Objective: Alzheimer's disease (AD) is a neurodegenerative disorder that severely impairs cognitive function across various age groups, ranging from early to late sixties. It progresses from mild to severe stages, so an accurate diagnostic tool is necessary for effective intervention and treatment planning. Methods: This work proposes a novel AD classification architecture that integrates depthwise separable convolutional layers with traditional convolutional layers to efficiently extract features from structural magnetic resonance imaging (sMRI) scans. This model benefits from excellent feature extraction and lightweight operation, which reduces the number of parameters without compromising accuracy. The model learns from scratch with optimized weight initialization, resulting in faster convergence and improved generalization. However, medical imaging datasets contain class imbalance as a major challenge, which often results in biased models with poor generalization to the underrepresented disease stages. A hybrid sampling approach combining SMOTE (synthetic minority oversampling technique) with the ENN (edited nearest neighbors) effectively handles the complications of class imbalance issue inherent in the datasets. An explainable activation space occlusion sensitivity map (ASOP) pixel attribution method is employed to highlight the critical regions of input images that influence the classification decisions across different stages of AD. Results and Conclusions: The proposed model outperformed several state-of-the-art transfer learning architectures, including VGG19, DenseNet201, EfficientNetV2S, MobileNet, ResNet152, InceptionV3, and Xception. It achieves noteworthy results in disease stage classification, with an accuracy of 98.87%, an F1 score of 98.86%, a precision of 98.80%, and recall of 98.69%. These results demonstrate the effectiveness of the proposed model for classifying stages of AD progression.
87Soluble TREM2 changes during the clinical course of Alzheimer’s disease: A meta-analysis
Soluble TREM2 changes during the clinical course of Alzheimer’s disease: A meta-analysis
5Dynamic Change of Intracellular Metabolism of Microglia Evaluated by Transcriptomics in an Alzheimer's Mouse Model.
Microglia play diverse roles in Alzheimer's disease (AD). Intracellular metabolism has been indicated an important factor in modulating the function of microglia. However, it is not clear whether the intracellular metabolism of microglia changes dynamically in different stages of AD. To determine whether microglia intracellular metabolism changes dynamically in different stages of AD. Microglia were extracted from APPSwe/PS1dE9 (APP/PS1) mice and wild-type littermates at 2, 4, and 8 months old by fluorescence-activated cell sorting and used for RNA-sequencing analysis and quantitative PCR. Morphologies of amyloid plaques and microglia were detected by immunofluorescence staining. Compared with control littermates, the microglia of APP/PS1 mice exhibited significant transcriptional changes at 2-month-old before microglia morphological alterations and the plaque formation. The changes continued drastically following age with defined morphological shift of microglia and amyloid plaque enhancement in brains. Further analysis of those genotype and age dependent transcriptomic changes revealed that differentially expressed genes were enriched in pathways related to energy metabolism. Compared with wild-type mice, there were changes of some vital genes related to glucose metabolism and lipid metabolism pathways in APP/PS1 mice at different ages. Glucose metabolism may play a major role in early activation of microglia, and lipid metabolism may be more important in later activation period. Our results showed that microglia actively participate in the pathological progress of AD. The intracellular metabolism of microglia changed significantly in different stages of AD, even preceding amyloid-β deposition.
Read more428A Deep Learning Approach for Automated Diagnosis and Multi-Class Classification of Alzheimer's Disease Stages Using Resting-State fMRI and Residual Neural Networks.
