbiased registration technique, and spatially normalized to a geometrically-centered average image based on healthy controls. Voxelwise statistical analyses revealed regional differences in atrophy rates, and these differences were correlated with clinical measures and biomarkers. Consistent with prior studies, we detected widespread cerebral atrophy in AD, and a more restricted atrophic pattern in MCI. In MCI, temporal lobe atrophy rates were correlated with changes in mini-mental state exam (MMSE) scores, clinical dementia rating (CDR), and logical/verbal learning memory scores. In AD, temporal atrophy rates were correlated with several biomarker indices, including a higher CSF level of p-tau protein, and a greater CSF tau/beta amyloid 1-42 (ABeta42) ratio. Temporal lobe atrophy was significantly faster in MCI subjects who converted to AD than in non-converters. Serial MRI scans can therefore be analyzed with nonlinear image registration to relate ongoing neurodegeneration to a variety of pathological biomarkers, cognitive changes, and conversion from MCI to AD, tracking disease progression in 3-dimensional detail.
The purpose of this study was to use serial imaging to gain insight into the sequence of pathologic events in Alzheimer's disease, and the clinical features associated with this sequence. We measured change in amyloid deposition over time using serial 11C Pittsburgh compound B (PIB) positron emission tomography and progression of neurodegeneration using serial structural magnetic resonance imaging. We studied 21 healthy cognitively normal subjects, 32 with amnestic mild cognitive impairment and 8 with Alzheimer's disease. Subjects were drawn from two sources—ongoing longitudinal registries at Mayo Clinic, and the Alzheimer's disease Neuroimaging Initiative (ADNI). All subjects underwent clinical assessments, MRI and PIB studies at two time points, approximately one year apart. PIB retention was quantified in global cortical to cerebellar ratio units and brain atrophy in units of cm3 by measuring ventricular expansion. The annual change in global PIB retention did not differ by clinical group (P = 0.90), and although small (median 0.042 ratio units/year overall) was greater than zero among all subjects (P < 0.001). Ventricular expansion rates differed by clinical group (P < 0.001) and increased in the following order: cognitively normal (1.3 cm3/year) < amnestic mild cognitive impairment (2.5 cm3/year) < Alzheimer's disease (7.7 cm3/year). Among all subjects there was no correlation between PIB change and concurrent change on CDR-SB (r = −0.01, P = 0.97) but some evidence of a weak correlation with MMSE (r =−0.22, P = 0.09). In contrast, greater rates of ventricular expansion were clearly correlated with worsening concurrent change on CDR-SB (r = 0.42, P < 0.01) and MMSE (r =−0.52, P < 0.01). Our data are consistent with a model of typical late onset Alzheimer's disease that has two main features: (i) dissociation between the rate of amyloid deposition and the rate of neurodegeneration late in life, with amyloid deposition proceeding at a constant slow rate while neurodegeneration accelerates and (ii) clinical symptoms are coupled to neurodegeneration not amyloid deposition. Significant plaque deposition occurs prior to clinical decline. The presence of brain amyloidosis alone is not sufficient to produce cognitive decline, rather, the neurodegenerative component of Alzheimer's disease pathology is the direct substrate of cognitive impairment and the rate of cognitive decline is driven by the rate of neurodegeneration. Neurodegeneration (atrophy on MRI) both precedes and parallels cognitive decline. This model implies a complimentary role for MRI and PIB imaging in Alzheimer's disease, with each reflecting one of the major pathologies, amyloid dysmetabolism and neurodegeneration.
Alzheimer's disease; amyloid imaging; magnetic resonance imaging, longitudinal imaging; mild cognitive impairment; Pittsburgh compound B
To develop and validate a tool for Alzheimer's disease (AD) diagnosis in individual subjects using support vector machine (SVM) based classification of structural MR (sMR) images.
Libraries of sMR scans of clinically well characterized subjects can be harnessed for the purpose of diagnosing new incoming subjects.
190 patients with probable AD were age- and gender-matched with 190 cognitively normal (CN) subjects. Three different classification models were implemented: Model I uses tissue densities obtained from sMR scans to give STructural Abnormality iNDex (STAND)-score; and Models II and III use tissue densities as well as covariates (demographics and Apolipoprotein E genotype) to give adjusted-STAND (aSTAND)-score. Data from 140 AD and 140 CN were used for training. The SVM parameter optimization and training was done by four-fold cross validation. The remaining independent sample of 50 AD and 50 CN were used to obtain a minimally biased estimate of the generalization error of the algorithm.
The CV accuracy of Model II and Model III aSTAND-scores was 88.5% and 89.3% respectively and the developed models generalized well on the independent test datasets. Anatomic patterns best differentiating the groups were consistent with the known distribution of neurofibrillary AD pathology.
