MR Görüntülerinde Evrişimli Sinir Ağlar Kullanılarak Alzheimer Hastalık Tespiti

Year 2023, Volume: 39 Issue: 3, 357 - 368, 31.12.2023

Sümeyye Aydın Murat Taşyürek Celal Öztürk

Abstract

Demans beyni etkileyen, hafıza ve düşünmede eksikliklere neden olan bir bulgudur. Demans türleri arasında en sık rastlanılan Alzheimer hastalığı (AH) özellikle yaşlı bireylerde görülen, beyinde hafıza kaybına ve gündelik hayatı aksatmaya neden olan bir rahatsızlıktır. Bu hastalığın tedavisi henüz mümkün olmasa da ilerlemesini engelleyecek tedbirler alınabilmektedir. Erken teşhis ile beyinde kalıcı hasarlar oluşmadan tedaviye başlanarak önemli ölçüde yol kat edilebilir. Manyetik rezonans (MR) görüntüleme AH teşhisinde yaygın olarak kullanılmaktadır, MR görüntülerinin analizi ile beyindeki hacimsel değişimler gözlemlenebilir ve AH sınıflandırılması yapılabilir. Derin öğrenme yöntemleri ile görüntü analizlerinde hızlı ve başarılı sonuçlar alındığı literatür taramasında gözlemlenmiştir. Bu çalışmada MR görüntüleri ile AH sınıflandırmasında derin öğrenme yöntemlerinin başarısı araştırılmıştır. AH sınıflandırması için iki boyutlu evrişimli sinir ağları (2D CNN) algoritması olan YOLO v4 ve üç boyutlu evrişimli sinir ağı modeli (3D CNN) önerilmiştir. 2D CNN modeli için MR verileri iki boyuta indirgenmiş ve önişlemlerden geçirilmiş haliyle eğitilip sınıflandırma işlemi yapılmıştır. Sınıflandırma sonuçlarında 3D CNN modeli 2D CNN modeline göre daha başarılı çıkmıştır. 2D CNN modelinde doğruluk değeri %68 iken bu değer 3D CNN modelinde %88’e yükselmiştir.

Keywords

Derin Öğrenme, Evrişimli Sinir Ağları, Alzheimer hastalığı

Alzheimer's Disease Detection Using Convolutional Neural Networks on MR Images

Abstract

Dementia is a disease that affects the brain and causes deficits in memory and thinking. Alzheimer's disease (AD), which is the most common type of dementia, is a disorder that causes memory loss and disruption of daily life, especially in elderly individuals. Although the treatment of this disease is not yet possible, make provision for can be taken to prevent its progression. Thanks to early diagnosis, treatment can go a long way before permanent damage to the brain occurs.. Magnetic resonance (MR) imaging is widely used in the diagnosis of AD, with the analysis of MR images, volumetric changes in the brain can be observed; With these analyzes, it is possible to identify and classify the stages of the disease. It has been observed in the literature review that fast and successful results are obtained in image analysis deep learning methods. In this study, MR images were used and the success of deep learning methods in AD classification was investigated. YOLO v4, a two-dimensional convolutional neural network (2D CNN) algorithm, and a three-dimensional convolutional neural network model (3D CNN) have been proposed for AD classification. For the 2D CNN model, the MR data was reduced to 2 dimensions and preprocessed, trained and classified. In the classification results, the 3D CNN model was quite successful compared to the 2D CNN model. While the accuracy value was 68% in the 2D CNN model, this value increased to 88% in the 3D CNN model.