Alzheimer's disease (AD) is an incurable neurodegenerative disorder accounting for 70%-80% dementia cases worldwide. Although, research on AD has increased in recent years, however, the complexity associated with brain structure and functions makes the early diagnosis of this disease a challenging task. Resting-state functional magnetic resonance imaging (rs-fMRI) is a neuroimaging technology that has been widely used to study the pathogenesis of neurodegenerative diseases. In literature, the computer-aided diagnosis of AD is limited to binary classification or diagnosis of AD and MCI stages. However, its applicability to diagnose multiple progressive stages of AD is relatively under-studied. This study explores the effectiveness of rs-fMRI for multi-class classification of AD and its associated stages including CN, SMC, EMCI, MCI, LMCI, and AD. A longitudinal cohort of resting-state fMRI of 138 subjects (25 CN, 25 SMC, 25 EMCI, 25 LMCI, 13 MCI, and 25AD) from Alzheimer's Disease Neuroimaging Initiative (ADNI) is studied. To provide a better insight into deep learning approaches and their applications to AD classification, we investigate ResNet-18 architecture in detail. We consider the training of the network from scratch by using single-channel input as well as performed transfer learning with and without fine-tuning using an extended network architecture. We experimented with residual neural networks to perform AD classification task and compared it with former research in this domain. The performance of the models is evaluated using precision, recall, f1-measure, AUC and ROC curves. We found that our networks were able to significantly classify the subjects. We achieved improved results with our fine-tuned model for all the AD stages with an accuracy of 100%, 96.85%, 97.38%, 97.43%, 97.40% and 98.01% for CN, SMC, EMCI, LMCI, MCI, and AD respectively. However, in terms of overall performance, we achieved state-of-the-art results with an average accuracy of 97.92% and 97.88% for off-the-shelf and fine-tuned models respectively. The Analysis of results indicate that classification and prediction of neurodegenerative brain disorders such as AD using functional magnetic resonance imaging and advanced deep learning methods is promising for clinical decision making and have the potential to assist in early diagnosis of AD and its associated stages.
Read more14Multimodal Neuroimaging Fusion for Alzheimer's Disease: An Image Colorization Approach With Mobile Vision Transformer
ABSTRACTMultimodal neuroimaging, combining data from different sources, has shown promise in the classification of the Alzheimer's disease (AD) stage. Existing multimodal neuroimaging fusion methods exhibit certain limitations, which require advancements to enhance their objective performance, sensitivity, and specificity for AD classification. This study uses the use of a Pareto‐optimal cosine color map to enhance classification performance and visual clarity of fused images. A mobile vision transformer (ViT) model, incorporating the swish activation function, is introduced for effective feature extraction and classification. Fused images from the Alzheimer's Disease Neuroimaging Initiative (ADNI), the Whole Brain Atlas (AANLIB), and Open Access Series of Imaging Studies (OASIS) datasets, obtained through optimized transposed convolution, are utilized for model training, while evaluation is achieved using images that have not been fused from the same databases. The proposed model demonstrates high accuracy in AD classification across different datasets, achieving 98.76% accuracy for Early Mild Cognitive Impairment (EMCI) versus LMCI, 98.65% for Late Mild Cognitive Impairment (LMCI) versus AD, 98.60% for EMCI versus AD, and 99.25% for AD versus Cognitive Normal (CN) in the ADNI dataset. Similarly, on OASIS and AANLIB, the precision of the AD versus CN classification is 99.50% and 96.00%, respectively. Evaluation metrics showcase the model's precision, recall, and F1 score for various binary classifications, emphasizing its robust performance.
Read more8Investigation of plasma metabolomics and neurotransmitter dysfunction in the process of Alzheimer's disease rat induced by amyloid beta 25-35.
