Deep CNN ResNet-18 based model with attention and transfer learning for Alzheimer's disease detection
IntroductionThe prevalence of age-related brain issues has risen in developed countries because of changes in lifestyle. Alzheimer's disease leads to a rapid and irreversible decline in cognitive abilities by damaging memory cells.MethodsA ResNet-18-based system is proposed, integrating Depth C...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-01-01
|
| Series: | Frontiers in Neuroinformatics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fninf.2024.1507217/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841554402544975872 |
|---|---|
| author | Sofia Biju Francis Sofia Biju Francis Jai Prakash Verma |
| author_facet | Sofia Biju Francis Sofia Biju Francis Jai Prakash Verma |
| author_sort | Sofia Biju Francis |
| collection | DOAJ |
| description | IntroductionThe prevalence of age-related brain issues has risen in developed countries because of changes in lifestyle. Alzheimer's disease leads to a rapid and irreversible decline in cognitive abilities by damaging memory cells.MethodsA ResNet-18-based system is proposed, integrating Depth Convolution with a Squeeze and Excitation (SE) block to minimize tuning parameters. This design is based on analyses of existing deep learning architectures and feature extraction techniques. Additionally, pre-trained ResNet-18 models were created with and without the SE block to compare ROC and accuracy values across different hyperparameters.ResultsThe proposed model achieved ROC values of 95% for Alzheimer's Disease (AD), 95% for Cognitively Normal (CN), and 93% for Mild Cognitive Impairment (MCI), with a maximum test accuracy of 88.51%. However, the pre-trained model with SE had 93.26% accuracy and ROC values of 98%, 99%, and 98%, while the model without SE had 94%, 97%, and 94% ROC values and 92.41% accuracy.DiscussionCollecting medical data can be expensive and raises ethical concerns. Small data sets are also prone to local minima issues in the cost function. A scratch model that experiences extensive hyperparameter tuning may end up being either overfitted or underfitted. Class imbalance also reduces performance. Transfer learning is most effective with small, imbalanced datasets, and pre-trained models with SE blocks perform better than others. The proposed model introduced a method to reduce training parameters and prevent overfitting from imbalanced medical data. Overall performance findings show that the suggested approach performs better than the state-of-the-art techniques. |
| format | Article |
| id | doaj-art-5654222847eb4ec0bb5231b213802442 |
| institution | Kabale University |
| issn | 1662-5196 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroinformatics |
| spelling | doaj-art-5654222847eb4ec0bb5231b2138024422025-01-08T13:40:55ZengFrontiers Media S.A.Frontiers in Neuroinformatics1662-51962025-01-011810.3389/fninf.2024.15072171507217Deep CNN ResNet-18 based model with attention and transfer learning for Alzheimer's disease detectionSofia Biju Francis0Sofia Biju Francis1Jai Prakash Verma2Department of Computer Science and Engineering, Institute of Technology, Nirma University, Gujarat, IndiaDepartment of Computer Engineering, NMIMS, MPSTME, Mumbai, IndiaDepartment of Computer Science and Engineering, Institute of Technology, Nirma University, Gujarat, IndiaIntroductionThe prevalence of age-related brain issues has risen in developed countries because of changes in lifestyle. Alzheimer's disease leads to a rapid and irreversible decline in cognitive abilities by damaging memory cells.MethodsA ResNet-18-based system is proposed, integrating Depth Convolution with a Squeeze and Excitation (SE) block to minimize tuning parameters. This design is based on analyses of existing deep learning architectures and feature extraction techniques. Additionally, pre-trained ResNet-18 models were created with and without the SE block to compare ROC and accuracy values across different hyperparameters.ResultsThe proposed model achieved ROC values of 95% for Alzheimer's Disease (AD), 95% for Cognitively Normal (CN), and 93% for Mild Cognitive Impairment (MCI), with a maximum test accuracy of 88.51%. However, the pre-trained model with SE had 93.26% accuracy and ROC values of 98%, 99%, and 98%, while the model without SE had 94%, 97%, and 94% ROC values and 92.41% accuracy.DiscussionCollecting medical data can be expensive and raises ethical concerns. Small data sets are also prone to local minima issues in the cost function. A scratch model that experiences extensive hyperparameter tuning may end up being either overfitted or underfitted. Class imbalance also reduces performance. Transfer learning is most effective with small, imbalanced datasets, and pre-trained models with SE blocks perform better than others. The proposed model introduced a method to reduce training parameters and prevent overfitting from imbalanced medical data. Overall performance findings show that the suggested approach performs better than the state-of-the-art techniques.https://www.frontiersin.org/articles/10.3389/fninf.2024.1507217/fullAlzheimer's diseasecognitive deteriorationResNet-18depth convolutionsqueeze and excitationtransfer learning |
| spellingShingle | Sofia Biju Francis Sofia Biju Francis Jai Prakash Verma Deep CNN ResNet-18 based model with attention and transfer learning for Alzheimer's disease detection Frontiers in Neuroinformatics Alzheimer's disease cognitive deterioration ResNet-18 depth convolution squeeze and excitation transfer learning |
| title | Deep CNN ResNet-18 based model with attention and transfer learning for Alzheimer's disease detection |
| title_full | Deep CNN ResNet-18 based model with attention and transfer learning for Alzheimer's disease detection |
| title_fullStr | Deep CNN ResNet-18 based model with attention and transfer learning for Alzheimer's disease detection |
| title_full_unstemmed | Deep CNN ResNet-18 based model with attention and transfer learning for Alzheimer's disease detection |
| title_short | Deep CNN ResNet-18 based model with attention and transfer learning for Alzheimer's disease detection |
| title_sort | deep cnn resnet 18 based model with attention and transfer learning for alzheimer s disease detection |
| topic | Alzheimer's disease cognitive deterioration ResNet-18 depth convolution squeeze and excitation transfer learning |
| url | https://www.frontiersin.org/articles/10.3389/fninf.2024.1507217/full |
| work_keys_str_mv | AT sofiabijufrancis deepcnnresnet18basedmodelwithattentionandtransferlearningforalzheimersdiseasedetection AT sofiabijufrancis deepcnnresnet18basedmodelwithattentionandtransferlearningforalzheimersdiseasedetection AT jaiprakashverma deepcnnresnet18basedmodelwithattentionandtransferlearningforalzheimersdiseasedetection |