Multi-Perspective Feature Extraction and Fusion Based on Deep Latent Space for Diagnosis of Alzheimer’s Diseases

Gao, Le; Hu, Zhongyi; Li, Rui; Lu, Xingjin; Li, Zuoyong; Zhang, Xiabin; Xu, Shiwei

doi:10.3390/brainsci12101348

Cited by 6 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The former was based on dynamically intercepting signals through windows of specific length and then performing functional connectivity analysis within each dynamic window. In recent years, a number of studies have been devoted to the extraction of high-level features from the dFC to achieve effective filtering of invalid information in the dFC and extraction of dynamic change features associated with the disease, contributing to exploring abnormal brain function networks and improving the classification accuracy of AD ( Sendi et al, 2021 ; Gao et al, 2022 ; Matsui and Yamashita, 2022 ; Qiao et al, 2022 ; Ghanbari et al, 2023 ; Penalba-Sánchez et al, 2023 ). The principle of CAP was to extract co-activation patterns in certain specific peak points of the BOLD signal time series using a clustering algorithm.…”

Section: Introductionmentioning

confidence: 99%

Triple-network analysis of Alzheimer’s disease based on the energy landscape

Li¹,

An²,

Zhou³

et al. 2023

Front. Neurosci.

View full text Add to dashboard Cite

IntroductionResearch on the brain activity during resting state has found that brain activation is centered around three networks, including the default mode network (DMN), the salient network (SN), and the central executive network (CEN), and switches between multiple modes. As a common disease in the elderly, Alzheimer’s disease (AD) affects the state transitions of functional networks in the resting state.MethodsEnergy landscape, as a new method, can intuitively and quickly grasp the statistical distribution of system states and information related to state transition mechanisms. Therefore, this study mainly uses the energy landscape method to study the changes of the triple-network brain dynamics in AD patients in the resting state.ResultsAD brain activity patterns are in an abnormal state, and the dynamics of patients with AD tend to be unstable, with an unusually high flexibility in switching between states. Also , the subjects’ dynamic features are correlated with clinical index.DiscussionThe atypical balance of large-scale brain systems in patients with AD is associated with abnormally active brain dynamics. Our study are helpful for further understanding the intrinsic dynamic characteristics and pathological mechanism of the resting-state brain in AD patients.

show abstract

Section: Introductionmentioning

confidence: 99%

Triple-network analysis of Alzheimer’s disease based on the energy landscape

Li¹,

An²,

Zhou³

et al. 2023

Front. Neurosci.

View full text Add to dashboard Cite

show abstract

“…Venugopalan et al ( 2021 ) used stacked de-noising auto-encoders to extract features from clinical and genetic data of AD, and used CNN for imaging data, then found that hippocampus, amygdala brain areas and the rey auditory verbal learning test (RAVLT) were significant changes in AD brain. Despite variable structures of CNNs, the deep features extracted from CNNs were used to explain their strong recognition power (Bi et al, 2020 ; Gao et al, 2022 ). One aim of this study was to find possible explanations of the learned features based on the connectivity matrices derived from PSI matrix of fMRI signals.…”

Section: Introductionmentioning

confidence: 99%

A deep learning framework for identifying Alzheimer's disease using fMRI-based brain network

Wang

Han

et al. 2023

Front. Neurosci.

View full text Add to dashboard Cite

BackgroundThe convolutional neural network (CNN) is a mainstream deep learning (DL) algorithm, and it has gained great fame in solving problems from clinical examination and diagnosis, such as Alzheimer's disease (AD). AD is a degenerative disease difficult to clinical diagnosis due to its unclear underlying pathological mechanism. Previous studies have primarily focused on investigating structural abnormalities in the brain's functional networks related to the AD or proposing different deep learning approaches for AD classification.ObjectiveThe aim of this study is to leverage the advantages of combining brain topological features extracted from functional network exploration and deep features extracted by the CNN. We establish a novel fMRI-based classification framework that utilizes Resting-state functional magnetic resonance imaging (rs-fMRI) with the phase synchronization index (PSI) and 2D-CNN to detect abnormal brain functional connectivity in AD.MethodsFirst, PSI was applied to construct the brain network by region of interest (ROI) signals obtained from data preprocessing stage, and eight topological features were extracted. Subsequently, the 2D-CNN was applied to the PSI matrix to explore the local and global patterns of the network connectivity by extracting eight deep features from the 2D-CNN convolutional layer.ResultsFinally, classification analysis was carried out on the combined PSI and 2D-CNN methods to recognize AD by using support vector machine (SVM) with 5-fold cross-validation strategy. It was found that the classification accuracy of combined method achieved 98.869%.ConclusionThese findings show that our framework can adaptively combine the best brain network features to explore network synchronization, functional connections, and characterize brain functional abnormalities, which could effectively detect AD anomalies by the extracted features that may provide new insights into exploring the underlying pathogenesis of AD.

