Deep learning and multimodal feature fusion for the aided diagnosis of Alzheimer's disease

Jia, Hongfei; Lao, Huan

doi:10.1007/s00521-022-07501-0

Cited by 20 publications

(4 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First , individual differences between different subjects are not fully considered in our scheme, which is an important issue for FBN analysis (Folville et al, 2020 ; Schabdach et al, 2022 ) and an important direction for our future improvements. Besides , since multi-modal data is taking an increasingly important place in brain analysis (Jia and Lao, 2022 ; Zhao et al, 2022 ), the performance of employing one-modal data is limited. Compared with the widely used multi-modal data model (Yu et al, 2021 ), how to adapt the scheme to the multi-modal data type is a key point in our future work.…”

Section: Discussionmentioning

confidence: 99%

Spatial-temporal data-augmentation-based functional brain network analysis for brain disorders identification

et al. 2023

View full text Add to dashboard Cite

IntroductionDue to the lack of devices and the difficulty of gathering patients, the small sample size is one of the most challenging problems in functional brain network (FBN) analysis. Previous studies have attempted to solve this problem of sample limitation through data augmentation methods, such as sample transformation and noise addition. However, these methods ignore the unique spatial-temporal information of functional magnetic resonance imaging (fMRI) data, which is essential for FBN analysis.MethodsTo address this issue, we propose a spatial-temporal data-augmentation-based classification (STDAC) scheme that can fuse the spatial-temporal information, increase the samples, while improving the classification performance. Firstly, we propose a spatial augmentation module utilizing the spatial prior knowledge, which was ignored by previous augmentation methods. Secondly, we design a temporal augmentation module by random discontinuous sampling period, which can generate more samples than former approaches. Finally, a tensor fusion method is used to combine the features from the above two modules, which can make efficient use of spatial-temporal information of fMRI simultaneously. Besides, we apply our scheme to different types of classifiers to verify the generalization performance. To evaluate the effectiveness of our proposed scheme, we conduct extensive experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset and REST-meta-MDD Project (MDD) dataset.ResultsExperimental results show that the proposed scheme achieves superior classification accuracy (ADNI: 82.942%, MDD: 63.406%) and feature interpretation on the benchmark datasets.DiscussionThe proposed STDAC scheme, utilizing both spatial and temporal information, can generate more diverse samples than former augmentation methods for brain disorder classification and analysis.

show abstract

Section: Discussionmentioning

confidence: 99%

Spatial-temporal data-augmentation-based functional brain network analysis for brain disorders identification

et al. 2023

View full text Add to dashboard Cite

show abstract

“…The first is to model the 4D information by applying the model directly to the time series of fMRI volumes in order to avoid information loss associated with applying CNNs to FC networks or mReHo transformation images, and the second is to overcome the overfitting problem that occurs with traditional CNNs that was replaced with CapsNet. From the comparison results, it has been observed that the proposed model and the studies introduced in [7] and [22], which model the 4D information from the fMRI volume sequence directly, were able to achieve better accuracies than the models introduced in [10], [12], [14], [15] (which apply CNNs to FC networks), and [17] (that apply CNNs to mReHO images). Moreover, despite CapsNet being designed to achieve better generalization than traditional CNNs, the method which applies CNNs with LSTM directly to fMRI volumes [22] still achieves better accuracy than our model.…”

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 96%

“…They applied CNNs to dynamic FC (dFC) networks for extracting the temporal features based on which an LSTM was used to model the sequential information of the dFC networks. To perform the spatiotemporal analysis avoiding the 4D information complexity and model the 4D features at the same time, Jia et al used mReHo transformation [16] that represents the whole time series of volumes as a single volume [17]. In the study, a 3D-PCANet was trained on the mReHo volumes as the classifier model.…”

Section: Related Work and Contributionmentioning

confidence: 99%

Use of 3D-CAPSNET and RNN models for 4D fMRI-based Alzheimer’s Disease Pre-detection

