Search

Introduction: Alzheimer's disease (AD) is a chronic form of the neurodegenerative disease marked by atrophy in different brain regions. A region-wise analysis is essential for performing AD detection, as each brain region has different functionalities depending on its location. This work aims to investigate supervoxel based volumetric features in place of traditional voxel-based features from the vital brain regions. Methods: In this work, the whole brain structural magnetic resonance imaging (MRI) is segmented into 116 regions using atlas-based segmentation. Important atrophic regions are used for further analysis based on a region ranking procedure from these segmented regions. The focus of this study is to perform supervoxel based partitioning for attaining features prominent for AD detection. Volumetric features are extracted from supervoxels belonging to the selected regions. An optimal feature set is obtained by using the support vector machine recursive elimination (SVM-RFE) method, and classification is performed using SVM. Results: ADNI dataset is used for experimentation. Results are obtained by iteratively fusing the features extracted from vital brain regions. The highest classification accuracy of 90.11%, the sensitivity of 86.11%, and the specificity of 93.4% are obtained by fusing features extracted from hippocampus and amygdala regions. Discussion: The highest classification accuracy reported in this work for AD detection is obtained by fusing features of the four most important regions, i.e., hippocampus and amygdala, in both left and right hemispheres. These regions are also known to affect the consolidation of memory and decision-making in medical science. Experimental results evaluated on the standard dataset depict that the proposed method performs better than the traditional as well as state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) is a prevalent, irreversible, chronic, and degenerative disorder whose diagnosis at the prodromal stage is critical. Mostly, single modality data, such as magnetic resonance imaging (MRI) or positron emission tomography (PET), are used to make predictions in AD studies. However, the metabolic and structural data fusion can provide a holistic view of AD-staging analysis. To achieve this objective, a novel multimodal fusion-based method is proposed in this article. An optimal fusion of MRI and PET is achieved by harnessing demon algorithm and discrete wavelet transform. Finally, the fused image features are extracted using ResNet-50, and these features are classified using robust energy least square twin support vector machine classifier. Experiments on the AD neuroimaging initiative dataset show descent accuracy of 97%, 94%, and 97.5% for cognitive normal (CN) versus AD, CN versus mild cognitive impairment (MCI), and AD versus MCI, respectively. The proposed model will be beneficial for health professionals in accurately diagnosing AD at an early stage. [ABSTRACT FROM AUTHOR]

