Furkan KARAKAYA, Caglar GURKAN, Abdulkadir BUDAK, Hakan KARATAŞ

Derin Evrişimli Sinir Ağlarını Kullanarak MRG Modalitesinde Alzheimer Hastalığının Sınıflandırılması ve Segmentasyonu

Çalışmada Alzheimer hastalığının analizi için sınıflandırma ve segmentasyon görevleri uygulanmıştır. Sınıflandırma görevinde transfer öğrenme kullanılarak 7 farklı model test edilmiştir. GoogLeNet modeli 0.9467 doğruluk, 0.9474 duyarlılık, 0.9811 özgüllük ve 0.9467 F1 skoru ile en iyi sınıflandırma performansını elde etmiştir. Segmentasyon görevinde, Alzheimer hastalığının segmentasyonu için U-Net mimari tasarımı kullanılmıştır. U-Net modeli 0.874 zar skoru, 0.776 IoU, 0.868 duyarlılık, 0.999 özgüllük, 0.879 kesinlik ve 0.999 doğruluk elde etmiştir. Pipeline oluşturmak için sınıflandırma ve segmentasyon modelleri birlikte kullanılmıştır. Sonuç olarak, bilgisayarlı görü destekli bir karar destek sistemi oluşturulmuştur.

Anahtar Kelimeler:

Alzheimer, Sınıflandırma, Segmentasyon, GoogLeNet, U-Net

Classification and Segmentation of Alzheimer Disease in MRI Modality using the Deep Convolutional Neural Networks

In the study, classification and segmentation tasks were implemented for analysis of Alzheimer's disease. In classification task, 7 different models were tested using transfer learning. The GoogLeNet model achieved the best classification performance with the accuracy of 0.9467, sensitivity of 0.9474, specificity of 0.9811, and F1-score of 0.9467. In segmentation task, U-Net architecture design was used for the segmentation of Alzheimer's disease. U-Net model achieved the dice of 0.874, IoU of 0.776, sensitivity of 0.868, specificity of 0.999, precision of 0.879, and accuracy of 0.999. In order to create the pipeline, classification and segmentation models were used together. Consequently, a computer vision-assisted decision support system was created.

Keywords:

Alzheimer, Classification, Segmentation, GoogLeNet, U-Net,

PDF

___

Alp Eren, H., Okyay, S., Adar, N., Üniversitesi, E. O., Fakültesi, M.-M., Bölümü, M., Anahtar, T., & Öz, K. (2021). ADOKEN: MR İÇİN DERİN ÖĞRENME TABANLI KARAR DESTEK YAZILIMI. Journal of Engineering Sciences and Design, 9(2), 406–413. https://doi.org/10.21923/JESD.887327
Öziç, M. Ü., & Özşen, S. (2020). Classification of 3b alzheimer’s mr images using voxel values in volumetric loss regions. El-Cezeri Journal of Science and Engineering, 7(3), 1152–1166. https://doi.org/10.31202/ecjse.728049
John, R., & Kunju, N. (2018). Detection of Alzheimer’s Disease Using Fractional Edge Detection. Journal of Biodiversity & Endangered Species, 09(03). https://doi.org/10.4172/2229-8711.1000230
Khvostikov, A., Aderghal, K., Benois-Pineau, J., Krylov, A., & Catheline, G. (2018). 3D CNN-based classification using sMRI and MD-DTI images for Alzheimer disease studies. https://ida.loni.usc.edu
Hong, X., Lin, R., Yang, C., Zeng, N., Cai, C., Gou, J., & Yang, J. (2019). Predicting Alzheimer’s Disease Using LSTM. IEEE Access, 7, 80893–80901. https://doi.org/10.1109/ACCESS.2019.2919385
Allioui, H., Sadgal, M., & Elfazziki, A. (2019). Deep MRI Segmentation: A Convolutional Method Applied to Alzheimer Disease Detection. IJACSA) International Journal of Advanced Computer Science and Applications, 10(11). www.ijacsa.thesai.org
Ahmed, S., Choi, K. Y., Lee, J. J., Kim, B. C., Kwon, G. R., Lee, K. H., & Jung, H. Y. (2019). Ensembles of Patch-Based Classifiers for Diagnosis of Alzheimer Diseases. IEEE Access, 7, 73373–73383. https://doi.org/10.1109/ACCESS.2019.2920011
Vieira, S., Pinaya, W. H. L., & Mechelli, A. (2017). Using deep learning to investigate the neuroimaging correlates of psychiatric and neurological disorders: Methods and applications. Neuroscience & Biobehavioral Reviews, 74, 58–75. https://doi.org/10.1016/J.NEUBIOREV.2017.01.002
SARVESH DUBEY. (2020). Alzheimer’s Dataset ( 4 class of Images) Images of MRI Segementation. https://www.kaggle.com/datasets/tourist55/alzheimers-dataset-4-class-of-images
ADNI | Alzheimer’s Disease Neuroimaging Initiative. (2003). https://adni.loni.usc.edu/
Weng, W., & Zhu, X. (2021). INet: Convolutional Networks for Biomedical Image Segmentation. IEEE Access, 9, 16591–16603. https://doi.org/10.1109/ACCESS.2021.3053408
Iqbal, H. (2018). Harisiqbal88/plotneuralnet v1. 0.0. URL: https://doi. org/10.5281/Zenodo.