Anomaly Analysis of Alzheimer’s Disease in PET Images Using an Unsupervised Adversarial Deep Learning Model

In this study, the anomaly analysis of Alzheimer’s disease using positron emission tomography (PET) images using an unsupervised proposed adversarial model is investigated. The model consists of three parts: a parallel-network encoder, which is comprised of a convolutional pipeline and a dilated convolutional pipeline that extracts global and local features and concatenates them, a decoder that reconstructs the input image from the obtained feature vector, and a discriminator that distinguishes if the input image image is real or fake. The hypothesis is that if the proposed model is trained with only normal brain images, the corresponding construction loss for normal images should be minimal. However, if the input image belongs to a class that is designated as an anomaly that which the model is not trained with, then the construction loss will be high. This will reflect during the anomaly score comparison between the normal and the anomalous image. A multi-case analysis is performed for three major classes using the Alzheimer’s Disease Neuroimaging Initiative dataset, Alzheimer’s disease, mild cognitive impairment, and normal control. The base parallel-encoder network shows better classification accuracy than the benchmark models, and the proposed model that is built on the parallel model outperforms the benchmark anomaly detection models. The proposed model gave out 96.03% and 75.21% in classification and area under the curve score, respectively. Additionally, a qualitative evaluation done by using Fréchet inception distance gave a better score than the state-of-the-art by three points.

Download Full-text

Diagnosis and Prognosis of Alzheimer?s Disease Via 3D CNN

International Journal of Recent Technology and Engineering - 2 ◽

10.35940/ijrte.c5776.118419 ◽

2019 ◽

Vol 8 (4) ◽

pp. 54-63

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Image Data ◽

Brain Images ◽

Early Symptom ◽

Diagnosis And Prognosis ◽

Positron Emission ◽

Proposed Model ◽

3D Cnn ◽

Leaky Rectified Linear Unit

Mild Cognitive Impairment (MCI) is an early symptom of Alzheimer’s disease (AD). The feature extraction and deep learning architecture of the convolutional neural network in 3D brain images is applied to the problem of Alzheimer’s disease. The Structural Magnetic Resonance (sMRI) and Positron Emission Tomography (PET) image of the patient’s brain are classified according to the vigorousness of the disease and is labelled to be either in MCI or in AD or Normal Control (NC) condition. In this paper, we proposed a model and presented the baseline convolutional CNN with four layers viz., Convolutional layer, Leaky Rectified Linear Unit(LReLU), S3Pool layer and Global average pooling. Further, the 3D image data is used to perform the binary and ternary classifications and its performance are examined. The strength of the network has improved interior resource utilization evaluated with medical images, sMRI and PET on hippocampal ROI. The results of our proposed CNN architecture have achieved an accuracy level of 0.945, 0.859 and 0.748 respectively, when compared to the conventional AlexNet based network. The obtained data from the ADNI database shows better performance with our proposed model.

Download Full-text

The Who, When, Why, and How of PET Amyloid Imaging in Management of Alzheimer’s Disease—Review of Literature and Interesting Images

Diagnostics ◽

10.3390/diagnostics9020065 ◽

2019 ◽

Vol 9 (2) ◽

pp. 65 ◽

Cited By ~ 8

Author(s):

Subapriya Suppiah ◽

Mellanie-Anne Didier ◽

Sobhan Vinjamuri

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Area Under The Curve ◽

Standard Uptake Value ◽

Amyloid Imaging ◽

Carrier Status ◽

Amyloid Pet ◽

Positron Emission ◽

Cutoff Points ◽

Pet Amyloid

Amyloid imaging using positron emission tomography (PET) has an emerging role in the management of Alzheimer’s disease (AD). The basis of this imaging is grounded on the fact that the hallmark of AD is the histological detection of beta amyloid plaques (Aβ) at post mortem autopsy. Currently, there are three FDA approved amyloid radiotracers used in clinical practice. This review aims to take the readers through the array of various indications for performing amyloid PET imaging in the management of AD, particularly using 18F-labelled radiopharmaceuticals. We elaborate on PET amyloid scan interpretation techniques, their limitations and potential improved specificity provided by interpretation done in tandem with genetic data such as apolipiprotein E (APO) 4 carrier status in sporadic cases and molecular information (e.g., cerebral spinal fluid (CSF) amyloid levels). We also describe the quantification methods such as the standard uptake value ratio (SUVr) method that utilizes various cutoff points for improved accuracy of diagnosing AD, such as a threshold of 1.122 (area under the curve 0.894), which has a sensitivity of 92.3% and specificity of 90.5%, whereas the cutoff points may be higher in APOE ε4 carriers (1.489) compared to non-carriers (1.313). Additionally, recommendations for future developments in this field are also provided.

Download Full-text

18F-labelled CCR1-receptor antagonist is not suitable for imaging of Alzheimer's disease

Nuklearmedizin ◽

10.3413/nukmed-0457-12-01 ◽

2012 ◽

Vol 51 (06) ◽

pp. 239-243 ◽

Cited By ~ 1

Author(s):

V.A. Holthoff ◽

P. Mäding ◽

R. Bergmann ◽

B. Pawelke ◽

G. Holl ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Pet Imaging ◽

Amyloid Β ◽

Qualitative Evaluation ◽

Compartment Analysis ◽

Positron Emission ◽

Enhanced Expression ◽

Ccr1 Antagonist ◽

The Brain

SummaryDiagnosis of Alzheimer’s disease (AD) with positron emission tomography (PET) using 18F-fluorodeoxyglucose (FDG) relies on typical alterations of brain glucose metabolism which are, however, not disease specific. Amyloid- β imaging has not entered clinical routine yet. Post mortem histological specimen of brain tissue from AD patients revealed enhanced expression of the chemotactic cytocine receptor 1 (CCR1). Participants, methods: CCR1-antagonist ZK811460 was labeled with fluorine-18 to explore its possible use as specific diagnostic tool in AD. Tracer characterization comprising PET imaging of brain and metabolite analysis was performed in AD patients and controls. Results: Neither qualitative evaluation nor quantitative compartment analysis of PET data did show any enhanced binding of the 18F-labeled CCR1-antagonist in the brain of AD patients or controls. Conclusion: 18F-ZK811460 did not fulfill the expectation as diagnostic tracer in PET imaging of AD.

