Diagnosis of Alzheimer’s disease via an attention-based multi-scale convolutional neural network

Liu, Zhenbing; Lu, Haoxiang; Pan, Xipeng; Xu, Mingchang; Lan, Rushi; Luo, Xiangfeng

doi:10.1016/j.knosys.2021.107942

Cited by 50 publications

(30 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Liu et al [15] developed a multi-scale CNN with a channel attention mechanism for enhanced AD diagnosis. They used preprocessing and segmentation to obtain the WM and GM datasets and model training.…”

Section: Related Studiesmentioning

confidence: 99%

“…The manual diagnosis of AD is based on recent developments in advanced neuroimaging techniques (such as magnetic resonance imaging (MRI), Computed Tomography (CT), Positron Emission Tomography (PET)), manual feature extraction, and clinical evaluations. MRI scans are the commonly utilized method that achieved unprecedented progress due to their non-invasive nature, high resolution, nonionizing radiation, and multidirectional imaging [15]. However, the brain structure is very complicated, and the imaging modalities involved are multi-modal and the curse of dimensionality, making the manual diagnosis time-consuming, error-prone, and tedious.…”

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

Section: Related Studiesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

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

show abstract

“…According to the characteristics of the image, a specific area of the image is selected, and through rapid eye movement scanning, the area is moved to the fovea area with high resolution and attention to this area is realized for more detailed observation and analyze. The selective attention mechanism is a key technology for humans to select specific regions of interest from a large amount of information input from the outside world [ 12 – 14 ].…”

Section: Related Workmentioning

confidence: 99%

Research on Image Segmentation Algorithm Based on Multimodal Hierarchical Attention Mechanism and Genetic Neural Network

Wang

Ma²

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Multimodal tasks based on attention mechanism and language face numerous problems. Based on multimodal hierarchical attention mechanism and genetic neural network, this paper studies the application of image segmentation algorithm in data completion and 3D scene reconstruction. The algorithm refers to the process of concentrating attention that humans subjectively pay attention to and calculates the difference between each pixel in the genetic neural network test image in the color space and the average value of the target image, which solves the problem of static feature maps and dynamic feature maps of image sequences. In addition, in view of the problem that the number of attention enhancement feature extraction modules is too large and the parameters are too large, the recursive mechanism is used as the feature extraction branch, and new model parameters are not added when the network depth is increased. The simulation results show that the accuracy of the improved image saliency detection algorithm based on the attention mechanism reaches 89.7%, and the difference between the average value of the single-point pixel and the target image is reduced to 0.132, which further promotes the practicability and reliability of the image segmentation model.

show abstract

“…Many research works have been carried out related to the ADMPIs [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17] processing and analysis. Some of the works are described as follows.…”

Section: Related Workmentioning

confidence: 99%

“…Alzheimer's disease (AD) [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17] is considered as a major concern throughout the globe in this era.…”

Section: Introductionmentioning

confidence: 99%

Alzheimer Disease MRI Preprocessed Images: A Machine Intelligent Based Approach for Classification and Analysis

Jena

K.²

2022

IJCSBE

View full text Add to dashboard Cite

Purpose: Alzheimer’s disease (AD) is considered as one of the most dangerous diseases in the present scenario. It is a brain disorder disease which leads to the destruction of the thinking skills and memory of human beings. It is very much essential for the early classification of AD magnetic resonance imaging (MRI) preprocessed images (ADMPIs) into several categories such as Mild_Demented (MID), Moderate_Demented (MOD), Non_Demented (ND), Very_Mild_Demented (VMD), etc. so that preventive measures can be taken at the earliest. Approach: In this work, a machine intelligent (MI) based approach is proposed for the classification of ADMPIs into the MID, MOD, ND and VMD types. This approach is focused on machine learning (ML) based methods such as Logistic Regression (LRG), Support Vector Machine (SVMN), Random Forest (RFS), Neural Network (NNT), Decision Tree (DTR), AdaBoost (ADB), Naïve Bayes (NBY), K-Nearest Neighbor (KNNH) and Stochastic Gradient Descent (SGDC) to carry out such classification. Result: The ML based methods have been implemented using Python based Orange 3.26.0. In this work, 1564 ADMPIs having 500, 64, 500 and 500 numbers of each type such as MID, MOD, ND and VMD respectively are taken from the Kaggle source. The performance of all the methods is assessed using the performance parameters such as classification accuracy (CA), F1, Precision (PR) and Recall (RC). From the results, it is found that the NNT method is capable of providing better classification results in terms of CA, F1, PR and RC as compared to other ML based methods such as SVMN, RFS, NNT, DTR, ADB, NBY, KNNH and SGD. Originality: In this work, a MI based approach is proposed to carry out the classification of ADMPIs into several types such as MID, MOD, ND and VMD types. The NNT method performs better in terms of CA, F1, PR and RC as compared to LRG, SVMN, RFS, DTR, ADB, NBY, KNNH and SGDC methods. Paper Type: Conceptual Research.

show abstract

Diagnosis of Alzheimer’s disease via an attention-based multi-scale convolutional neural network

Cited by 50 publications

References 51 publications

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

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

Research on Image Segmentation Algorithm Based on Multimodal Hierarchical Attention Mechanism and Genetic Neural Network

Alzheimer Disease MRI Preprocessed Images: A Machine Intelligent Based Approach for Classification and Analysis

Contact Info

Product

Resources

About