Diagnosis and prognosis of Alzheimer's disease using brain morphometry and white matter connectomes.
ABSTRACT: Accurate, reliable prediction of risk for Alzheimer's disease (AD) is essential for early, disease-modifying therapeutics. Multimodal MRI, such as structural and diffusion MRI, is likely to contain complementary information of neurodegenerative processes in AD. Here we tested the utility of the multimodal MRI (T1-weighted structure and diffusion MRI), combined with high-throughput brain phenotyping-morphometry and structural connectomics-and machine learning, as a diagnostic tool for AD. We used, firstly, a clinical cohort at a dementia clinic (National Health Insurance Service-Ilsan Hospital [NHIS-IH]; N = 211; 110 AD, 64 mild cognitive impairment [MCI], and 37 cognitively normal with subjective memory complaints [SMC]) to test the diagnostic models; and, secondly, Alzheimer's Disease Neuroimaging Initiative (ADNI)-2 to test the generalizability. Our machine learning models trained on the morphometric and connectome estimates (number of features = 34,646) showed optimal classification accuracy (AD/SMC: 97% accuracy, MCI/SMC: 83% accuracy; AD/MCI: 97% accuracy) in NHIS-IH cohort, outperforming a benchmark model (FLAIR-based white matter hyperintensity volumes). In ADNI-2 data, the combined connectome and morphometry model showed similar or superior accuracies (AD/HC: 96%; MCI/HC: 70%; AD/MCI: 75% accuracy) compared with the CSF biomarker model (t-tau, p-tau, and Amyloid β, and ratios). In predicting MCI to AD progression in a smaller cohort of ADNI-2 (n = 60), the morphometry model showed similar performance with 69% accuracy compared with CSF biomarker model with 70% accuracy. Our comparisons of the classifiers trained on structural MRI, diffusion MRI, FLAIR, and CSF biomarkers showed the promising utility of the white matter structural connectomes in classifying AD and MCI in addition to the widely used structural MRI-based morphometry, when combined with machine learning.
Project description:Combining machine learning with neuroimaging data has a great potential for early diagnosis of mild cognitive impairment (MCI) and Alzheimer's disease (AD). However, it remains unclear how well the classifiers built on one population can predict MCI/AD diagnosis of other populations. This study aimed to employ a spectral graph convolutional neural network (graph-CNN), that incorporated cortical thickness and geometry, to identify MCI and AD based on 3089 T1-weighted MRI data of the ADNI-2 cohort, and to evaluate its feasibility to predict AD in the ADNI-1 cohort (n = 3602) and an Asian cohort (n = 347). For the ADNI-2 cohort, the graph-CNN showed classification accuracy of controls (CN) vs. AD at 85.8% and early MCI (EMCI) vs. AD at 79.2%, followed by CN vs. late MCI (LMCI) (69.3%), LMCI vs. AD (65.2%), EMCI vs. LMCI (60.9%), and CN vs. EMCI (51.8%). We demonstrated the robustness of the graph-CNN among the existing deep learning approaches, such as Euclidean-domain-based multilayer network and 1D CNN on cortical thickness, and 2D and 3D CNNs on T1-weighted MR images of the ADNI-2 cohort. The graph-CNN also achieved the prediction on the conversion of EMCI to AD at 75% and that of LMCI to AD at 92%. The find-tuned graph-CNN further provided a promising CN vs. AD classification accuracy of 89.4% on the ADNI-1 cohort and >90% on the Asian cohort. Our study demonstrated the feasibility to transfer AD/MCI classifiers learned from one population to the other. Notably, incorporating cortical geometry in CNN has the potential to improve classification performance.