This paper presents preliminary evidence that application of SVM-based classification of an individual sMR scan relative to a library of scans can provide useful information in individual subjects for diagnosis of AD. Including demographic and genetic information in the classification algorithm slightly improves diagnostic accuracy.
support vector machines; classification; diagnosis; Alzheimer's
Measures of brain change can be computed from sequential MRI scans, providing valuable information on disease progression, e.g., for patient monitoring and drug trials. Tensor-based morphometry (TBM) creates maps of these brain changes, visualizing the 3D profile and rates of tissue growth or atrophy, but its sensitivity depends on the contrast and geometric stability of the images. A s part of the Alzheimer’s Disease Neuroimaging Initiative (ADNI), 17 normal elderly subjects were scanned twice (at a 2-week interval) with several 3D 1.5 T MRI pulse sequences: high and low flip angle SPGR/FLASH (from which Synthetic T1 images were generated), MP-RAGE, IR-SPGR (N = 10) and MEDIC (N = 7) scans. For each subject and scan type, a 3D deformation map aligned baseline and follow-up scans, computed with a nonlinear, inverse-consistent elastic registration algorithm. Voxelwise statistics, in ICBM stereotaxic space, visualized the profile of mean absolute change and its cross-subject variance; these maps were then compared using permutation testing. Image stability depended on: (1) the pulse sequence; (2) the transmit/receive coil type (birdcage versus phased array); (3) spatial distortion corrections (using MEDIC sequence information); (4) B1-field intensity inhomogeneity correction (using N3). SPGR/FLASH images acquired using a birdcage coil had least overall deviation. N3 correction reduced coil type and pulse sequence differences and improved scan reproducibility, except for Synthetic T1 images (which were intrinsically corrected for B1-inhomogeneity). No strong evidence favored B0 correction. Although SPGR/FLASH images showed least deviation here, pulse sequence selection for the ADNI project was based on multiple additional image analyses, to be reported elsewhere.
To characterize the shape of the trajectories of Alzheimer’s Disease (AD) biomarkers as a function of MMSE.
Longitudinal registries from the Mayo Clinic and the Alzheimer’s Disease Neuroimaging Initiative (ADNI).
Two different samples (n=343 and n=598) were created that spanned the cognitive spectrum from normal to AD dementia. Subgroup analyses were performed in members of both cohorts (n=243 and n=328) who were amyloid positive at baseline.
Main Outcome Measures
The shape of biomarker trajectories as a function of MMSE, adjusted for age, was modeled and described as baseline (cross-sectional) and within-subject longitudinal effects. Biomarkers evaluated were cerebro spinal fluid (CSF) Aβ42 and tau; amyloid and fluoro deoxyglucose position emission tomography (PET) imaging, and structural magnetic resonance imaging (MRI).
Baseline biomarker values generally worsened (i.e., non-zero slope) with lower baseline MMSE. Baseline hippocampal volume, amyloid PET and FDG PET values plateaued (i.e., non-linear slope) with lower MMSE in one or more analyses. Longitudinally, within-subject rates of biomarker change were associated with worsening MMSE. Non-constant within-subject rates (deceleration) of biomarker change were found in only one model.
Biomarker trajectory shapes by MMSE were complex and were affected by interactions with age and APOE status. Non-linearity was found in several baseline effects models. Non-constant within-subject rates of biomarker change were found in only one model, likely due to limited within-subject longitudinal follow up. Creating reliable models that describe the full trajectories of AD biomarkers will require significant additional longitudinal data in individual participants.
Alzheimer’s disease biomarkers; Magnetic Resonance Imaging; cerebro spinal fluid; amyloid PET imaging; FDG PET imaging
The Alzheimer's Disease Neuroimaging Initiative (ADNI) is a longitudinal multisite observational study of healthy elders, mild cognitive impairment (MCI), and Alzheimer's disease. Magnetic resonance imaging (MRI), (18F)-fluorode-oxyglucose positron emission tomography (FDG PET), urine serum, and cerebrospinal fluid (CSF) biomarkers, as well as clinical/psychometric assessments are acquiredat multiple time points. All data will be cross-linked and made available to the general scientific community. The purpose of this report is to describe the MRI methods employed in ADNI. The ADNI MRI core established specifications thatguided protocol development. A major effort was devoted toevaluating 3D T1-weighted sequences for morphometric analyses. Several options for this sequence were optimized for the relevant manufacturer platforms and then compared in a reduced-scale clinical trial. The protocol selected for the ADNI study includes: back-to-back 3D magnetization prepared rapid gradient echo (MP-RAGE) scans; B1-calibration scans when applicable; and an axial proton density-T2 dual contrast (i.e., echo) fast spin echo/turbo spin echo (FSE/TSE) for pathology detection. ADNI MRI methods seek to maximize scientific utility while minimizing the burden placed on participants. The approach taken in ADNI to standardization across sites and platforms of the MRI protocol, postacquisition corrections, and phantom-based monitoring of all scanners could be used as a model for other multisite trials.
MRI; Alzheimer's disease; clinical trials; imaging methods; imaging standardization