Alzheimer's disease diagnosis and classification using deep learning techniques.

Multimodal neuroimaging has gained traction in Alzheimer's Disease (AD) diagnosis by integrating information from multiple imaging modalities to enhance classification accuracy. However, effectively handling heterogeneous data sources and overcoming the challenges posed by multiscale transform methods remains a significant hurdle. This article proposes a novel approach to address these challenges. To harness the power of diverse neuroimaging data, we employ a strategy that leverages optimized convolution techniques. These optimizations include varying kernel sizes and the incorporation of instance normalization, both of which play crucial roles in feature extraction from magnetic resonance imaging (MRI) and positron emission tomography (PET) images. Specifically, varying kernel sizes allow us to adapt the receptive field to different image characteristics, enhancing the model's ability to capture relevant information. Furthermore, we employ transposed convolution, which increases spatial resolution of feature maps, and it is optimized with varying kernel sizes and instance normalization. This heightened resolution facilitates the alignment and integration of data from disparate MRI and PET data. The use of larger kernels and strides in transposed convolution expands the receptive field, enabling the model to capture essential cross-modal relationships. Instance normalization, applied to each modality during the fusion process, mitigates potential biases stemming from differences in intensity, contrast, or scale between modalities. This enhancement contributes to improved model performance by reducing complexity and ensuring robust fusion. The performance of the proposed fusion method is assessed on three distinct neuroimaging datasets, which include: Alzheimer's Disease Neuroimaging Initiative (ADNI), consisting of 50 participants each at various stages of AD for both MRI and PET (Cognitive Normal, AD, and Early Mild Cognitive); Open Access Series of Imaging Studies (OASIS), consisting of 50 participants each at various stages of AD for both MRI and PET (Cognitive Normal, Mild Dementia, Very Mild Dementia); and whole-brain atlas neuroimaging (AANLIB) (consisting of 50 participants each at various stages of AD for both MRI and PET (Cognitive Normal, AD). To evaluate the quality of the fused images generated via our method, we employ a comprehensive set of evaluation metrics, including Structural Similarity Index Measurement (SSIM), which assesses the structural similarity between two images; Peak Signal-to-Noise Ratio (PSNR), which measures how closely the generated image resembles the ground truth; Entropy (E), which assesses the amount of information preserved or lost during fusion; the Feature Similarity Indexing Method (FSIM), which assesses the structural and feature similarities between two images; and Edge-Based Similarity (EBS), which measures the similarity of edges between the fused and ground truth images. The obtained fused image is further evaluated using a Mobile Vision Transformer. In the classification of AD vs. Cognitive Normal, the model achieved an accuracy of 99.00%, specificity of 99.00%, and sensitivity of 98.44% on the AANLIB dataset.

Reducing variations in multi-center Alzheimer's disease classification with convolutional adversarial autoencoder.

Alzheimer's disease damages brain neurons, resulting in memory loss. Early and accurate diagnosis of the disease is crucial for implementing preventive measures. However, differentiating between Alzheimer’s and healthy data in older adults is challenging due to the similarities in their brain patterns and intensities, complicating researchers' efforts to make an accurate diagnosis. Therefore, the research aims to use machine learning to improve diagnosis and classification of the disease, such as support vector machines (SVM), decision trees, and feedforward neural networks (FFNN). Classification algorithms were applied to the Alzheimer’s disease dataset, including 2149 cases, and the models were evaluated through metrics (Accuracy, Precision, Recall, specificity, F1 Score, F2 Score, F3 score, and AUC). Following data analysis and obtaining the results, we reached the Decision Tree model excels across all metrics, achieving high scores in accuracy (96.32%), precision (94.63%), recall (95%), specificity (97.05%), and AUC (94.96%). This demonstrates its ability to correctly identify true positives and negatives, and reduce false positives and negatives, makes it the most reliable model for accurately classifying Alzheimer's disease cases. In contrast, the SVM linear and FFANN models offer a good balance with accuracy (83.53% and 83.57%), specificity (89.27% and 91.72%), and AUC (89.63% and 89.84%). However, their lower recall (73.03% and 68.68%) compared to the Decision Tree may result in missed positive cases, making them less effective for classification. The SVM RBF model is the least effective option, with high precision and specificity but poor performance across all other metrics and lacks overall balance, resulting in a high number of false negatives. metrics and lacks overall balance, resulting in a high number of false negatives. Conclusion: The decision tree model outperforms other models, making it the best choice for Alzheimer's disease classification.