İsmail,

Menekşe Dalveren

2024

Turkish Journal of Science and Technology

View full text Add to dashboard Cite

An early prediction of Alzheimer’s disease (AD) progression can help slow down cognitive decline more effectively. Several studies have been devoted to applying different methods based on convolutional neural networks (CNNs) for automated AD diagnosis using resting-state functional magnetic resonance imaging (rs-fMRI). The methods introduced in these studies encounter two major challenges. First, fMRI datasets suffer from being of small size resulting in overfitting. Second, the 4D information of fMRI sessions needs to be efficiently modeled. Some of the studies applied their deep learning methods to functional connectivity matrices generated from fMRI data to model the 4D information, or to fMRI data as separate 2D slices or 3D volumes. However, this results in information loss in both types of methods. In this study, a new model based on Capsule network (CapsNet) and recurrent neural network (RNN) is proposed to model the spatiotemporal (4D) information of fMRI data for AD diagnosis. Experiments were conducted to evaluate the efficiency of the proposed model. According to the results, it has been observed that the proposed model could achieve 94.5% and 61.8% accuracy for the AD versus normal control (NC) and late mild cognitive impairment (lMCI) versus early mild cognitive impairment (eMCI) classification tasks, respectively.

show abstract

“…The number of deaths from Alzheimer's disease in 2020 increased by 15,925 compared to the 5 years before 2023, and 44,729 more deaths were recorded for all dementias, including Alzheimer's disease (Chua, 2023 ). Traditional machine learning (ML) techniques such as pre-processing (Wen et al, 2020 ), feature extraction (Rathore et al, 2017 ), feature selection (Balaji et al, 2023 ), feature fusion (Jia and Lao, 2022 ), and classification (Tanveer et al, 2020 ) have been employed by researchers as a four-step channel in the past few years. Classification is the bottommost step in which each object accredits a label, in either a supervised or unsupervised ML technique (Bondi et al, 2017 ).…”

Section: Introductionmentioning

confidence: 99%

Prediction of Alzheimer's disease stages based on ResNet-Self-attention architecture with Bayesian optimization and best features selection

Yaqoob,

Khan,

Masood

et al. 2024

Front. Comput. Neurosci.

View full text Add to dashboard Cite

Alzheimer's disease (AD) is a neurodegenerative illness that impairs cognition, function, and behavior by causing irreversible damage to multiple brain areas, including the hippocampus. The suffering of the patients and their family members will be lessened with an early diagnosis of AD. The automatic diagnosis technique is widely required due to the shortage of medical experts and eases the burden of medical staff. The automatic artificial intelligence (AI)-based computerized method can help experts achieve better diagnosis accuracy and precision rates. This study proposes a new automated framework for AD stage prediction based on the ResNet-Self architecture and Fuzzy Entropy-controlled Path-Finding Algorithm (FEcPFA). A data augmentation technique has been utilized to resolve the dataset imbalance issue. In the next step, we proposed a new deep-learning model based on the self-attention module. A ResNet-50 architecture is modified and connected with a self-attention block for important information extraction. The hyperparameters were optimized using Bayesian optimization (BO) and then utilized to train the model, which was subsequently employed for feature extraction. The self-attention extracted features were optimized using the proposed FEcPFA. The best features were selected using FEcPFA and passed to the machine learning classifiers for the final classification. The experimental process utilized a publicly available MRI dataset and achieved an improved accuracy of 99.9%. The results were compared with state-of-the-art (SOTA) techniques, demonstrating the improvement of the proposed framework in terms of accuracy and time efficiency.

show abstract

Deep learning and multimodal feature fusion for the aided diagnosis of Alzheimer's disease

Cited by 20 publications

References 39 publications

Spatial-temporal data-augmentation-based functional brain network analysis for brain disorders identification

Spatial-temporal data-augmentation-based functional brain network analysis for brain disorders identification

Use of 3D-CAPSNET and RNN models for 4D fMRI-based Alzheimer’s Disease Pre-detection

Prediction of Alzheimer's disease stages based on ResNet-Self-attention architecture with Bayesian optimization and best features selection

Contact Info

Product

Resources

About