Alzheimer's disease (AD) has become one of the major diseases endangering the health of the elderly. Clarifying the features of each AD animal model is valuable for understanding the onset and progression of diseases and developing potential treatments in the pharmaceutical industry. In this study, we aimed to clarify plasma metabolomics and neurotransmitter dysfunction in the process of AD model rat induced by amyloid beta 25-35 (Aβ 25-35). Firstly, Morris Water Maze (MWM) test was used to investigate cognitive impairment in AD rat after 2, 4 and 8 weeks of modelling. Based on this, the effects on levels of AD-related enzymes and eight neurotransmitters were analyzed. And plasma metabolomics analysis based on ultra high-performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (UHPLC-Q-TOF-MS) was used to research the metabolic disturbances in the process of AD rat. The results shown the injury on the spatial learning ability of AD rats was gradually aggravated within 4 weeks, reached the maximum at 4 weeks and then was stable until 8 weeks. During 8 weeks of modeling, the levels of enzymes including β-secretase, γ-secretase, glycogen synthase kinase-3β (GSK-3β), acetyl cholinesterase (AchE) and nitric oxide synthase (NOS) were significant increased in the plasma of AD rats. The neurotransmitter dysfunction was mainly involved in γ-aminobutyric acid (GABA), acetyl choline (Ach), glutamic acid (Glu), 5-hydroxytryptamine (5-HT), dopamine (DA) and norepinephrine (NE). 17 endogenous metabolites correlated with AD were successfully detected in the metabolomics analysis. These metabolites were mainly involved in fatty acids, sphingolipids, and sterols metabolisms, vitamin metabolism, and amino acid metabolism. These metabolites might be the potential biomarkers that correctly mark different stages of AD. The study on peripheral plasma indices reflecting the process of AD laid the foundation for understand the pathophysiology of AD and find an effective and radical cure. And the rules of endogenous metabolic disorder in AD rats also have a certain guiding significance for the future study of food–drug interactions at different stages of AD.
Read more3Sex differences in functional connectivity and the predictive role of the connectome-based predictive model in Alzheimer's disease.
Alzheimer's disease (AD) is a neurodegenerative disease characterized by cognitive decline. Sex differences in the progression of AD exist, but the neural mechanisms are not well understood. The purpose of the current study was to explore sex differences in brain functional connectivity (FC) at different stages of AD and their predictive ability on Montreal Cognitive Assessment (MoCA) scores using connectome-based predictive modeling (CPM). Resting-state functional magnetic resonance imaging was collected from 81 AD patients (44 females), 78 amnestic mild cognitive impairment patients (44 females), and 92 healthy controls (50 females). The FC analysis was conducted and the interaction effect between sex and group was investigated using two-factor variance analysis. The CPM was used to predict MoCA scores. There were sex-by-group interaction effects on FC between the left dorsolateral superior frontal gyrus and left middle temporal gyrus, left precuneus and right calcarine fissure surrounding cortex, left precuneus and left middle occipital gyrus, left middle temporal gyrus and left precentral gyrus, and between the left middle temporal gyrus and right cuneus. In the CPM, the positive network predictive model significantly predicted MoCA scores in both males and females. There were significant sex-by-group interaction effects on FC between the left precuneus and left middle occipital gyrus, and between the left middle temporal gyrus and right cuneus could predict MoCA scores in female patients. Our results suggest that there are sex differences in FC at different stages of AD. The sex-specific FC can further predict MoCA scores at individual level.
Read more
16Plasma oligomer beta-amyloid is associated with disease severity and cerebral amyloid deposition in Alzheimer’s disease spectrum
BackgroundMultimer detection system-oligomeric amyloid-β (MDS-OAβ) is a measure of plasma OAβ, which is associated with Alzheimer’s disease (AD) pathology. However, the relationship between MDS-OAβ and disease severity of AD is not clear. We aimed to investigate MDS-OAβ levels in different stages of AD and analyze the association between MDS-OAβ and cerebral Aβ deposition, cognitive function, and cortical thickness in subjects within the AD continuum.MethodsIn this cross-sectional study, we analyzed a total 126 participants who underwent plasma MDS-OAβ, structural magnetic resonance image of brain, and neurocognitive measures using Korean version of the Consortium to Establish a Registry for Alzheimer’s Disease, and cerebral Aβ deposition or amyloid positron emission tomography (A-PET) assessed by [18F] flutemetamol PET. Subjects were divided into 4 groups: N = 39 for normal control (NC), N = 31 for A-PET-negative mild cognitive impairment (MCI) patients, N = 30 for A-PET-positive MCI patients, and N = 22 for AD dementia patients. The severity of cerebral Aβ deposition was expressed as standard uptake value ratio (SUVR).ResultsCompared to the NC (0.803 ± 0.27), MDS-OAβ level was higher in the A-PET-negative MCI group (0.946 ± 0.137) and highest in the A-PET-positive MCI group (1.07 ± 0.17). MDS-OAβ level in the AD dementia group was higher than in the NC, but it fell to that of the A-PET-negative MCI group level (0.958 ± 0.103). There were negative associations between MDS-OAβ and cognitive function and both global and regional cerebral Aβ deposition (SUVR). Cortical thickness of the left fusiform gyrus showed a negative association with MDS-OAβ when we excluded the AD dementia group.ConclusionsThese findings suggest that MDS-OAβ is not only associated with neurocognitive staging, but also with cerebral Aβ burden in patients along the AD continuum.