Background: Machine learning is now a powerful tool to help improve medical disorders diagnosis. One of its critical applications is the classification or clustering of neurodegenerative disease by pattern recognition methods based on biomedical signals and medical images. Early detection of these diseases is always useful and vital. In this study, we focused on Alzheimer’s disease (AD) as a type of dementia leading to problems with memory, thinking, and behavior. This disease was named after Dr. Alois Alzheimer in 1906 when he inspected a female patient who died of an unusual mental illness. According to recent studies, four stages are introduced for AD, including pre-dementia, early AD, moderate AD, and advanced AD. There are several methods for AD diagnosis that include mental status evaluation, physical exam, and neurological exam, based on different imaging techniques such as magnetic resonance imaging (MRI). Several methods have been introduced until now for the classification and detection of AD using machine learning algorithms, such as the classification of AD with discrete wavelet transform (DWT) and single linear discriminant analysis (LDA) classifiers and differentiation of AD from normal based on T2-weighted MRI with shearlet transform (ST) and K-nearest neighbors (KNN) classifiers. Objectives: In this work, we proposed a methodology based on DWT with three-level decomposition feature extraction (Figure 1). Methods: Based on statistics (mean, variance, skewness) of features and principal component analysis (PCA) for dimension reduction, we used five classifiers, including multi-layer perceptron neural network (MLPNN) (Figure 2), KNN, support vector machine (SVM) (Figure 3), and naive Bayesian (NB) with the majority vote method to fuse them into one ensemble classifier. Results: The proposed methodology was evaluated using 100 T2-weighted MRI of AD and cognitive normal (CN) subjects, which were chosen from the Harvard Medical School website. The accuracy, specificity, and sensitivity achieved from our methodology were 95%, 90%, and 100%, respectively by using a 10-fold cross-validation strategy. Conclusion: Our study showed that the stacking method for classification of AD and CN was better than using one classifier and comparable with state-of-the-art methods.

The disease popularly known as Alzheimer’s disease (AD), which causes cognitive impairment, is extensively distributed throughout the world. The lack of a cure or strategy to stop the growth of the disease notwithstanding, it is one of the most extensively studied illnesses in medicine and healthcare. However, patients with AD have a variety of treatment options to select from, including pharmaceuticals or non-drug alternatives, to help them maintain their quality of life. Over time, patients will need to be treated differently based on how far into the disease process they are. In this regard, early detection and classification of the illness stages can have a substantial impact on the treatment of symptoms. Early detection of Alzheimer’s disease is critical for the analysis of patient medications and the selection of appropriate specialists. The goal of this project is to provide a comprehensive comparison of the techniques used in the detection and classification of Alzheimer’s disease. In medical research, machine learning algorithms can be used to uncover facts, particularly in disease prediction. Support vector machines (SVM), decision trees, Bayes classifiers, K-nearest neighbors (KNN), gradient boosting, and other machine learning algorithms are used to diagnose a variety of diseases and conditions. The application of machine learning algorithms can result in the quick and highly accurate prediction and classification of this disease. This project helps in analyses of different machine learning techniques and algorithms that are used to predict and classify and provides an extensive accuracy comparison of the machine learning techniques used for the prediction and classification of Alzheimer’s disease.

Alzheimer's Disease (AD) is considered one of the most prevalent neurological disorders, primarily affecting elderly people and adversely impacting their brain functions. This disease is characterized by the gradual deterioration of cognitive functions, particularly memory, leading to varying information loss levels, typically classified within the framework of dementia. In this context, Artificial Intelligence (AI) methods, especially Convolutional Neural Networks (CNN) and Optuna optimization, have emerged as effective technique to provide efficient approach (or model) for classifying AD levels. The proposed approach is called the Optuna-CNN (O-CNN). It has the ability to identify the optimal CNN architecture by employing the optimization of Optuna. It is designed to classify AD into four levels (or stages): Mild Dementia (MD), Moderate Dementia (MoD), No Dementia (ND) and Very Mild Dementia (VMD). Two datasets are utilized here: the Best Alzheimer's MRI Dataset (BA-MRI) and the Alzheimer's MRI Dataset (AD-MRI), both are obtained from Kaggle platform. Following extensive training and implementations of the O-CNN method, high classification accuracies 94.79% for the AD-MRI dataset and 99.18% for the BA-MRI dataset are achieved.

