A transfer learning approach for multiclass classification of Alzheimer's disease using MRI images

Khan, Rizwan; Akbar, Saeed; Mehmood, Atif; Shahid, Farah; Munir, Khushboo; Ilyas, Naveed; Asif, M.; Zheng, Zhonglong

doi:10.3389/fnins.2022.1050777

Cited by 22 publications

(6 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, this model is unable to identify alterations in the brain networks of patients with mildly impaired functional working brain networks [ 24 ]. An accurate system was proposed by [ 25 ] that was based on transfer learning for the classification of AD at different stages. This approach categorizes normal, early-mild, late-mild, and AD brains.…”

Section: Literature Reviewmentioning

confidence: 99%

Improving Alzheimer’s Disease Classification in Brain MRI Images Using a Neural Network Model Enhanced with PCA and SWLDA

Ahmad,

Siddiqi,

Alhujaili

et al. 2023

Healthcare

View full text Add to dashboard Cite

The examination of Alzheimer’s disease (AD) using adaptive machine learning algorithms has unveiled promising findings. However, achieving substantial credibility in medical contexts necessitates a combination of notable accuracy, minimal processing time, and universality across diverse populations. Therefore, we have formulated a hybrid methodology in this study to classify AD by employing a brain MRI image dataset. We incorporated an averaging filter during preprocessing in the initial stage to reduce extraneous details. Subsequently, a combined strategy was utilized, involving principal component analysis (PCA) in conjunction with stepwise linear discriminant analysis (SWLDA), followed by an artificial neural network (ANN). SWLDA employs a combination of forward and backward recursion methods to choose a restricted set of features. The forward recursion identifies the most interconnected features based on partial Z-test values. Conversely, the backward recursion method eliminates the least correlated features from the same feature space. After the extraction and selection of features, an optimized artificial neural network (ANN) was utilized to differentiate the various classes of AD. To demonstrate the significance of this hybrid approach, we utilized publicly available brain MRI datasets using a 10-fold cross-validation strategy. The proposed method excelled over existing state-of-the-art systems, attaining weighted average recognition rates of 99.35% and 96.66%, respectively, across all the datasets.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Improving Alzheimer’s Disease Classification in Brain MRI Images Using a Neural Network Model Enhanced with PCA and SWLDA

Ahmad,

Siddiqi,

Alhujaili

et al. 2023

Healthcare

View full text Add to dashboard Cite

show abstract

“…Nonetheless, the rise of deep learning techniques, particularly transfer learning, has shown promise in enhancing AD detection from neuroimaging scans. However, it is worth noting that these methods often demand resource-intensive computational models [28].…”

Section: Related Workmentioning

confidence: 99%

Enhancing Early Dementia Detection: A Machine Learning Approach Leveraging Cognitive and Neuroimaging Features for Optimal Predictive Performance

Irfan,

Shahrestani,

Elkhodr

2023

Applied Sciences

View full text Add to dashboard Cite

Dementia, including Alzheimer’s Disease (AD), is a complex condition, and early detection remains a formidable challenge due to limited patient records and uncertainty in identifying relevant features. This paper proposes a machine learning approach to address this issue, utilizing cognitive and neuroimaging features for training predictive models. This study highlighted the viability of cognitive test scores in dementia detection—a procedure that offers the advantage of simplicity. The AdaBoost Ensemble model, trained on cognitive features, displayed a robust performance with an accuracy rate of approximately 83%. Notably, this model surpassed benchmark models such as the Artificial Neural Network, Support Vector Machine, and Naïve Bayes. This study underscores the potential of cognitive tests and machine learning for early dementia detection.

show abstract

“…Similarly, (19) aimed to extract useful AD biomarkers from structural MRI (sMRI) and classify brain images into AD, MCI, and cognitively normal (CN) groups. Leveraging transfer learning, (20) proposed a robust approach for classifying various stages of AD, highlighting the potential of deep learning in the realm of AD diagnosis. Underlining the importance of multi-modality approaches, (21) emphasized the need for efficient evaluation in identifying AD and its phases, advocating for the integration of diverse data sources for comprehensive analysis.…”

Section: Introductionmentioning

confidence: 99%

“…However, these approaches require further validation and standardization to ensure their clinical applicability. Furthermore, the development of multimodal, multi-kernel, and ensemble approaches presents opportunities for improved classification of different AD stages and enhanced diagnosis and research in AD (8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21) . The exploration of transfer learning, lightweight models, and deep neural networks also indicates the need for more efficient and accessible AD detection methods that can be widely applied (9,16,18) .…”

Section: Introductionmentioning

confidence: 99%

Data Complexity-Based Evaluation Using Brain Magnetic Resonance Images to Determine Alzheimer’s Disease

Kumar

2024

IJST

View full text Add to dashboard Cite

Objectives: This study aims to develop a robust diagnostic model for Alzheimer's Disease using a curated MRI dataset from Kaggle, integrating the BABC algorithm with a Random Forest classifier for precise classification. Methods: Feature extraction employed a modified LeNet model, effectively capturing crucial low-intensity pixel information. Additionally, the study integrated a binary version of the BABC algorithm for streamlined feature selection, enhancing data dimensionality and overall model efficacy. Findings: The study's findings mark a notable breakthrough in Alzheimer's Disease (AD) diagnosis. The model, which seamlessly combines the Biologically Inspired Artificial Bee Colony (BABC) algorithm and the Random Forest (RF) classifier, achieved an impressive accuracy of 98.97%. Furthermore, it exhibited a high recall rate of 97.22% and maintained precise precision at 97.59%. The robust F1-Score reached 98.63%, and a noteworthy specificity of 96.12% highlighted the model's ability to accurately classify diverse AD stages. Notably, the incorporation of a comprehensive confusion matrix enriched the study, offering profound insights into the model's predictive prowess. This comprehensive evaluation underscores the model's reliability and effectiveness in identifying various AD stages, positioning it as a promising tool for clinical diagnosis and research in neurodegenerative disorders. The proposed study's diagnostic model facilitates early intervention and effective treatment by identifying individuals at risk of Alzheimer's disease through cognitive assessments, neuroimaging, and biomarker analysis. Novelty: The integration of BABC with RF enhances AD diagnosis, promising improved accuracy and specificity, and fostering more effective management and treatment strategies.

show abstract

A transfer learning approach for multiclass classification of Alzheimer's disease using MRI images

Cited by 22 publications

References 55 publications

Improving Alzheimer’s Disease Classification in Brain MRI Images Using a Neural Network Model Enhanced with PCA and SWLDA

Improving Alzheimer’s Disease Classification in Brain MRI Images Using a Neural Network Model Enhanced with PCA and SWLDA

Enhancing Early Dementia Detection: A Machine Learning Approach Leveraging Cognitive and Neuroimaging Features for Optimal Predictive Performance

Data Complexity-Based Evaluation Using Brain Magnetic Resonance Images to Determine Alzheimer’s Disease

Contact Info

Product

Resources

About