Project description:The role of structural brain magnetic resonance imaging (MRI) is becoming more and more emphasized in the early diagnostics of Alzheimer's disease (AD). This study aimed to assess the improvement in classification accuracy that can be achieved by combining features from different structural MRI analysis techniques. Automatically estimated MR features used are hippocampal volume, tensor-based morphometry, cortical thickness and a novel technique based on manifold learning. Baseline MRIs acquired from all 834 subjects (231 healthy controls (HC), 238 stable mild cognitive impairment (S-MCI), 167 MCI to AD progressors (P-MCI), 198 AD) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database were used for evaluation. We compared the classification accuracy achieved with linear discriminant analysis (LDA) and support vector machines (SVM). The best results achieved with individual features are 90% sensitivity and 84% specificity (HC/AD classification), 64%/66% (S-MCI/P-MCI) and 82%/76% (HC/P-MCI) with the LDA classifier. The combination of all features improved these results to 93% sensitivity and 85% specificity (HC/AD), 67%/69% (S-MCI/P-MCI) and 86%/82% (HC/P-MCI). Compared with previously published results in the ADNI database using individual MR-based features, the presented results show that a comprehensive analysis of MRI images combining multiple features improves classification accuracy and predictive power in detecting early AD. The most stable and reliable classification was achieved when combining all available features.
Project description:Prediction of conversion from mild cognitive impairment (MCI) to Alzheimer's disease (AD) is of major interest in AD research. A large number of potential predictors have been proposed, with most investigations tending to examine one or a set of related predictors. In this study, we simultaneously examined multiple features from different modalities of data, including structural magnetic resonance imaging (MRI) morphometry, cerebrospinal fluid (CSF) biomarkers and neuropsychological and functional measures (NMs), to explore an optimal set of predictors of conversion from MCI to AD in an Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort. After FreeSurfer-derived MRI feature extraction, CSF and NM feature collection, feature selection was employed to choose optimal subsets of features from each modality. Support vector machine (SVM) classifiers were then trained on normal control (NC) and AD participants. Testing was conducted on MCIc (MCI individuals who have converted to AD within 24 months) and MCInc (MCI individuals who have not converted to AD within 24 months) groups. Classification results demonstrated that NMs outperformed CSF and MRI features. The combination of selected NM, MRI and CSF features attained an accuracy of 67.13%, a sensitivity of 96.43%, a specificity of 48.28%, and an AUC (area under curve) of 0.796. Analysis of the predictive values of MCIc who converted at different follow-up evaluations showed that the predictive values were significantly different between individuals who converted within 12 months and after 12 months. This study establishes meaningful multivariate predictors composed of selected NM, MRI and CSF measures which may be useful and practical for clinical diagnosis.
Project description:Objective: Late-life depression (LLD) is associated with development of different types of dementia. Identification of LLD patients, who will develop cognitive decline, i.e., the early stage of dementia would help to implement interventions earlier. The purpose of this study was to assess whether structural brain magnetic resonance imaging (MRI) in LLD patients can predict mild cognitive impairment (MCI) or dementia 1 year prior to the diagnosis. Methods: LLD patients underwent brain MRI at baseline and repeated clinical assessment after 1-year. Structural brain measurements were obtained using Freesurfer software (v. 5.1) from the T1W brain MRI images. MRI-based Random Forest classifier was used to discriminate between LLD who developed MCI or dementia after 1-year follow-up and cognitively stable LLD. Additionally, a previously established Random Forest model trained on 185 patients with Alzheimer's disease (AD) vs. 225 cognitively normal elderly from the Alzheimer's disease Neuroimaging Initiative was tested on the LLD data set (ADNI model). Results: MCI and dementia diagnoses were predicted in LLD patients with 76%/68%/84% accuracy/sensitivity/specificity. Adding the baseline Mini-Mental State Examination (MMSE) scores to the models improved accuracy/sensitivity/specificity to 81%/75%/86%. The best model predicted MCI status alone using MRI and baseline MMSE scores with accuracy/sensitivity/specificity of 89%/85%/90%. The most important region for all the models was right ventral diencephalon, including hypothalamus. Its volume correlated negatively with the number of depressive episodes. ADNI model trained on AD vs. Controls using SV could predict MCI-DEM patients with 67% accuracy. Conclusion: LDD patients developing MCI and dementia can be discriminated from LLD patients remaining cognitively stable with good accuracy based on baseline structural MRI alone. Baseline MMSE score improves prediction accuracy. Ventral diencephalon, including the hypothalamus might play an important role in preservation of cognitive functions in LLD.
