Search

BACKGROUND:Current methods of amyloid PET interpretation based on the binary classification of global amyloid signal fail to identify early phases of amyloid deposition. A recent analysis of 18F-florbetapir PET data from the Alzheimer's disease Neuroimaging Initiative cohort suggested a hierarchical four-stage model of regional amyloid deposition that resembles neuropathologic estimates and can be used to stage an individual's amyloid burden in vivo. Here, we evaluated the validity of this in vivo amyloid staging model in an independent cohort of older people with subjective memory complaints (SMC). We further examined its potential association with subtle cognitive impairments in this population at elevated risk for Alzheimer's disease (AD). METHODS:The monocentric INSIGHT-preAD cohort includes 318 cognitively intact older individuals with SMC. All individuals underwent 18F-florbetapir PET scanning and extensive neuropsychological testing. We projected the regional amyloid uptake signal into the previously proposed hierarchical staging model of in vivo amyloid progression. We determined the adherence to this model across all cases and tested the association between increasing in vivo amyloid stage and cognitive performance using ANCOVA models. RESULTS:In total, 156 participants (49%) showed evidence of regional amyloid deposition, and all but 2 of these (99%) adhered to the hierarchical regional pattern implied by the in vivo amyloid progression model. According to a conventional binary classification based on global signal (SUVRCereb = 1.10), individuals in stages III and IV were classified as amyloid-positive (except one in stage III), but 99% of individuals in stage I and even 28% of individuals in stage II were classified as amyloid-negative. Neither in vivo amyloid stage nor conventional binary amyloid status was significantly associated with cognitive performance in this preclinical cohort. CONCLUSIONS:The proposed hierarchical staging scheme of PET-evidenced

BackgroundCurrent methods of amyloid PET interpretation based on the binary classification of global amyloid signal fail to identify early phases of amyloid deposition. A recent analysis of 18F-florbetapir PET data from the Alzheimer's disease Neuroimaging Initiative cohort suggested a hierarchical four-stage model of regional amyloid deposition that resembles neuropathologic estimates and can be used to stage an individual's amyloid burden in vivo. Here, we evaluated the validity of this in vivo amyloid staging model in an independent cohort of older people with subjective memory complaints (SMC). We further examined its potential association with subtle cognitive impairments in this population at elevated risk for Alzheimer's disease (AD).MethodsThe monocentric INSIGHT-preAD cohort includes 318 cognitively intact older individuals with SMC. All individuals underwent 18F-florbetapir PET scanning and extensive neuropsychological testing. We projected the regional amyloid uptake signal into the previously proposed hierarchical staging model of in vivo amyloid progression. We determined the adherence to this model across all cases and tested the association between increasing in vivo amyloid stage and cognitive performance using ANCOVA models.ResultsIn total, 156 participants (49%) showed evidence of regional amyloid deposition, and all but 2 of these (99%) adhered to the hierarchical regional pattern implied by the in vivo amyloid progression model. According to a conventional binary classification based on global signal (SUVRCereb = 1.10), individuals in stages III and IV were classified as amyloid-positive (except one in stage III), but 99% of individuals in stage I and even 28% of individuals in stage II were classified as amyloid-negative. Neither in vivo amyloid stage nor conventional binary amyloid status was significantly associated with cognitive performance in this preclinical cohort.ConclusionsThe proposed hierarchical staging scheme of PET-evidenced a

BACKGROUND: Current methods of amyloid PET interpretation based on the binary classification of global amyloid signal fail to identify early phases of amyloid deposition. A recent analysis of 18F-florbetapir PET data from the Alzheimers disease Neuroimaging Initiative cohort suggested a hierarchical four-stage model of regional amyloid deposition that resembles neuropathologic estimates and can be used to stage an individuals amyloid burden in vivo. Here, we evaluated the validity of this in vivo amyloid staging model in an independent cohort of older people with subjective memory complaints (SMC). We further examined its potential association with subtle cognitive impairments in this population at elevated risk for Alzheimers disease (AD). METHODS: The monocentric INSIGHT-preAD cohort includes 318 cognitively intact older individuals with SMC. All individuals underwent 18F-florbetapir PET scanning and extensive neuropsychological testing. We projected the regional amyloid uptake signal into the previously proposed hierarchical staging model of in vivo amyloid progression. We determined the adherence to this model across all cases and tested the association between increasing in vivo amyloid stage and cognitive performance using ANCOVA models. RESULTS: In total, 156 participants (49%) showed evidence of regional amyloid deposition, and all but 2 of these (99%) adhered to the hierarchical regional pattern implied by the in vivo amyloid progression model. According to a conventional binary classification based on global signal (SUVRCereb = 1.10), individuals in stages III and IV were classified as amyloid-positive (except one in stage III), but 99% of individuals in stage I and even 28% of individuals in stage II were classified as amyloid-negative. Neither in vivo amyloid stage nor conventional binary amyloid status was significantly associated with cognitive performance in this preclinical cohort. CONCLUSIONS: The proposed hierarchical staging scheme of PET-evi