Read more10Vascular factors in Alzheimer’s disease
The vascular factor in Alzheimer’s disease (AD), affecting its development and progression, is one of the most urgent problems of modern neuroangiology. The research investigates the characteristics of cerebral angioarchitectonics identified at different stages of AD. The research included 106 patients: 1) The Test Group—47 patients suffering from various stages of AD; 2) The Control Group—59 patients suffering from the most common lesions of the brain accompanied by neurodegenerative changes, the development of dementia and cognitive impairment, but not having AD. All the patients underwent: the testing of cognitive functions (MMSE), the determination of severity of dementia (CDR) and AD stages (TDR), computed tomography (CT), magnetic resonance imaging (MRI), scintigraphy of the brain (SG), rheoencephalography (REG), and cerebral multigated angiography (MUGA). Patients with AD different stages showed the following changes in angioarchitectonics and microcirculation: Absence of pronounced atherosclerotic lesions of intracranial vessels, reduction of the capillary bed in the temporal and temporo-parietal regions, development of multiple arteriovenous shunts in the same areas, early venous discharge, abnormal expansion of venous trunks that receive blood from arteriovenous shunts, venous congestion at the border of the frontal and parietal region, increased looping of intracranial arteries. Control Group patients had no combination of the abovementioned changes. These vascular changes are specific for AD and are in fact the vascular factor of this disease; they may also be called dyscirculatory angiopathy of Alzheimer’s type (DAAT). Patients suffering from other diseases that are accompanied by neurodegenerative changes in the brain, dementia and cognitive impairment do not have them.
Read more27Proteomic profiling of brain cortex tissues in a Tau transgenic mouse model of Alzheimer’s disease
Proteomic profiling of brain cortex tissues in a Tau transgenic mouse model of Alzheimer’s disease
8MRI-based mild cognitive impairment and Alzheimer's disease classification using an algorithm of combination of variational autoencoder and other machine learning classifiers
Background Correctly diagnosing mild cognitive impairment (MCI) and Alzheimer's disease (AD) is important for patient selection in drug discovery. Research outcomes on stage diagnosis using neuroimages combined with cerebrospinal fluid and genetic biomarkers are expensive and time-consuming. Only structural magnetic resonance imaging (sMRI) scans from two internationally recognized datasets are employed as input as well as test and independent validation to determine the classification of dementia by the machine learning algorithm. Objective We extract the reduced dimensional latent feature vector from the sMRI scans using a variational autoencoder (VAE). The objective is to classify AD, MCI, and control (CN) using MRI and without any other information. Methods The extracted feature vectors from MRI scans by VAE are used as input conditions for different advanced machine-learning classifiers. Classification of AD/CN/MCI are conducted using the output of VAE from MRI images and different artificial intelligence/machine learning classifier models in two cohorts. Results Using only MRI scans, the primary goal of the study is to test the ability to classify AD from CN and MCI cases. The current study achieved classification accuracies of AD versus CN 75.45% (F1-score = 79.52%), AD versus MCI 81.41% (F1-Score = 87.06%), and autopsy-confirmed AD versus MCI 92.75% (F1-Score = 95.52%) in test sets and AD versus CN 86.16% (F1-score = 92.03%) and AD versus MCI 70.03% (F1-Score = 82.1%) in validation data set. Conclusions By overcoming the data leakage problem, the autopsy-confirmed machine learning classification model is tested in two independent cohorts. External validation by an independent cohort improved the quality and novelty of the classification algorithm.
