Boosting the diagnostic power of amyloid-β PET using a data-driven spatially informed classifier for decision support

Abstract Background Amyloid-β (Aβ) PET has emerged as clinically useful for more accurate diagnosis of patients with cognitive decline. Aβ deposition is a necessary cause or response to the cellular pathology of Alzheimer’s disease (AD). Usual clinical and research interpretation of amyloid PET does not fully utilise all information regarding the spatial distribution of signal. We present a data-driven, spatially informed classifier to boost the diagnostic power of amyloid PET in AD. Methods Voxel-wise k-means clustering of amyloid-positive voxels was performed; clusters were mapped to brain anatomy and tested for their associations by diagnostic category and disease severity with 758 amyloid PET scans from volunteers in the AD continuum from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). A machine learning approach based on this spatially constrained model using an optimised quadratic support vector machine was developed for automatic classification of scans for AD vs non-AD pathology. Results This classifier boosted the accuracy of classification of AD scans to 81% using the amyloid PET alone with an area under the curve (AUC) of 0.91 compared to other spatial methods. This increased sensitivity to detect AD by 15% and the AUC by 9% compared to the use of a composite region of interest SUVr. Conclusions The diagnostic classification accuracy of amyloid PET was improved using an automated data-driven spatial classifier. Our classifier highlights the importance of considering the spatial variation in Aβ PET signal for optimal interpretation of scans. The algorithm now is available to be evaluated prospectively as a tool for automated clinical decision support in research settings.

Download Full-text

Nonlinear Classification of EEG recordings from patients with Alzheimer’s Disease using Gaussian Process Latent Variable Model

10.1101/2020.05.07.20093922 ◽

2020 ◽

Cited By ~ 2

Author(s):

S. Rajintha. A. S. Gunawardena ◽

Fei He ◽

Ptolemaios Sarrigiannis ◽

Daniel J. Blackburn

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Gaussian Process ◽

Manifold Learning ◽

Latent Variable ◽

Time Window ◽

Support Vector ◽

Nonlinear Dimensionality Reduction ◽

Variable Model

AbstractIn this work, nonlinear temporal features from multi-channel EEGs are used for the classification of Alzheimer’s disease patients from healthy individuals. This was achieved by temporal manifold learning using Gaussian Process Latent Variable Models (GPLVM) as a nonlinear dimensionality reduction technique. Classification of the extracted features was undertaken using a nonlinear Support Vector Machine. Comparisons were made against the linear counterpart, Principle Component Analysis while exploring the effect of the time window or EEG epoch length used. It was demonstrated that temporal manifold learning using GPLVM is better in extracting features that attain high separability and prediction accuracy. This work aims to set the significance of using GPLVM temporal manifold learning for EEG feature extraction in the classification of Alzheimer’s disease.

Download Full-text

O4-07-04: Support vector machine-based classification of 18F-flutemetamol scans in Alzheimer's disease, MCI and controls: comparison with visual reads and structural MRI-based classification

Alzheimer s & Dementia ◽

10.1016/j.jalz.2011.05.2015 ◽

2011 ◽

Vol 7 ◽

pp. S697-S697

Author(s):

Rik Vandenberghe ◽

Natalie Nelissen ◽

Koen Van Laere ◽

Lennart Thurfjell ◽

Chris Buckley ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Support Vector Machine ◽

Structural Mri ◽

Support Vector

Download Full-text

Decreased salivary lactoferrin levels are specific to Alzheimer’s disease

10.21203/rs.3.rs-19912/v1 ◽

2020 ◽

Author(s):

Marta González-Sánchez ◽

Fernando Bartolome ◽

Desiree Antequera ◽

Veronica Puertas-Martín ◽

Pilar González ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Diagnostic Performance ◽