Download Full-text

Relationship of Hippocampal Volume to Amyloid Burden across Diagnostic Stages of Alzheimer's Disease

Dementia and Geriatric Cognitive Disorders ◽

10.1159/000441351 ◽

2015 ◽

Vol 41 (1-2) ◽

pp. 68-79 ◽

Cited By ~ 8

Author(s):

Paula T. Trzepacz ◽

Helen Hochstetler ◽

Peng Yu ◽

Peter Castelluccio ◽

Michael M. Witte ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Logistic Regression ◽

Area Under The Curve ◽

Hippocampal Volume ◽

Elastic Net ◽

Amyloid Pet ◽

Or Education ◽

Positron Emission ◽

Relationship Of

Aims: To assess how hippocampal volume (HV) from volumetric magnetic resonance imaging (vMRI) is related to the amyloid status at different stages of Alzheimer's disease (AD) and its relevance to patient care. Methods: We evaluated the ability of HV to predict the florbetapir positron emission tomography (PET) amyloid positive/negative status by group in healthy controls (HC, n = 170) and early/late mild cognitive impairment (EMCI, n = 252; LMCI, n = 136), and AD dementia (n = 75) subjects from the Alzheimer's Disease Neuroimaging Initiative Grand Opportunity (ADNI-GO) and ADNI2. Logistic regression analyses, including elastic net classification modeling with 10-fold cross-validation, were used with age and education as covariates. Results: HV predicted amyloid status only in LMCI using either logistic regression [area under the curve (AUC) = 0.71, p < 0.001] or elastic net classification modeling [positive predictive value (PPV) = 72.7%]. In EMCI, age (AUC = 0.70, p < 0.0001) and age and/or education (PPV = 63.1%), but not HV, predicted amyloid status. Conclusion: Using clinical neuroimaging, HV predicted amyloid status only in LMCI, suggesting that HV is not a biomarker surrogate for amyloid PET in clinical applications across the full diagnostic spectrum.

Download Full-text

Detecting Alzheimer’s disease biomarkers with a brief tablet-based cognitive battery: sensitivity to Aβ and tau PET

Alzheimer s Research & Therapy ◽

10.1186/s13195-021-00776-w ◽

2021 ◽

Vol 13 (1) ◽

Author(s):

Elena Tsoy ◽

Amelia Strom ◽

Leonardo Iaccarino ◽

Sabrina J. Erlhoff ◽

Collette A. Goode ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Health Assessment ◽

Area Under The Curve ◽

Cognitive Measures ◽

Regional Patterns ◽

Logistic Regression Models ◽

Positron Emission ◽

Tau Pet

Abstract Background β-amyloid (Aβ) and tau positron emission tomography (PET) detect the pathological changes that define Alzheimer’s disease (AD) in living people. Cognitive measures sensitive to Aβ and tau burden may help streamline identification of cases for confirmatory AD biomarker testing. Methods We examined the association of Brain Health Assessment (BHA) tablet-based cognitive measures with dichotomized Aβ -PET status using logistic regression models in individuals with mild cognitive impairment (MCI) or dementia (N = 140; 43 Aβ-, 97 Aβ+). We also investigated the relationship between the BHA tests and regional patterns of tau-PET signal using voxel-wise regression analyses in a subsample of 60 Aβ+ individuals with MCI or dementia. Results Favorites (associative memory), Match (executive functions and speed), and Everyday Cognition Scale scores were significantly associated with Aβ positivity (area under the curve [AUC] = 0.75 [95% CI 0.66–0.85]). We found significant associations with tau-PET signal in mesial temporal regions for Favorites, frontoparietal regions for Match, and occipitoparietal regions for Line Orientation (visuospatial skills) in a subsample of individuals with MCI and dementia. Conclusion The BHA measures are significantly associated with both Aβ and regional tau in vivo imaging markers and could be used for the identification of patients with suspected AD pathology in clinical practice.

Download Full-text

Formation of paired helical filaments in Alzheimer's disease

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100111628 ◽

1984 ◽

Vol 42 ◽

pp. 302-303

Author(s):

J. Metuzals ◽

D. F. Clapin ◽

V. Montpetit

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Frontal Lobe ◽

Giant Axon ◽

Paired Helical Filaments ◽

Normal Brain ◽

Protein Assemblies ◽

Electron Microscopic ◽

Helical Symmetry ◽

Structural Levels

Information on the conformation of paired helical filaments (PHF) and the neurofilamentous (NF) network is essential for an understanding of the mechanisms involved in the formation of the primary lesions of Alzheimer's disease (AD): tangles and plaques. The structural and chemical relationships between the NF and the PHF have to be clarified in order to discover the etiological factors of this disease. We are investigating by stereo electron microscopic and biochemical techniques frontal lobe biopsies from patients with AD and squid giant axon preparations. The helical nature of the lesion in AD is related to pathological alterations of basic properties of the nervous system due to the helical symmetry that exists at all hierarchic structural levels in the normal brain. Because of this helical symmetry of NF protein assemblies and PHF, the employment of structure reconstruction techniques to determine the conformation, particularly the handedness of these structures, is most promising. Figs. 1-3 are frontal lobe biopsies.

Download Full-text