Project description:Based on previous studies, a preclinical classification for Alzheimer's disease (AD) has been proposed. However, 1) specificity of the different neuronal injury (NI) biomarkers has not been studied, 2) subjects with subtle cognitive impairment but normal NI biomarkers (SCINIB) have not been included in the analyses and 3) progression to mild cognitive impairment (MCI) or dementia of the AD type (DAT), referred to here as MCI/DAT, varies between studies. Therefore, we analyzed data from 486 cognitively normal (CN) and 327 DAT subjects in the AD Neuroimaging Initiative (ADNI)-1/GO/2 cohorts.In the ADNI-1 cohort (median follow-up of 6 years), 6.3% and 17.0% of the CN subjects developed MCI/DAT after 3 and 5 years follow-up, respectively. NI biomarker cutoffs [structural magnetic resonance imaging (MRI), fluorodeoxyglucose positron emission tomography (FDG-PET) and cerebrospinal fluid (CSF) tau] were established in DAT patients and memory composite scores were calculated in CN subjects in a cross-sectional sample (n?=?160). In the complete longitudinally followed CN ADNI cohort (n?=?326, median follow-up of 2 years), CSF and MRI values predicted an increased conversion to MCI/DAT. Different NI biomarkers showed important disagreements for classifying subjects as abnormal NI [kappa?=?(-0.05)-(0.33)] and into AD preclinical groups. SCINIB subjects (5.0%) were more prevalent than AD preclinical stage 3 subjects (3.4%) and showed a trend for increased progression to MCI/DAT.Different NI biomarkers lead to different classifications of ADNI subjects, while structural MRI and CSF tau measures showed the strongest predictive value for progression to MCI/DAT. The newly defined SCINIB category of ADNI subjects is more prevalent than AD preclinical stage individuals.
Project description:Mild cognitive impairment (MCI) is the prodromal stage of Alzheimer's disease (AD). Identifying MCI subjects who are at high risk of converting to AD is crucial for effective treatments. In this study, a deep learning approach based on convolutional neural networks (CNN), is designed to accurately predict MCI-to-AD conversion with magnetic resonance imaging (MRI) data. First, MRI images are prepared with age-correction and other processing. Second, local patches, which are assembled into 2.5 dimensions, are extracted from these images. Then, the patches from AD and normal controls (NC) are used to train a CNN to identify deep learning features of MCI subjects. After that, structural brain image features are mined with FreeSurfer to assist CNN. Finally, both types of features are fed into an extreme learning machine classifier to predict the AD conversion. The proposed approach is validated on the standardized MRI datasets from the Alzheimer's Disease Neuroimaging Initiative (ADNI) project. This approach achieves an accuracy of 79.9% and an area under the receiver operating characteristic curve (AUC) of 86.1% in leave-one-out cross validations. Compared with other state-of-the-art methods, the proposed one outperforms others with higher accuracy and AUC, while keeping a good balance between the sensitivity and specificity. Results demonstrate great potentials of the proposed CNN-based approach for the prediction of MCI-to-AD conversion with solely MRI data. Age correction and assisted structural brain image features can boost the prediction performance of CNN.
Project description:In North America, the Alzheimer's Disease Neuroimaging Initiative (ADNI) has established a platform to track the brain changes of Alzheimer's disease. A pilot study has been carried out in Europe to test the feasibility of the adoption of the ADNI platform (pilot E-ADNI).Seven academic sites of the European Alzheimer's Disease Consortium (EADC) enrolled 19 patients with mild cognitive impairment (MCI), 22 with AD, and 18 older healthy persons by using the ADNI clinical and neuropsychological battery. ADNI compliant magnetic resonance imaging (MRI) scans, cerebrospinal fluid, and blood samples were shipped to central repositories. Medial temporal atrophy (MTA) and white matter hyperintensities (WMH) were assessed by a single rater by using visual rating scales.Recruitment rate was 3.5 subjects per month per site. The cognitive, behavioral, and neuropsychological features of the European subjects were very similar to their U.S. counterparts. Three-dimensional T1-weighted MRI sequences were successfully performed on all subjects, and cerebrospinal fluid samples were obtained from 77%, 68%, and 83% of AD patients, MCI patients, and controls, respectively. Mean MTA score showed a significant increase from controls (left, right: 0.4, 0.3) to MCI patients (0.9, 0.8) to AD patients (2.3, 2.0), whereas mean WMH score did not differ among the three diagnostic groups (between 0.7 and 0.9). The distribution of both MRI markers was comparable to matched US-ADNI subjects.Academic EADC centers can adopt the ADNI platform to enroll MCI and AD patients and older controls with global cognitive and structural imaging features remarkably similar to those of the US-ADNI.