Group Versus ◽

Amyloid Β ◽

University Hospital ◽

Control Group ◽

Amyloid Pet ◽

Prodromal Ad ◽

Study Participants

Abstract Background Efforts focused on developing new less invasive biomarkers for early Alzheimer’s disease (AD) diagnosis are substantial. Evidences of infectious pathogens in AD brains may suggest a deteriorated innate immune system in AD pathophysiology. We previously demonstrated reduced salivary levels of Lf in AD patients, one of the major antimicrobial peptides. Methods To assess the clinical utility of salivary Lf for AD diagnosis, we examine the relationship between salivary Lf and cerebral amyloid-β (Aβ) load in two different cross-sectional cohorts including patients with different neurodegenerative disorders. Study participants for cohort 1 (n = 116) were enrolled from the 12 de Octubre University Hospital Neurology Service in Madrid (Spain) and Pablo de Olavide University in Sevilla (Spain). Study participants for cohort 2 (n = 142) were enrolled as part of the Atherobrain - Heart to Head (H2H) project. Participants underwent neurological and neuropsychological examination, saliva sampling, and amyloid-Positron-Emission Tomography (PET) neuroimaging. Results The diagnostic performance of salivary Lf in the cohort 1 had an area under the curve [AUC] of 0.95 (0.911-0.992) for the differentiation of the prodromal AD/AD group positive for amyloid-PET (PET + ) versus healthy group, and 0.97 (0.924-1) versus the frontotemporal dementia (FTD) group. In the cohort 2, salivary Lf had also an excellent diagnostic performance in the health control group versus prodromal AD comparison: AUC 0.93 (95% CI 0.876-0.989). Salivary Lf detected prodromal AD and AD dementia distinguishing them from other dementias as FTD with over 87% sensitivity and 91% specificity. Conclusion Therefore, salivary Lf seems to have a very good diagnostic performance to detect AD. Our findings support the possible utility of salivary Lf as a new non-invasive and cost-effective AD biomarker.

Download Full-text

Longitudinal Observation of Early-Onset Alzheimer’s Disease Dementia with Positive CSF Biomarkers/Negative Amyloid-β Positron-Emission Tomography

Journal of the Korean Neurological Association ◽

10.17340/jkna.2021.3.19 ◽

2021 ◽

Vol 39 (3) ◽

pp. 214-218

Author(s):

Min Hye Kim ◽

Joonho Lee ◽

Hong Nam Kim ◽

In Ja Shin ◽

Keun Lee ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Positron Emission Tomography ◽

Early Onset ◽

Brain Magnetic Resonance Imaging ◽

Amyloid Β ◽

Emission Tomography ◽

Csf Biomarkers ◽

Amyloid Pet ◽

Positron Emission

We report a 61-year-old woman with clinical course for Alzheimer’s disease (AD) dementia and discordant amyloid-β positron-emission tomography (PET) and cerebrospinal fluid biomarkers. Brain magnetic resonance imaging revealed remarkable atrophy in the hippocampus. However, repeated delayed <sup>18</sup>F-flutemetamol brain amyloid PET images with 1 year-interval revealed no amyloid deposition, whereas her CSF revealed low Aβ42, high total tau and p-tau181. This discordant amyloid-β PET and CSF biomarkers in this early-onset AD dementia might be associated with her low resilience or mixed pathology.

Download Full-text

Associations of Fully Automated CSF and Novel Plasma Biomarkers With Alzheimer Disease Neuropathology at Autopsy

Neurology ◽

10.1212/wnl.0000000000012513 ◽

2021 ◽

pp. 10.1212/WNL.0000000000012513

Author(s):

Michel J. Grothe ◽

Alexis Moscoso ◽

Nicholas J. Ashton ◽

Thomas K. Karikari ◽

Juan Lantero-Rodriguez ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Csf Biomarkers ◽