A (GGGGCC)n repeat expansion in C9orf72 gene is the major cause of frontotemporal dementia (FTD) and amyotrophic lateral sclerosis (ALS). The relations between the repeats size and the age at disease onset (AO) or the clinical phenotype (FTD vs. ALS) were investigated in 125 FTD, ALS, and presymptomatic carriers. Positive correlations were found between repeats number and the AO (p < 10e−4) but our results suggested that the association was mainly driven by age at collection (p < 10e−4). A weaker association was observed with clinical presentation (p = 0.02), which became nonsignificant after adjustment for the age at collection in each group. Importantly, repeats number variably expanded or contracted over time in carriers with multiple blood samples, as well as through generations in parent-offspring pairs, conversely to what occurs in several expansion diseases with anticipation at the molecular level. Finally, this study establishes that measure of repeats number in lymphocytes is not a reliable biomarker predictive of the AO or disease outcome in C9orf72 long expansion carriers.

A large number of papers have introduced novel machine learning and feature extraction methods for automatic classification of AD. However, they are difficult to reproduce because key components of the validation are often not readily available. These components include selected participants and input data, image preprocessing and cross-validation procedures. The performance of the different approaches is also difficult to compare objectively. In particular, it is often difficult to assess which part of the method provides a real improvement, if any. We propose a framework for reproducible and objective classification experiments in AD using three publicly available datasets (ADNI, AIBL and OASIS). The framework comprises: i) automatic conversion of the three datasets into BIDS format, ii) a modular set of preprocessing pipelines, feature extraction and classification methods, together with an evaluation framework, that provide a baseline for benchmarking the different components. We demonstrate the use of the framework for a large-scale evaluation on 1960 participants using T1 MRI and FDG PET data. In this evaluation, we assess the influence of different modalities, preprocessing, feature types, classifiers, training set sizes and datasets. Performances were in line with the state-of-the-art. FDG PET outperformed T1 MRI for all classification tasks. No difference in performance was found for the use of different atlases, image smoothing, partial volume correction of FDG PET images, or feature type. Linear SVM and L2-logistic regression resulted in similar performance and both outperformed random forests. The classification performance increased along with the number of subjects used for training. Classifiers trained on ADNI generalized well to AIBL and OASIS. All the code of the framework and the experiments is publicly available at: https://gitlab.icm-institute.org/aramislab/AD-ML.

In recent years, the number of papers on Alzheimer's disease classification has increased dramatically, generating interesting methodological ideas on the use machine learning and feature extraction methods. However, practical impact is much more limited and, eventually, one could not tell which of these approaches are the most efficient. While over 90\% of these works make use of ADNI an objective comparison between approaches is impossible due to variations in the subjects included, image pre-processing, performance metrics and cross-validation procedures. In this paper, we propose a framework for reproducible classification experiments using multimodal MRI and PET data from ADNI. The core components are: 1) code to automatically convert the full ADNI database into BIDS format; 2) a modular architecture based on Nipype in order to easily plug-in different classification and feature extraction tools; 3) feature extraction pipelines for MRI and PET data; 4) baseline classification approaches for unimodal and multimodal features. This provides a flexible framework for benchmarking different feature extraction and classification tools in a reproducible manner. We demonstrate its use on all (1519) baseline T1 MR images and all (1102) baseline FDG PET images from ADNI 1, GO and 2 with SPM-based feature extraction pipelines and three different classification techniques (linear SVM, anatomically regularized SVM and multiple kernel learning SVM). The highest accuracies achieved were: 91% for AD vs CN, 83% for MCIc vs CN, 75% for MCIc vs MCInc, 94% for AD-A$\beta$+ vs CN-A$\beta$- and 72% for MCIc-A$\beta$+ vs MCInc-A$\beta$+. The code is publicly available at https://gitlab.icm-institute.org/aramislab/AD-ML (depends on the Clinica software platform, publicly available at http://www.clinica.run).