Project description:Atrophic changes in early Alzheimer's disease (AD) and amnestic mild cognitive impairment (MCI) have been proposed as biomarkers for detection and monitoring. We analyzed magnetic resonance imaging (MRI) atrophy rate from baseline to 1 year in 4 groups of participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI): AD (n = 152), converters from MCI to probable AD (MCI-C, n = 60), stable MCI (MCI-S, n = 261), and healthy controls (HC, n = 200). Scans were analyzed using multiple methods, including voxel-based morphometry (VBM), regions of interest (ROIs), and automated parcellation, permitting comparison of annual percent change (APC) in neurodegeneration markers. Effect sizes and the sample required to detect 25% reduction in atrophy rates were calculated. The influence of APOE genotype on APC was also evaluated. AD patients and converters from MCI to probable AD demonstrated high atrophy APCs across regions compared with minimal change in healthy controls. Stable MCI subjects showed intermediate atrophy rates. APOE genotype was associated with APC in key regions. In sum, APC rates are influenced by APOE genotype, imminent MCI to AD conversion, and AD-related neurodegeneration.
Project description:Various neuroimaging measures are being evaluated for tracking Alzheimer's disease (AD) progression in therapeutic trials, including measures of structural brain change based on repeated scanning of patients with magnetic resonance imaging (MRI). Methods to compute brain change must be robust to scan quality. Biases may arise if any scans are thrown out, as this can lead to the true changes being overestimated or underestimated. Here we analyzed the full MRI dataset from the first phase of Alzheimer's Disease Neuroimaging Initiative (ADNI-1) from the first phase of Alzheimer's Disease Neuroimaging Initiative (ADNI-1) and assessed several sources of bias that can arise when tracking brain changes with structural brain imaging methods, as part of a pipeline for tensor-based morphometry (TBM). In all healthy subjects who completed MRI scanning at screening, 6, 12, and 24months, brain atrophy was essentially linear with no detectable bias in longitudinal measures. In power analyses for clinical trials based on these change measures, only 39AD patients and 95 mild cognitive impairment (MCI) subjects were needed for a 24-month trial to detect a 25% reduction in the average rate of change using a two-sided test (?=0.05, power=80%). Further sample size reductions were achieved by stratifying the data into Apolipoprotein E (ApoE) ?4 carriers versus non-carriers. We show how selective data exclusion affects sample size estimates, motivating an objective comparison of different analysis techniques based on statistical power and robustness. TBM is an unbiased, robust, high-throughput imaging surrogate marker for large, multi-site neuroimaging studies and clinical trials of AD and MCI.
Project description:Regional volume atrophy and functional degeneration are key imaging hallmarks of Alzheimer's disease (AD) in structural and functional magnetic resonance imaging (MRI), respectively. We jointly explored regional volume atrophy and functional connectivity to better characterize neuroimaging data of AD and mild cognitive impairment (MCI). All data were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. We compared regional volume atrophy and functional connectivity in 10 subcortical regions using structural MRI and resting-state functional MRI (rs-fMRI). Neuroimaging data of normal controls (NC) (n = 35), MCI (n = 40), and AD (n = 30) were compared. Significant differences of regional volumes and functional connectivity measures between groups were assessed using permutation tests in 10 regions. The regional volume atrophy and functional connectivity of identified regions were used as features for the random forest classifier to distinguish among three groups. The features of the identified regions were also regarded as connectional fingerprints that could distinctively separate a given group from the others. We identified a few regions with distinctive regional atrophy and functional connectivity patterns for NC, MCI, and AD groups. A three label classifier using the information of regional volume atrophy and functional connectivity of identified regions achieved classification accuracy of 53.33% to distinguish among NC, MCI, and AD. We identified distinctive regional atrophy and functional connectivity patterns that could be regarded as a connectional fingerprint.