Amyloid Pet ◽

Braak Stage ◽

Neuritic Plaques ◽

Neurofilament Light ◽

T Tau

Objective:To study cerebrospinal fluid (CSF) biomarkers of Alzheimer’s disease (AD) analyzed by fully automated Elecsys immunoassays in comparison to neuropathologic gold standards, and compare their accuracy to plasma phosphorylated tau (p-tau181) measured using a novel Simoa method.Methods:We studied ante-mortem Elecsys-derived CSF biomarkers in 45 individuals who underwent standardized post-mortem assessments of AD and non-AD neuropathologic changes at autopsy. In a subset of 26 participants, we also analysed ante-mortem levels of plasma p-tau181 and neurofilament light (NfL). Reference biomarker values were obtained from 146 amyloid-PET-negative healthy controls (HC).Results:All CSF biomarkers clearly distinguished pathology-confirmed AD dementia (N=27) from HC (AUCs=0.86-1.00). CSF total-tau (t-tau), p-tau181, and their ratios with Aβ1-42, also accurately distinguished pathology-confirmed AD from non-AD dementia (N=8; AUCs=0.94-0.97). In pathology-specific analyses, intermediate-to-high Thal amyloid phases were best detected by CSF Aβ1-42 (AUC[95% CI]=0.91[0.81-1]), while intermediate-to-high CERAD neuritic plaques and Braak tau stages were best detected by CSF p-tau181 (AUC=0.89[0.79-0.99] and 0.88[0.77-0.99], respectively). Optimal Elecsys biomarker cut-offs were derived at 1097/229/19 pg/ml for Aβ1-42, t-tau, and p-tau181. In the plasma subsample, both plasma p-tau181 (AUC=0.91[0.86-0.96]) and NfL (AUC=0.93[0.87-0.99]) accurately distinguished pathology-confirmed AD (N=14) from HC. However, only p-tau181 distinguished AD from non-AD dementia cases (N=4; AUC=0.96[0.88-1.00]), and showed a similar, though weaker, pathologic specificity for neuritic plaques (AUC=0.75[0.52-0.98]) and Braak stage (AUC=0.71[0.44-0.98]) as CSF p-tau181.Conclusions:Elecsys-derived CSF biomarkers detect AD neuropathologic changes with very high discriminative accuracy in-vivo. Preliminary findings support the use of plasma p-tau181 as an easily accessible and scalable biomarker of AD pathology.Classification of Evidence:This study provides Class II evidence that fully-automated CSF t-tau and p-tau181measurements discriminate between autopsy-confirmed Alzheimer's disease and other dementias.

Download Full-text

Classification of Alzheimer’s Disease and Mild Cognitive Impairment Based on Cortical and Subcortical Features from MRI T1 Brain Images Utilizing Four Different Types of Datasets

Journal of Healthcare Engineering ◽

10.1155/2020/3743171 ◽

2020 ◽

Vol 2020 ◽

pp. 1-14 ◽

Cited By ~ 1

Author(s):

Saidjalol Toshkhujaev ◽

Kun Ho Lee ◽

Kyu Yeong Choi ◽

Jang Jae Lee ◽

Goo-Rak Kwon ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Cognitive Impairment ◽

Mild Cognitive Impairment ◽

State Of The Art ◽

Principal Component ◽

Label Propagation ◽

Support Vector ◽

Brain Volumes

Alzheimer’s disease (AD) is one of the most common neurodegenerative illnesses (dementia) among the elderly. Recently, researchers have developed a new method for the instinctive analysis of AD based on machine learning and its subfield, deep learning. Recent state-of-the-art techniques consider multimodal diagnosis, which has been shown to achieve high accuracy compared to a unimodal prognosis. Furthermore, many studies have used structural magnetic resonance imaging (MRI) to measure brain volumes and the volume of subregions, as well as to search for diffuse changes in white/gray matter in the brain. In this study, T1-weighted structural MRI was used for the early classification of AD. MRI results in high-intensity visible features, making preprocessing and segmentation easy. To use this image modality, we acquired four types of datasets from each dataset’s server. In this work, we downloaded 326 subjects from the National Research Center for Dementia homepage, 123 subjects from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) homepage, 121 subjects from the Alzheimer’s Disease Repository Without Borders homepage, and 131 subjects from the National Alzheimer’s Coordinating Center homepage. In our experiment, we used the multiatlas label propagation with expectation–maximization-based refinement segmentation method. We segmented the images into 138 anatomical morphometry images (in which 40 features belonged to subcortical volumes and the remaining 98 features belonged to cortical thickness). The entire dataset was split into a 70 : 30 (training and testing) ratio before classifying the data. A principal component analysis was used for dimensionality reduction. Then, the support vector machine radial basis function classifier was used for classification between two groups—AD versus health control (HC) and early mild cognitive impairment (MCI) (EMCI) versus late MCI (LMCI). The proposed method performed very well for all four types of dataset. For instance, for the AD versus HC group, the classifier achieved an area under curve (AUC) of more than 89% for each dataset. For the EMCI versus LMCI group, the classifier achieved an AUC of more than 80% for every dataset. Moreover, we also calculated Cohen kappa and Jaccard index statistical values for all datasets to evaluate the classification reliability. Finally, we compared our results with those of recently published state-of-the-art methods.