Efficient multimodel method based on transformers and CoAtNet for Alzheimer's diagnosis

Functional magnetic resonance imaging (fMRI) is widely used to capture brain functional activities, but its high-dimensional nature and temporal complexity pose significant challenges for feature representation. To address these issues, this study proposes an end-to-end deep learning network for fMRI feature encoding and validates its effectiveness through Alzheimer's disease (AD) classification task. The proposed network first employs a 3D CNN to perform spatial encoding of fMRI at each time point, extracting local spatial patterns of brain images. Then, a transformer attention block specifically designed for 3D MRI is designed, which enhances spatial feature modeling through 3D positional encoding and captures long-range dependencies between brain regions. Finally, a cascaded transformer module is constructed to integrate spatial features across different time points, modeling the dynamic changes in brain activity. Experimental results on two fMRI datasets from ADNI database demonstrate that the proposed method improves feature representation and AD classification performance. By effectively capturing both the spatial and temporal characteristics of fMRI data, this approach provides a robust solution for automated fMRI feature encoding. The code is available at https://github.com/Yanteng32/CTF_FMRI.

Alzheimer's disease (AD) is a debilitating neurodegenerative disease. Early diagnosis of AD and its precursor, mild cognitive impairment (MCI), is crucial for timely intervention and management. Radiomics involves extracting quantitative features from medical images and analyzing them using advanced computational algorithms. These characteristics have the potential to serve as biomarkers for disease classification, treatment response prediction, and patient stratification. Of note, Magnetic resonance imaging (MRI) radiomics showed a promising result for diagnosing and classifying AD, and MCI from normal subjects. Thus, we aimed to systematically evaluate the diagnostic performance of the MRI radiomics for this task. A comprehensive search of the current literature was conducted using relevant keywords in PubMed/MEDLINE, Embase, Scopus, and Web of Science databases from inception to August 5, 2023. Original studies discussing the diagnostic performance of MRI radiomics for the classification of AD, MCI, and normal subjects were included. Method quality was evaluated with the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) and the Radiomics Quality Score (RQS) tools. We identified 13 studies that met the inclusion criteria, involving a total of 5448 participants. The overall quality of the included studies was moderate to high. The pooled sensitivity and specificity of MRI radiomics for differentiating AD from normal subjects were 0.92 (95% CI [0.85; 0.96]) and 0.91 (95% CI [0.85; 0.95]), respectively. The pooled sensitivity and specificity of MRI radiomics for differentiating MCI from normal subjects were 0.74 (95% CI [0.60; 0.85]) and 0.79 (95% CI [0.70; 0.86]), respectively. Also, the pooled sensitivity and specificity of MRI radiomics for differentiating AD from MCI were 0.73 (95% CI [0.64; 0.80]) and 0.79 (95% CI [0.64; 0.90]), respectively. MRI radiomics has promising diagnostic performance in differentiating AD, MCI, and normal subjects. It can potentially serve as a non-invasive and reliable tool for early diagnosis and classification of AD and MCI.

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.

Alzheimer's disease (AD) is the most general type of dementia, which concludes in memory-related problems. For researchers, the AD diagnosis at an early stage is a complex task. Alzheimer's disease is a chronic neurodegenerative disease. The risk of further degeneration will be decreased by early diagnosis of the disease. Other than predicting the possible growth of the disease, the recent studies concentrate on disease classification state. In this research, a Siamese-LSTM (long short-term memory) model is proposed for the enhanced multiple stages of Alzheimer's disease classification. The BoG bag of visual word method is utilized in order to improve the potency of texture based features (i.e., GLCM). The dataset used in this research work is ADNI. For multi-class classification, the samples are classified into three classes such as normal, AD, and MCI. The performance of the Alzheimer's disease stage is calculated in terms of accuracy, precision, and recall. Moreover, in addition, the sensitivity and specificity are evaluated.

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.

Background/Objectives: Translating brain volumetric biomarkers to individual-level Alzheimer's disease (AD) diagnosis remains challenging due to difficulty interpreting raw volumes without longitudinal monitoring or matched controls. We tested a classification model using population-referenced volumetric percentiles to distinguish AD from cognitively normal (CN) subjects and evaluated its generalization across independent cohorts. Methods: Brain volumes from 95 regions were extracted using an automated segmentation pipeline and converted to age and sex adjusted percentiles using a reference population (N = 1833). A logistic regression classifier was trained on ADNI subjects (N = 873; AD = 183, CN = 690) split into training (60%), validation (20%), and test (20%) sets. The model was evaluated on two independent validation datasets: the held-out ADNI validation set and an external Korean cohort (N = 72; AD = 36, CN = 36) acquired with different scanner protocols and demographic characteristics. Results: The model achieved excellent discrimination across all evaluation sets: ADNI validation (AUC = 0.963, accuracy = 90.3%), ADNI test (AUC = 0.960, accuracy = 89.7%), and Korean external validation (AUC = 0.981, accuracy = 87.5%). The minimal validation gap (0.018) demonstrated robust generalization. Positive coefficients for ventricular regions reflected AD-associated atrophy patterns, while negative coefficients for medial temporal structures indicated their contribution within multivariate patterns distinguishing AD from normal aging. Conclusions: Population-referenced brain volumetric percentiles enable accurate AD classification with robust generalization across populations and scanner protocols. By contextualizing individual brain structure relative to normative populations while accounting for age and sex, this approach demonstrates potential for clinical translation as an accessible neuroimaging-based diagnostic tool.

