Improved neural network with multi-task learning for Alzheimer's disease classification
Alzheimer's disease(AD) poses a significant challenge due to its widespread prevalence and the lack of effective treatments, highlighting the urgent need for early detection. This research introduces an enhanced neural network, named ADnet, which is based on the VGG16 model, to detect Alzheimer...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-02-01
|
| Series: | Heliyon |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405844024024368 |
| _version_ | 1797267647892553728 |
|---|---|
| author | Xin Zhang Le Gao Zhimin Wang Yong Yu Yudong Zhang Jin Hong |
| author_facet | Xin Zhang Le Gao Zhimin Wang Yong Yu Yudong Zhang Jin Hong |
| author_sort | Xin Zhang |
| collection | DOAJ |
| description | Alzheimer's disease(AD) poses a significant challenge due to its widespread prevalence and the lack of effective treatments, highlighting the urgent need for early detection. This research introduces an enhanced neural network, named ADnet, which is based on the VGG16 model, to detect Alzheimer's disease using two-dimensional MRI slices. ADNet incorporates several key improvements: it replaces traditional convolution with depthwise separable convolution to reduce model parameters, replaces the ReLU activation function with ELU to address potential issues with exploding gradients, and integrates the SE(Squeeze-and-Excitation) module to enhance feature extraction efficiency. In addition to the primary task of MRI feature extraction, ADnet is simultaneously trained on two auxiliary tasks: clinical dementia score regression and mental state score regression. Experimental results demonstrate that compared to the baseline VGG16, ADNet achieves a 4.18% accuracy improvement for AD vs. CN classification and a 6% improvement for MCI vs. CN classification. These findings highlight the effectiveness of ADnet in classifying Alzheimer's disease, providing crucial support for early diagnosis and intervention by medical professionals. The proposed enhancements represent advancements in neural network architecture and training strategies for improved AD classification. |
| first_indexed | 2024-03-07T22:00:45Z |
| format | Article |
| id | doaj.art-0a80ca08bbab4aa6bdb7693e2c019ce9 |
| institution | Directory Open Access Journal |
| issn | 2405-8440 |
| language | English |
| last_indexed | 2024-04-25T01:19:55Z |
| publishDate | 2024-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Heliyon |
| spelling | doaj.art-0a80ca08bbab4aa6bdb7693e2c019ce92024-03-09T09:28:13ZengElsevierHeliyon2405-84402024-02-01104e26405Improved neural network with multi-task learning for Alzheimer's disease classificationXin Zhang0Le Gao1Zhimin Wang2Yong Yu3Yudong Zhang4Jin Hong5School of Electronic and Information Engineering, Wuyi University, Jiangmen, 529000, ChinaSchool of Electronic and Information Engineering, Wuyi University, Jiangmen, 529000, China; Corresponding author.School of Electronic and Information Engineering, Wuyi University, Jiangmen, 529000, ChinaSchool of Computer Science, Shaanxi Normal University, Xi'an, 710062, ChinaSchool of Computing and Mathematic Sciences, University of Leicester, Leicester, LE17RH, UKSchool of Information Engineering, Nanchang University, Nanchang, 330031, ChinaAlzheimer's disease(AD) poses a significant challenge due to its widespread prevalence and the lack of effective treatments, highlighting the urgent need for early detection. This research introduces an enhanced neural network, named ADnet, which is based on the VGG16 model, to detect Alzheimer's disease using two-dimensional MRI slices. ADNet incorporates several key improvements: it replaces traditional convolution with depthwise separable convolution to reduce model parameters, replaces the ReLU activation function with ELU to address potential issues with exploding gradients, and integrates the SE(Squeeze-and-Excitation) module to enhance feature extraction efficiency. In addition to the primary task of MRI feature extraction, ADnet is simultaneously trained on two auxiliary tasks: clinical dementia score regression and mental state score regression. Experimental results demonstrate that compared to the baseline VGG16, ADNet achieves a 4.18% accuracy improvement for AD vs. CN classification and a 6% improvement for MCI vs. CN classification. These findings highlight the effectiveness of ADnet in classifying Alzheimer's disease, providing crucial support for early diagnosis and intervention by medical professionals. The proposed enhancements represent advancements in neural network architecture and training strategies for improved AD classification.http://www.sciencedirect.com/science/article/pii/S2405844024024368Alzheimer's diseaseVGG16 networkMulti-task learning |
| spellingShingle | Xin Zhang Le Gao Zhimin Wang Yong Yu Yudong Zhang Jin Hong Improved neural network with multi-task learning for Alzheimer's disease classification Heliyon Alzheimer's disease VGG16 network Multi-task learning |
| title | Improved neural network with multi-task learning for Alzheimer's disease classification |
| title_full | Improved neural network with multi-task learning for Alzheimer's disease classification |
| title_fullStr | Improved neural network with multi-task learning for Alzheimer's disease classification |
| title_full_unstemmed | Improved neural network with multi-task learning for Alzheimer's disease classification |
| title_short | Improved neural network with multi-task learning for Alzheimer's disease classification |
| title_sort | improved neural network with multi task learning for alzheimer s disease classification |
| topic | Alzheimer's disease VGG16 network Multi-task learning |
| url | http://www.sciencedirect.com/science/article/pii/S2405844024024368 |
| work_keys_str_mv | AT xinzhang improvedneuralnetworkwithmultitasklearningforalzheimersdiseaseclassification AT legao improvedneuralnetworkwithmultitasklearningforalzheimersdiseaseclassification AT zhiminwang improvedneuralnetworkwithmultitasklearningforalzheimersdiseaseclassification AT yongyu improvedneuralnetworkwithmultitasklearningforalzheimersdiseaseclassification AT yudongzhang improvedneuralnetworkwithmultitasklearningforalzheimersdiseaseclassification AT jinhong improvedneuralnetworkwithmultitasklearningforalzheimersdiseaseclassification |