Neurodegenerative disorders like Alzheimer's disease (AD) are irreversible and show atrophies in the area of the cerebral cortex of brain. AD leads to loss of memory and other cognitive impairments. The AD subjects are evaluated based on magnetic resonance imaging scans. The data may have the problem of class imbalance, noise and outliers which is a great challenge for classification. Support vector machines and twin support vector machine-based classifiers may not effectively deal with these problems as both these models assume that all the samples are equally important for the separating hyperplane. To overcome these issues, we propose intuitionistic fuzzy least square twin support vector machine for class imbalance problems (IFLSTSVM) and class specific-IFLSTSVM (CS-IFLSTSVM). To minimize the effects of class imbalance, the samples are appropriately weighted to minimize their effect on the optimal hyperplane. Moreover, we use intuitionistic fuzzy scores to overcome the issues of noise and outliers. Intuitionistic fuzzy score values generate appropriate weights by considering both the distance of the samples from the class centroid as well as the heterogeneity of the samples. The proposed models IFLSTSVM and CS-IFLSTSVM are efficient as they need to solve a system of linear equations. In Alzheimer's disease diagnosis, the proposed IFLSTSVM and CS-IFLSTSVM models showed better performance in MCI_vs_AD and CN_vs_MCI cases, respectively. Moreover, the proposed models showed better performance in the diagnosis of breast cancer classification. The statistical analysis carried out over KEEL and UCI data leads to the superiority of the proposed models. The source code of the proposed model is available at https://github.com/mtanveer1/Diagnosis-of-Alzheimer-s-disease-via-Intuitionistic-fuzzy-least-squares-twin-SVM. • To handle the CIL issue, training data are assigned weights based on imbalance ratio. • The regularization term improves the generalization performance of proposed models. • The proposed models solve system of linear equations. Hence, it is very efficient. • The weights reduce the impact of noise/outliers in classifying imbalanced data. • The results on biomedical and UCI data show the superiority of proposed models. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) is the most pervasive form of dementia, resulting in severe psychosocial effects such as affecting personality, reasoning, emotions, and memory. Several neuroimaging techniques are available to correctly identify the structural changes in the brain, out of which the most popular is structural T-1 weighted Magnetic Resonance Imaging (MRI). From 3D MRI, sagittal plane slices provide more clear information related to the hippocampus, amygdala, corpus callosum, and several vital regions of the brain, which defines the extent of degeneration of the AD. Although diverse analysis of machine learning (ML) and deep learning (DL) based algorithm is already proposed for diagnosis of AD, still there is scope of research for early prediction so that treatment can be started either by medication or by improving the lifestyle. This paper proposed a DL model for all level feature extraction and fuzzy hyperplane based least square twin support vector machine (FLS-TWSVM) for the classification of the extracted features for early diagnosis of AD (FDN-ADNet) using extracted sagittal plane slices from 3D MRI images. Model is trained over the online available ADNI dataset and triangular fuzzy function is applied for the construction of hyperplane for classification. The proposed model attains the highest accuracy of 97.15%, 97.29% and 95% for CN vs AD, CN vs MCI and AD vs MCI classification, respectively when compared with the several state of the art networks. • Alzheimer's disease (AD) is the most prevailing irreversible mental neural disorder which leads to memory deterioration. • Sagittal plane of MRI provides more visual features of the mid brain regions. • Pre-processing is done on MRI images for sagittal plane extraction, image registration and key slice extraction. • ResNet101 deep learning network (DLN) is implemented to extract the features from sagittal plane slices. • Fuzzy based classifier is used for classification of the features extracted from DLN for classification of CN vs AD, Cn vs MCI, AD vs MCI. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) is a degenerative neurological ailment that begins with memory loss and ultimately leads to a total loss of mental capacity. Researchers are interested in using magnetic resonance imaging (MRI) and positron emission tomography (PET) to find people with mild cognitive impairment (MCI), which is a stage before Alzheimer's disease (AD). Significant hippocampal loss and temporal lobe atrophy characterize the transition from MCI to AD, which can be visualized using T1-W structural MRI. PET visualizes brain glucose metabolism, which indicates neuronal activity, making it a viable neuroimaging method for AD diagnosis. The extraction and fusion of structural and metabolite information about brain alterations contained in multimodal data is crucial for achieving an appropriate classification result. Therefore, in this work a new end-to-end coupled-GAN (CGAN) architecture is introduced. The proposed CGANC network consists of two sub-models: a CGAN for extraction of fused features from multimodal data, and a CNN classifier to classify these features. The proposed CGAN model is trained to encode MRI and PET images into a shared latent space. The fused features are extracted from this shared latent space and then are classified according to particular stage of AD. In order to test the effectiveness of the suggested approach, experiments are done on the publicly available ADNI dataset and compared with state-of-the-art methods. The proposed method's source code will be made freely available at https://github.com/ChandrajitChoudhury/CGAN-AD. • MRI and PET images are fused to utilize structural and metabolic features for AD diagnosis. • An adversarial learning-based method is proposed to extract fused features. • A coupled adversarial subnetwork and a classification subnetwork have been designed. • Experiments are done on the publicly available ADNI dataset. [ABSTRACT FROM AUTHOR]

Deep kernel learning has been well explored for multi-class classification tasks; however, relatively less work is done for one-class classification (OCC). OCC needs samples from only one class to train the model. Most recently, kernel regularized least squares (KRL) method-based deep architecture is developed for the OCC task. This paper introduces a novel extension of this method by embedding minimum variance information within this architecture. This embedding improves the generalization capability of the classifier by reducing the intra-class variance. In contrast to traditional deep learning methods, this method can effectively work with small-size datasets. We conduct a comprehensive set of experiments on 18 benchmark datasets (13 biomedical and 5 other datasets) to demonstrate the performance of the proposed classifier. We compare the results with 16 state-of-the-art one-class classifiers. Further, we also test our method for 2 real-world biomedical datasets viz.; detection of Alzheimer's disease from structural magnetic resonance imaging data and detection of breast cancer from histopathological images. Proposed method exhibits more than 5% F 1 score compared to existing state-of-the-art methods for various biomedical benchmark datasets. This makes it viable for application in biomedical fields where relatively less amount of data is available. [ABSTRACT FROM AUTHOR]