|
|
Course Classification of Cognitive Disorders Based on Improved Attention Mechanism |
LI Mei-mei1,2,HU Chun-hai1,ZHOU Ying2,SONG Xin2 |
1. Institute of Electrical Engineering,Yanshan University, Qinhuangdao, Hebei 066004, China
2. Northeastern University at Qinhuangdao, Qinhuangdao, Hebei 066004, China |
|
|
Abstract Alzheimer's disease (AD) is a neurodegenerative disease with a slow onset process and continuous deterioration over time. With the trend of aging, the number of AD patients is increasing. Therefore, how to make an early accurate diagnosis and positive intervention is an urgent problem to be solved. In order to improve the efficiency of computer-aided diagnosis and promote the study of pathophysiological mechanism of diseases, an improved two-dimension dual-path fusion network based on the SE module is proposed. A reduction coefficient module is added to the network to increase the proportion of useful information in images. The weight function of the channel attention module is redesigned to increase the difference between feature maps, and the network is combined with the two-dimensional dual-path network to increase the network emphasis, so as to achieve better classification performance and prevent model overfitting. The ADNI dataset is used to classify AD, EMCI and NC. The experimental results show that the accuracy of the proposed model is improved by 5.59% and 8.11% compared with the VGG and two-dimensional dual-path fusion model, respectively, which verifies the feasibility of the proposed method.
|
Received: 26 September 2022
Published: 21 February 2023
|
|
|
|
|
|
Fan F J, Bai Y, Ji H F. Denoising Method of EEG Signal Based on EEMD-ICA[J]. Acta Metrologica Sinica, 2021, 42(3): 395-400.
|
[3] |
Liu S, Cai W, Wen L, et al. Multi-Channel neurodegenerative pattern analysis and its application in Alzheimers disease characterization[J]. Comput Med Imaging Graph, 2014, 38: 436-444.
|
[4] |
Sarraf S, Desouza D, Andersonea J, et al. DeepAD: Alzheimers Disease Classification via Deep Convolutional Neural Networks using MRI and fMRI[J]. bioRxiv, 2016: 070441.
|
[5] |
Yang L, Wang Y, Wu G, et al. Discriminant analysis of longitudinal cortical thickness changes in Alzheimers disease using dynamic and network features[J]. Neurobiol Aging, 2012, 33(2):427. e15-427. e30.
|
[7] |
张世辉, 闫晓蕊, 桑榆. 融合残差及通道注意力机制的单幅图像去雨方法[J]. 计量学报, 2021, 42(1): 20-28.
|
[8] |
李鹤喜, 李记花, 李威龙. 一种基于注意机制和卷积神经网络视觉模型[J]. 计量学报, 2021, 42(7): 840-845.
|
[9] |
周影.基于深度学习的阿尔兹海默病分类研究[D]. 沈阳:东北大学, 2022.
|
[12] |
Li M, Hu C, Liu Z, et al. MRI Segmentation of Brain Tissue and Course Classification in Alzheimers Disease[J]. Electronics, 2022, 11, 1288.
|
[2] |
Suk H I, Lee S W, Shen D. Latent feature representation with stacked auto-encoder for AD/MCI diagnosis[J]. Brain Structure and Function, 2015, 220: 841-859.
|
[10] |
Hinrichs C, Singh V, Xu G, et al. Predictive markers for AD in a multi-modality framework: An analysis of MCIprogression in the ADNI population[J]. Neuroimage 2011, 55:574-589.
|
[13] |
Atif M. Early Diagnosis of Alzheimers Disease through Transfer Learning[D]. Xian: Xidian University, 2021.
|
[15] |
Forouzannezhad P, Abbaspour A, Li C F, et al. A deep neural network approach for early diagnosis of mild cognitive impairment using multiple features [C]//AMLA. ICMLA2018. America, 2018, 33:1341-1346.
|
[17] |
Kim J, Lee B. Automated discrimination of dementia spectrum disorders using extreme learning machine and structural t1 MRI features[C]//IEEE. EMBC2017. Jeju Island, Korea, 2017: 1990-1993.
|
[1] |
樊凤杰, 白洋, 纪会芳. 基于EEMD-ICA的脑电去噪算法研究[J]. 计量学报, 2021, 42(3): 395-400.
|
|
Li H X, Li J H, Li W L. A Visual Model Based on Attention Mechanism andConvolutional Neural Network[J]. Acta Metrologica Sinica, 2021, 42(7): 840-845.
|
|
Zhang S X, Yan X R, Sang Y. Single Image Rain Removal Method by Fusing Residual andChannel Attention Mechanism[J]. Acta Metrologica Sinica, 2021, 42(1): 20-28.
|
[16] |
Fan Z, Li J, Zhang L, et al. U-net based analysis of MRI for Alzheimers disease diagnosis[J]. Neural Computing and Applications, 2021: 13587-13599.
|
[6] |
Liu S Q, Liu S D, Cai W D, et al. Multimodal Neuroimaging Feature Learning for Multiclass Diagnosis of Alzheimers Disease[J]. IEEE Transactions on Biomedical Engineering, 2015, 62(4): 1132-1140.
|
[14] |
Korolev S, Safiullin A, Belyaev M, et al. Residual and plain convolutional neural networks for 3d brain MRI classification[C]//IEEE. ISBI 2017. Melbourne, Australia, 2017: 835-838.
|
[11] |
Saeeda N, Abida A, Ahmad Z. Transfer learning using freeze features for Alzheimer neurological disorder detection using ADNI dataset[J]. Multimedia Syst, 2022, 28(1): 85-94.
|
|
|
|