Multi-modal data Alzheimer’s disease detection based on 3D convolution

Kong, Zhaokai; Zhang, Mengyi; Zhu, Wei; Yang, Yi; Wang, Tian; Zhang, Baochang

doi:10.1016/j.bspc.2022.103565

Cited by 53 publications

(19 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…New techniques have been developed to tackle the difficulties presented by high‐resolution 3D data in sMRI. One such innovation is the sMRI, PatchNet, which uses explainable patch localization and selection to diagnose AD 26 . Additionally, vision transformer models have shown promise in enhancing performance for larger‐scale population neuroimaging diagnostic and prognostic tasks 27 .…”

Section: Related Studymentioning

confidence: 99%

Structural biomarker‐based Alzheimer's disease detection via ensemble learning techniques

Shukla,

Tiwari,

Tiwari

2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Alzheimer's disease (AD) is a degenerative neurological disorder with incurable characteristics. To identify the substantial solution, we used a structural biomarker (structural magnetic resonance imaging) to see the neurostructural changes in the different regions of the brain of AD, mild cognitive impairment, and cognitive normal subjects. In this study, we detected the AD and their subtypes by using the traditional machine learning and ensemble learning models. It is also identified the relative impact score of various cortical and subcortical regions of AD and their subtypes. Experimental study contains two levels of classification: binary and multiclass. The Ensemble_LR_SVM model in binary classification has 99% accuracy in detection. Random forest model in the multiclass has 82% of accuracy. In the cortical‐subcortical analysis, the right hemisphere's parahippocampal and entorhinal regions were discovered to be the most influential. Similarly, the inferior temporal and isthmus cingulate regions had a significant influence in the left hemisphere.

show abstract

Section: Related Studymentioning

confidence: 99%

Structural biomarker‐based Alzheimer's disease detection via ensemble learning techniques

Shukla,

Tiwari,

Tiwari

2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

show abstract

“…The ResNet-18 network achieved the highest classification accuracy of 98.63%. Kong et al [3] developed a deep learning-based strategy that involved a novel MRI and PET image fusion and 3D CNN for AD multi-classification methods. The ADNI dataset of 740 3D images was used.…”

Section: Related Studiesmentioning

confidence: 99%

“…AD is an irreversible, progressive, and ultimately fatal brain degenerative disorder that affects middle-aged and older people. When the disease is discovered, most patients have already progressed to an advanced stage [3]. As a result, AD gradually deteriorates memory and thinking abilities and the ability to carry out even the most basic duties of daily life by destroying the brain cells.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A3C-TL-GTO: Alzheimer Automatic Accurate Classification Using Transfer Learning and Artificial Gorilla Troops Optimizer

Baghdadi

Malki

Balaha

et al. 2022

Sensors

View full text Add to dashboard Cite

Alzheimer’s disease (AD) is a chronic disease that affects the elderly. There are many different types of dementia, but Alzheimer’s disease is one of the leading causes of death. AD is a chronic brain disorder that leads to problems with language, disorientation, mood swings, bodily functions, memory loss, cognitive decline, mood or personality changes, and ultimately death due to dementia. Unfortunately, no cure has yet been developed for it, and it has no known causes. Clinically, imaging tools can aid in the diagnosis, and deep learning has recently emerged as an important component of these tools. Deep learning requires little or no image preprocessing and can infer an optimal data representation from raw images without prior feature selection. As a result, they produce a more objective and less biased process. The performance of a convolutional neural network (CNN) is primarily affected by the hyperparameters chosen and the dataset used. A deep learning model for classifying Alzheimer’s patients has been developed using transfer learning and optimized by Gorilla Troops for early diagnosis. This study proposes the A3C-TL-GTO framework for MRI image classification and AD detection. The A3C-TL-GTO is an empirical quantitative framework for accurate and automatic AD classification, developed and evaluated with the Alzheimer’s Dataset (four classes of images) and the Alzheimer’s Disease Neuroimaging Initiative (ADNI). The proposed framework reduces the bias and variability of preprocessing steps and hyperparameters optimization to the classifier model and dataset used. Our strategy, evaluated on MRIs, is easily adaptable to other imaging methods. According to our findings, the proposed framework was an excellent instrument for this task, with a significant potential advantage for patient care. The ADNI dataset, an online dataset on Alzheimer’s disease, was used to obtain magnetic resonance imaging (MR) brain images. The experimental results demonstrate that the proposed framework achieves 96.65% accuracy for the Alzheimer’s Dataset and 96.25% accuracy for the ADNI dataset. Moreover, a better performance in terms of accuracy is demonstrated over other state-of-the-art approaches.

show abstract

“…3D CNN has been applied to the staging of the AD spectrum. Kong and his colleagues 15 initially trained a 3D sparse autoencoder to learn the filters on randomly chosen 3D patches of the sMRI and then used those pretrained kernels as the first convolution layer of a 3D CNN. Li et al 16 proposed a hybrid convolutional and recurrent neural network by combining 3D DenseNets and (bidirectional gated recurrent unit) BGRU for AD diagnosis based on hippocampus volumes.…”

Section: Introductionmentioning

confidence: 99%

Three-round learning strategy based on 3D deep convolutional GANs for Alzheimer’s disease staging

Kang

Lin

Sun

et al. 2023

Sci Rep

View full text Add to dashboard Cite

Accurately diagnosing of Alzheimer's disease (AD) and its early stages is critical for prompt treatment or potential intervention to delay the the disease’s progression. Convolutional neural networks (CNNs) models have shown promising results in structural MRI (sMRI)-based diagnosis, but their performance, particularly for 3D models, is constrained by the lack of labeled training samples. To address the overfitting problem brought on by the insufficient training sample size, we propose a three-round learning strategy that combines transfer learning with generative adversarial learning. In the first round, a 3D Deep Convolutional Generative Adversarial Networks (DCGAN) model was trained with all available sMRI data to learn the common feature of sMRI through unsupervised generative adversarial learning. The second round involved transferring and fine-tuning, and the pre-trained discriminator (D) of the DCGAN learned more specific features for the classification task between AD and cognitively normal (CN). In the final round, the weights learned in the AD versus CN classification task were transferred to the MCI diagnosis. By highlighting brain regions with high prediction weights using 3D Grad-CAM, we further enhanced the model's interpretability. The proposed model achieved accuracies of 92.8%, 78.1%, and 76.4% in the classifications of AD versus CN, AD versus MCI, and MCI versus CN, respectively. The experimental results show that our proposed model avoids overfitting brought on by a paucity of sMRI data and enables the early detection of AD.

show abstract

Multi-modal data Alzheimer’s disease detection based on 3D convolution

Cited by 53 publications

References 26 publications

Structural biomarker‐based Alzheimer's disease detection via ensemble learning techniques

Structural biomarker‐based Alzheimer's disease detection via ensemble learning techniques

A3C-TL-GTO: Alzheimer Automatic Accurate Classification Using Transfer Learning and Artificial Gorilla Troops Optimizer

Three-round learning strategy based on 3D deep convolutional GANs for Alzheimer’s disease staging

Contact Info

Product

Resources

About