Read moreA short‐hybrid scale for detecting anosognosia from the prodromal stage of Alzheimer’s disease
BackgroundWhile not originally developed for this purpose, evidence suggests that the Healthy Aging Brain Care Monitor questionnaire (HABC‐M; Monahan et al, 2012, 2014) may be a valuable instrument for assessing unawareness of cognitive decline (i.e., anosognosia) from preclinical stages of Alzheimer’s disease (AD) (Cacciamani et al, 2017). This is relevant because anosognosia may delay AD diagnosis and therapeutic interventions, with adverse consequences for patients and caregivers. However, several open questions remain: What is the actual effectiveness of this instrument across the AD spectrum? How do the cognitive, functional, and psychological subscales of the HABC‐M differ in their sensitivity to anosognosia in AD? Does the caregiver burden subscale provide additional information about AD patient’s anosognosia? The present study aimed to address these specific questions.MethodsThe HABC‐M was administered to 60 participants (30 AD patients and 30 healthy controls) and their partners (caregivers/informants). A measure of self‐awareness was then estimated by computing the discrepancy scores between the participants and their partners for the different subscales of HABC‐M (with the exception of the caregiver burden subscale, which is a caregiver‐exclusive questionnaire). To evaluate whether these measures can discriminate between patients at different stages of AD and controls, we fit linear models adjusted for age, sex, and education, and computed receiver operating characteristic curves.ResultsAll HABC‐M discrepancy scores, as well as the caregiver burden score, were able to discriminate controls from patients in distinct AD stages. For best performance, the area‐under‐curve for the HABC‐M cognitive subscale discriminated prodromal AD versus controls with a value 0.88 [95% CI: 0.78;0.97]. Adding the caregiver burden score raised it to 0.94 [0.86;0.99].ConclusionThe HABC‐M cognitive subscale (six items) can detect anosognosia from the earliest stages of AD. Separately, the caregiver burden subscale (four items) may corroborate the suspicion of anosognosia. A short‐hybrid scale combining these 10 items, instead of the 31 items of the total HABC‐M questionnaire, could provide a practical and efficient instrument for detection of anosognosia in the clinical setting. Importantly, such an instrument could further contribute to a timely AD diagnosis.
Read more3Incompleteness features in the descriptive discourse of Chinese elders with and without Alzheimer’s disease
Alzheimer’s disease (AD) can manifest itself with prominent language dysfunction. Incompleteness in discourse refers to the lack of indispensable sentence-constructing elements that hinder communication fluency and accuracy. The current study investigates how the pattern of incompleteness is associated with the descriptive discourse produced by elders without AD and those with different stages of AD. The Chinese discourse samples were collected from the picture description of 40 elders with mild probable AD (Mini-Mental State Examination (MMSE) 21–26, Montreal Cognitive Assessment Scale-Basic (MoCA-B) 15–19), 40 elders with moderate probable AD (MMSE 11–20, MoCA-B 10–14), and 40 controls (MMSE 26–29, MoCA-B 24–29). The total production of incomplete sentences and six incompleteness features were examined. The Mild AD, Moderate AD, and Control groups differed in the total output of the incomplete sentence. Group differences also emerged in four incompleteness features: missing subject, missing predicate, missing object, and missing functional word. The Moderate AD group differed from the Mild AD group with respect to most significant features, while Mild AD and Control groups were very similar. The results suggested that AD impairs the sentence construction ability of Chinese elders, especially at the later stage. These statistically significant differences between the groups might provide some references when diagnosing the risk and possibility of cognitive impairment of Chinese elders, facilitating the design of clinical evaluation or screening for probable AD.
Read more184Early changes in extracellular matrix in Alzheimer's disease.