show abstract

“…Vaithinathan et al designed a new magnetic resonance T1-weighted texture extraction technique for Alzheimer's disease classification for early AD diagnosis, starting from the texture of the brain, and achieved good classification results [21].In addition to this, the literature combines 2D images and 3D images to perform studies on the early diagnostic classification of AD [22,23].Bi X et al designed two deep learning methods for functional brain network classification. The convolutional learning method learns deep region connectivity features and the recursive learning method learns deep adjacency location features, in addition, an Extreme Learning Machine (ELM) enhancement structure was implemented to further improve the learning capability, and the results show that the proposed method provides satisfactory learning capability in AD detection applications [24].Kanghan et al used unsupervised learning based on convolutional autoencoders (CAE) to solve AD and NC classification tasks and supervised transfer learning to solve pMCI and sMCI classification tasks [25].In addition there is the application of autoencoders in combination with long and short-term memory networks to resting-state functional magnetic resonance imaging for the early diagnosis of Alzheimer's disease [26].…”

Section: Introductionmentioning

confidence: 99%

Application of CNN-SCN in early diagnosis of Alzheimer's disease

Tong

Huang

et al. 2022

Preprint

View full text Add to dashboard Cite

Background: The early and effective diagnosis of Alzheimer's disease(AD) and mild cognitive impairment(MCI) has received increasing attention in recent years.There are many machine learning and deep learning methods that are widely used in neural image analysis, among which structural magnetic resonance images play an important role in early diagnosis and intervention for patients with Alzheimer's disease and mild cognitive impairment as a data pattern, and many studies have constructed many network models based on structural magnetic resonance images, but most of the existing studies do not pay enough attention to the spatial structure information in structural magnetic resonance images, and do not go deeper to explore the potential connection between spatial slices and slices,thus making the network models less accurate, poor generalization ability and other problems.We believe that the lesion areas are highly correlated with other areas,and there is a certain relationship between spatial context slices.To explore deeper the connection between spatial context slices and solve the above problems,this research is designed to develop a new deep learning method to effectively detect or predict AD by digging into the deeper spatial contextual structural information. Results: In this paper, we design a spatial context network based on 3D convolutional neural network to learn the multi-level structural features of brain MRI images for AD classification.In addition, this paper designs a new deep learning framework to adapt spatial contextual association networks for early diagnosis of Alzheimer's disease. The experimental results show that the model has good stability, accuracy and generalization ability.Our experimental results showed an accuracy of 92.6% in the AD/CN group, 74.9% in the AD/MCI group, and 76.3% in the MCI/CN group, and several experiments showed that the model has excellent generalization ability. Conclusions: From the comparison of a large number of experimental results, it can be seen that the spatial contextual domain association network can effectively mine the potential information of spatial contextual structural features. The network has better results for the cases that need to go for spatial structure contextual association features.

show abstract

Multi-Perspective Feature Extraction and Fusion Based on Deep Latent Space for Diagnosis of Alzheimer’s Diseases

Cited by 6 publications

References 31 publications

Triple-network analysis of Alzheimer’s disease based on the energy landscape

Triple-network analysis of Alzheimer’s disease based on the energy landscape

A deep learning framework for identifying Alzheimer's disease using fMRI-based brain network

Application of CNN-SCN in early diagnosis of Alzheimer's disease

Contact Info

Product

Resources

About