Download Full-text

Elevated Tau PET Signal Depends on Abnormal Amyloid Levels and Correlates with Cognitive Impairment in Elderly Persons without Dementia

Journal of Alzheimer s Disease ◽

10.3233/jad-200526 ◽

2020 ◽

Vol 78 (1) ◽

pp. 395-404 ◽

Cited By ~ 1

Author(s):

Rui-Qi Zhang ◽

Shi-Dong Chen ◽

Xue-Ning Shen ◽

Yu-Xiang Yang ◽

Jia-Ying Lu ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Executive Function ◽

Amyloid Β ◽

Amyloid Pet ◽

Elderly Persons ◽

Cross Sectional ◽

Memory Decline ◽

Tau Pet

Background: The recent developed PET ligands for amyloid-β (Aβ) and tau allow these two neuropathological hallmarks of Alzheimer’s disease (AD) to be mapped and quantified in vivo and to be examined in relation to cognition. Objective: To assess the associations among Aβ, tau, and cognition in non-demented subjects. Methods: Three hundred eighty-nine elderly participants without dementia from the Alzheimer’s Disease Neuroimaging Initiative underwent tau and amyloid PET scans. Cross-sectional comparisons and longitudinal analyses were used to evaluate the relationship between Aβ and tau accumulation. The correlations between biomarkers of both pathologies and performance in memory and executive function were measured. Results: Increased amyloid-PET retention was associated with greater tau-PET retention in widespread cortices. We observed a significant tau increase in the temporal composite regions of interest over 24 months in Aβ+ but not Aβ– subjects. Finally, tau-PET retention but not amyloid-PET retention significantly explained the variance in memory and executive function. Higher level of tau was associated with greater longitudinal memory decline. Conclusion: These findings suggested PET-detectable Aβ plaque pathology may be a necessary antecedent for tau-PET signal elevation. Greater tau-PET retention may demonstrate poorer cognition and predict prospective memory decline in non-demented subjects.

Download Full-text

Unravelling the Association Between Amyloid-PET and Cerebrospinal Fluid Biomarkers in the Alzheimer’s Disease Spectrum: Who Really Deserves an A+?

Journal of Alzheimer s Disease ◽

10.3233/jad-210593 ◽

2021 ◽

pp. 1-12

Author(s):

Luca Sacchi ◽

Tiziana Carandini ◽

Giorgio Giulio Fumagalli ◽

Anna Margherita Pietroboni ◽

Valeria Elisa Contarino ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Cerebrospinal Fluid ◽

Amyloid Β ◽

Standardized Uptake Value ◽

Csf Biomarkers ◽

Amyloid Pet ◽

Linear Regression Models ◽

Csf Analysis ◽

Disease Spectrum

Background: Association between cerebrospinal fluid (CSF)-amyloid-β (Aβ)42 and amyloid-PET measures is inconstant across the Alzheimer’s disease (AD) spectrum. However, they are considered interchangeable, along with Aβ 42/40 ratio, for defining ‘Alzheimer’s Disease pathologic change’ (A+). Objective: Herein, we further characterized the association between amyloid-PET and CSF biomarkers and tested their agreement in a cohort of AD spectrum patients. Methods: We include ed 23 patients who underwent amyloid-PET, MRI, and CSF analysis showing reduced levels of Aβ 42 within a 365-days interval. Thresholds used for dichotomization were: Aβ 42 < 640 pg/mL (Aβ 42+); pTau > 61 pg/mL (pTau+); and Aβ 42/40 < 0.069 (ADratio+). Amyloid-PET scans were visually assessed and processed by four pipelines (SPMCL, SPMAAL, FSGM, FSWC). Results: Different pipelines gave highly inter-correlated standardized uptake value ratios (SUVRs) (rho = 0.93–0.99). The most significant findings were: pTau positive correlation with SPMCL SUVR (rho = 0.56, p = 0.0063) and Aβ 42/40 negative correlation with SPMCL and SPMAAL SUVRs (rho = –0.56, p = 0.0058; rho = –0.52, p = 0.0117 respectively). No correlations between CSF-Aβ 42 and global SUVRs were observed. In subregion analysis, both pTau and Aβ 42/40 values significantly correlated with cingulate SUVRs from any pipeline (R2 = 0.55–0.59, p < 0.0083), with the strongest associations observed for the posterior/isthmus cingulate areas. However, only associations observed for Aβ 42/40 ratio were still significant in linear regression models. Moreover, combining pTau with Aβ 42 or using Aβ 42/40, instead of Aβ 42 alone, increased concordance with amyloid-PET status from 74% to 91% based on visual reads and from 78% to 96% based on Centiloids. Conclusion: We confirmed that, in the AD spectrum, amyloid-PET measures show a stronger association and a better agreement with CSF-Aβ 42/40 and secondarily pTau rather than Aβ 42 levels.

Download Full-text