Alzheimer's Disease (AD) classification is very helpful for timely treatment and intervention in its early stage so as to reduce the incidence rate of AD and delay AD progression in elderly population. In this paper, a feature extraction method utilizing Euclidean distance map was proposed. Then the features for AD classification were selected in accordance with the feature importance score calculated by LightGBM (LGB). At last, the LGB classifier was applied to classifying AD patient utilizing selected features. The proposed AD classification method was evaluated on real dataset. The experimental results show that the proposed feature extraction and selection methods are effective, and the LGB classifier using selected features is superior to other conventional classifiers.

Alzheimer’s Disease (AD) is the most common single cause of dementia in our ageing society. Traditionally thought of as an untreatable degenerative condition, recent advances in drug therapy have challenged this view. The disease is characterised by an insidious decline in cognitive and non-cognitive function. Classically, short and long-term memory is impaired while language skills, concentration and attention are often affected. This results in impaired ability to learn and retain new skills as well as the loss of existing ones. Non-cognitive function is the global term used to describe problems such as depression, agitation, personality changes, delusions and hallucinations. These factors have a significant impact on patient behaviour and a very real impact on the quality of life for both patients and caregivers. Diagnosis of AD is clinically based, and using the NINCDS-ADRDA criteria (Table 1) [1], a diagnosis of probable or possible AD can be made. Definitive diagnosis relies on pathological confirmation, which in the majority of cases is rarely completed. With the development of AD specific treatments, definition of AD from other types of dementia is very important. Table 1 NINCDS-ADRDA Criteria for clinical diagnosis of Alzheimer’s disease. Pathogenesis The pathogenesis of AD has not yet been elucidated. It is widely accepted that a combination of genetic susceptibility factors and environmental triggers are responsible for late onset sporadic AD, the most common form of the disease. An understanding of the disease mechanism remains elusive, and is the key to developing a disease modifying agent. Currently, it is proposed that beta amyloid protein, abnormal tau protein or possibly both play key factors in the development of disease. It has been widely postulated that oxidative damage and a slow inflammatory process are two possible mechanisms involved. As yet, no product with proven disease modifying properties is available, and current treatments offer symptomatic benefit only. The development of acetylcholinesterase (AChe) inhibitor drugs has followed the finding that cholinergic pathways in the cerebral cortex and basal forebrain are compromised in AD [2] and the resultant cholinergic deficit contributes to the cognitive impairment of these patients [3]. Although many believe this ‘cholinergic hypothesis’ to be important, others feel it represents a less significant component of the disease process [4]. Many other neurotransmitters are affected in AD, and the relative importance of each in relation to clinical findings has not been fully elucidated. Initial work focused on the use of acetylcholine precursors, using a similar rationale to dopamine therapy in Parkinson’s disease. A series of small trials using precursors such as choline and phosphatidylcholine showed no reliable improvement in cognitive function, with only 10 out of 43 trials reporting any positive effect [5]. There has been renewed interest in muscarinic agonists drugs, which when first introduced, had major problems with adverse cholinergic effects. Better understanding of the molecular pathology of muscarinic receptors and their subtypes has led to the development of more specific agonists. Drugs such as xanomeline, milameline, and civimeline have reached clinical trials, and the improvements seen in cognitive function are reviewed by Avery et al. [6]. There are also claims that these drugs have disease modifying properties, with effects on APP processing and tau phosphorylation. Muscarinic agonists remain in trial, but have yet to fulfil their potential in AD treatment. The only group of drugs currently licensed for AD treatment is the AChe inhibitors, which act through inhibition of the enzyme acetylcholinesterase (AChe), responsible for the breakdown of ACh in the neural synapse. A meta-analysis of the early AChe inhibitor treatments was encouraging [7] and these proceeded to larger placebo controlled double-blind trials.