Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 May 8;13(9):1654.
doi: 10.3390/diagnostics13091654.

Automatic Analysis of MRI Images for Early Prediction of Alzheimer's Disease Stages Based on Hybrid Features of CNN and Handcrafted Features

Ahmed Khalid  1 , Ebrahim Mohammed Senan  2 , Khalil Al-Wagih  2 , Mamoun Mohammad Ali Al-Azzam  1 , Ziad Mohammad Alkhraisha  1
Affiliations

Automatic Analysis of MRI Images for Early Prediction of Alzheimer's Disease Stages Based on Hybrid Features of CNN and Handcrafted Features

Ahmed Khalid et al. Diagnostics (Basel). .

Abstract

Alzheimer's disease (AD) is considered one of the challenges facing health care in the modern century; until now, there has been no effective treatment to cure it, but there are drugs to slow its progression. Therefore, early detection of Alzheimer's is vital to take needful measures before it develops into brain damage which cannot be treated. Magnetic resonance imaging (MRI) techniques have contributed to the diagnosis and prediction of its progression. MRI images require highly experienced doctors and radiologists, and the analysis of MRI images takes time to analyze each slice. Thus, deep learning techniques play a vital role in analyzing a huge amount of MRI images with high accuracy to detect Alzheimer's and predict its progression. Because of the similarities in the characteristics of the early stages of Alzheimer's, this study aimed to extract the features in several methods and integrate the features extracted from more than one method into the same features matrix. This study contributed to the development of three methodologies, each with two systems, with all systems aimed at achieving satisfactory accuracy for the detection of AD and predicting the stages of its progression. The first methodology is by Feed Forward Neural Network (FFNN) with the features of GoogLeNet and DenseNet-121 models separately. The second methodology is by FFNN network with combined features between GoogLeNet and Dense-121 models before and after high-dimensionality reduction of features using the Principal Component Analysis (PCA) algorithm. The third methodology is by FFNN network with combined features between GoogLeNet and Dense-121 models separately and features extracted by Discrete Wavelet Transform (DWT), Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM) methods called handcrafted features. All systems yielded super results in detecting AD and predicting the stages of its progression. With the combined features of the DenseNet-121 and handcrafted, the FFNN achieved an accuracy of 99.7%, sensitivity of 99.64%, AUC of 99.56%, precision of 99.63%, and a specificity of 99.67%.

Keywords: AD; CNN; DWT; FFNN; GLCM; LBP; fusion features.

PubMed Disclaimer

Conflict of interest statement

The authors confirm there is no conflict of interest between them.

Figures

Figure 1
Figure 1
Methodology framework for diagnosing MRI images for early detection of AD progression stages.
Figure 2
Figure 2
MRI image samples for the AI dataset (a) before improved images and (b) after improved image.
Figure 3
Figure 3
Displaying the distribution of MRI images for the AD dataset before and after applying the data augmentation.
Figure 4
Figure 4
Basic methodology for diagnosing MRI images of the AD by FFNN according to CNN features.
Figure 5
Figure 5
Basic methodology for diagnosing MRI images of the AD dataset by FFNN according to the fusion of CNN features.
Figure 6
Figure 6
Basic methodology for diagnosing MRI images of the AD dataset by FFNN according to the fusion of CNN features and handcrafted feature.
Figure 7
Figure 7
Confusion matrix of FFNN performance for detection of AD based on features (a) GoogLeNet and (b) DenseNet-121.
Figure 8
Figure 8
Confusion matrix of FFNN performance for AD detection based on combined features of the GoogLeNet and DenseNet-121 (a) before PCA and (b) after PCA.
Figure 9
Figure 9
Confusion matrix of FFNN performance for MRI for detection of AD based on combined features between (a) GoogLeNet and handcrafted features and (b) DenseNet-121 and handcrafted features.
Figure 10
Figure 10
Display error histogram for diagnosing MRI for detection of Alzheimer’s progression by FFNN with combined features of (a) GoogLeNet and handcrafted features and (b) DenseNet-121 and handcrafted features.
Figure 11
Figure 11
Display cross-entropy for diagnosing MRI for detection of Alzheimer’s progression by FFNN with combined features of (a) GoogLeNet and handcrafted features and (b) DenseNet-121 and handcrafted features.
Figure 12
Figure 12
Display gradient and validation checks for recognizing MRI of Alzheimer’s progression by FFNN with combined features of (a) GoogLeNet and handcrafted features and (b) DenseNet-121 and handcrafted features.
Figure 13
Figure 13
Confusion matrix for generalization performance of FFNN for MRI image analysis of an ADNI dataset based on fused features between (a) GoogLeNet and handcrafted and (b) DenseNet-121 and handcrafted.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Grimmer T., Henriksen G., Wester H.J., Förstl H., Klunk W.E., Mathis C.A., Kurz A., Drzezga A. The clinical severity of Alzheimer’s disease is associated with PIB uptake in PET. Neurobiol. Aging. 2009;30:1902–1909. doi: 10.1016/j.neurobiolaging.2008.01.016. - DOI - PubMed
    1. Aël Chetelat G., Baron J.C. Early diagnosis of Alzheimer’s disease: Contribution of structural neuroimaging. [(accessed on 15 August 2022)];Neuroimage. 2003 18:525–541. doi: 10.1016/S1053-8119(02)00026-5. Available online: https://www.sciencedirect.com/science/article/pii/S1053811902000265. - DOI - PubMed
    1. Duchesne S., Caroli A., Geroldi C., Barillot C., Frisoni G.B., Collins D.L. MRI-based automated computer classification of probable AD versus normal controls. IEEE. 2008;27:509–520. doi: 10.1109/TMI.2007.908685. - DOI - PubMed
    1. Goren A., Montgomery W., Kahle-Wrobleski K., Nakamura T., Ueda K. Impact of caring for persons with Alzheimer’s disease or dementia on caregivers’ health outcomes: Findings from a community-based survey in Japan. BMC Geriatr. 2016;16:1–12. doi: 10.1186/s12877-016-0298-y. - DOI - PMC - PubMed
    1. Eisenmenger L.B., Peret A., Famakin B.M., Spahic A., Roberts G.S., Bockholt J.H., Paulsen J.S. Vascular Contributions to Alzheimer’s Disease. Transl. Res. 2022 doi: 10.1016/j.trsl.2022.12.003. - DOI - PMC - PubMed

LinkOut - more resources