Although changes in extracellular matrix (ECM) scaffold have been reported previously in Alzheimer's disease (AD) compared to normal ageing, it is not known how alterations in the numerous components of the perivascular ECM might occur at different stages of AD. This study therefore investigates potential changes in basement membrane-associated ECM molecules in relation to increasing Braak stages. Thirty patients were divided into three groups (control subject, subclinical AD and AD patients). ECM levels of collagen IV, perlecan and fibronectin as well as human platelet endothelial cell adhesion molecule (hPECAM) were quantified by immunohistochemistry. Von Willebrand factor staining was measured to assess vessel density. Expression levels were correlated with the presence of amyloid plaques. Collagen IV, perlecan and fibronectin expression was increased in subclinical AD and AD patients when compared to controls, in frontal and temporal cortex, whilst no further increase was detected between subclinical AD and AD. These changes were not associated with an increase in vessel density, which was instead decreased in the temporal cortex of AD patients. In contrast, hPECAM levels remained unchanged. Finally, we found similar pattern in levels of amyloid deposition between the different Braak stages and showed that changes in ECM components correlated with amyloid deposition. Present data support the hypothesis that significant ECM changes occur during the early stages of AD. ECM changes affecting brain microvascular functions could therefore drive disease progression and provide potential new early investigational biomarkers in AD.
Read more7Clinical and biomarker trajectories in sporadic Alzheimer's disease: A longitudinal study
BackgroundIt has been documented that β‐amyloid (Aβ) deposition, tau pathology and neuronal degeneration precedes clinical symptoms in autosomal dominant Alzheimer's disease (AD). Since the divergence on pathogenesis between sporadic and familial AD limits the extension of these findings to sporadic AD, where and magnitude of pathologic processes in sporadic AD are of significant importance to clinical management.MethodIn this longitudinal study, we analyzed data from 982 ADNI participants with abnormal amyloid accumulations who underwent cerebrospinal fluid (CSF) test, brain imaging and cognitive assessments. We compared the levels of clinical outcomes and biomarkers at baseline and the rates of change in follow‐up between different stages of AD. The slope of the regression model over years was applied to estimate rates of change for Aβ deposition, tau alteration, hypometabolism, brain atrophy, and cognitive decline.ResultIn total 167 healthy controls, 133 patients with preclinical AD, 451 patients with MCI due to AD, and 231 patients with dementia due to AD were studied. The levels of clinical and biomarkers at baseline and their change rates at follow‐up significantly differed among the four groups. In terms of average years, Aβ deposition was observed over a range of 45·6 years for CSF Aβ, and 31·1 years with amyloid imaging, although Aβ markers began to be abnormal at 27·8 years for CSF Aβ and 26·3 years for neuroimaging. When Aβ markers exceeded the threshold, CSF tau and other non‐ Aβ markers started to change, and patients were estimated to take 29·1 years (CSF tau),12·7 years (memory), 12·2 years (hippocampus atrophy), 10·1years (hypometabolism) to transition from clinically normal to dementia. Finally, CSF p‐tau significantly increased as dementia onset approached.ConclusionOur results reveal the trajectory of biomarkers in sporadic Alzheimer's disease is led by Aβ accumulation, followed by CSF tau change, memory deficits, brain atrophy, hypometabolism, cognitive decline, and lastly CSF p‐tau increase.
Read more44The role of TREM2 in Alzheimer's disease: from the perspective of Tau.
Triggering receptor expressed on myeloid cells 2 (TREM2), a pattern recognition receptor abundantly expressed on microglia, has been identified as one of the risk factors for Alzheimer's disease (AD). Several studies have already demonstrated the relationship between TREM2 and Tau. TREM2 mutations and altered expression play an important role in Tau phosphorylation. Furthermore, the level of Tau phosphorylation is correlated with soluble TREM2 (sTREM2). However, in different stages of AD, TREM2 seems to have varying effects on Tau pathology. The explicit interaction between TREM2 and Tau, as well as how they affect AD pathology, remains unclear, and there is much evidence to the contrary that requires rational interpretation. Reviewing the dual roles of TREM2 in AD will help identify a more appropriate development strategy for targeting TREM2 to treat AD. Therefore, this review focuses on the interplay between Tau and TREM2 in relation to AD.
Read more
Copyright 2026 Cactus Communications. All rights reserved.