Keywords

Deep Learning, Convolutional Neural Network, Alzheimer

References

• [1] Keskin, M., 2021. Alzheimer Hastalarına Bakım Verenlerin Kaygı, Stres, Başetme Stilleri ve Yaşam Deneyimleri. Hacettepe Üniversitesi, Sosyal Bilimler Enstitüsü, Doktora Tezi, Ankara, 14.
• [2] 2020 Alzheimer's disease facts and figures. Alzheimer's & Dementia., 16: 391-460.
• [3] Breijyeh, Z.; Karaman, R., 2020. Comprehensive Review on Alzheimer’s Disease: Causes and Treatment. Molecules, 25: 5789. https://doi.org/10.3390/molecules25245789
• [4] Gray, K. R., Aljabar, P., Heckemann, R. A., Hammers, A., Rueckert, D., and Alzheimer’s Disease Neuroimaging, I., 2013. Random forest-based similarity measures for multi-modal classification of Alzheimer’s disease. Neuroimage 65, 167–175.
• [5] 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.
• [6] Dukart, J., Mueller, K., Barthel, H., Villringer, A., Sabri, O., Schroeter, M. L., & Alzheimer's Disease Neuroimaging Initiative, 2013. Meta-analysis based SVM classification enables accurate detection of Alzheimer's disease across different clinical centers using FDG-PET and MRI. Psychiatry Research: Neuroimaging, 212(3), 230 236.
• [7] Shankar, K., Lakshmanaprabu, S. K., Khanna, A., Tanwar, S., Rodrigues, J. J., & Roy, N., R., 2019. Alzheimer detection using Group Grey Wolf Optimization based features with convolutional classifier. Computers & Electrical Engineering, 77, 230-243.
• [8] Hosseini-Asl, E., Keynton, R., El-Baz, A., 2016. Alzheimer's disease diagnostics by adaptation of 3D convolutional network, 2016 IEEE International Conference on Image Processing (ICIP),126-130, doi: 10.1109/ICIP.2016.7532332.
• [9] Payan, A., Montana, G., 2015. Predicting Alzheimer's disease: a neuroimaging study with 3D convolutional neural networks. arXiv preprint arXiv:1502.02506
• [10] Sahumbaiev, I., Popov, A., Ramírez, J., Górriz, J., M., Ortiz, A., 2018. 3D-CNN HadNet classification of MRI for Alzheimer’s Disease diagnosis, 2018 IEEE Nuclear Science Symposium and Medical Imaging Conference Proceedings (NSS/MIC), 1-4, doi: 10.1109/NSSMIC.2018.8824317.
• [11] Hon, M. and Khan, N., 2017. Towards Alzheimer's Disease Classification through Transfer Learning. arXivpreprint arXiv:1711.11117.
• [12] Korolev S, Safiullin A, Belyaev M, Dodonova Y. Residual and plain convolutional neural networks for 3d brain MRI classification Sergey Korolev Amir Safiullin Mikhail Belyaev Skolkovo Institute of Science and Technology Institute for Information Transmission Problems, 2017 IEEE 14th Int. Symp. Biomed. Imaging (ISBI 2017). 2017;835–838.
• [13] Sarraf, S., Tofighi, G., 2016. Deep learning-based pipeline to recognize Alzheimer's disease using fMRI data, 2016 Future Technologies Conference (FTC), 816-820.
• [14] Helaly, H., A., Badawy, M., Haikal, A., Y, 2022. Deep Learning Approach for Early Detection of Alzheimer’s Disease. Cognitive Computation, 14, 1711–1727.
• [15] S. M. Smith. Fast robust automated brain extraction. Human Brain Mapping, 17(3):143-155, November 2002.
• [16] Vu, T., D., Yang, H., -J., Nguyen, V., Q., Oh, A., -R., Kim, M., -S., 2017. Multimodal learning using convolution neural network and Sparse Autoencoder, 2017 IEEE International Conference on Big Data and Smart Computing (BigComp), 309-312.
• [17] Janiesch, C., Zschech, P., Heinrich, K., 2021. Machine learning and deep learning. Electronic Markets, 31(3), 685-695.
• [18] Redmon, J., Divvala, S., Girshick, R. and Farhadi, A., 2016. You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, 779-788.
• [19] Thuan, D., 2021. Evolution of yolo algorithm and yolov5: the state-of-the-art object detection algorithm, Oulu University of Applied Sciences, Bachelor’s Thesis. Conference
• [20] Lan, W., Dang, J., Wang, Y., Wang, S., 2018. Pedestrian Detection Based on YOLO Network Model, 2018 IEEE International on Mechatronics and Automation (ICMA), 1547-1551, doi: 10.1109/ICMA.2018.8484698.
• [21] Rothe, R., Guillaumin, M., Gool, L. V., 2014, November. Non-maximum suppression for object detection by passing messages between windows. In Asian conference on computer vision, 290-306.
• [22] Sambasivarao, K., MR Görüntülerinde Evrişimli Sinir Ağlar Kullanılarak Alzheimer Hastalık Tespiti (Web sayfası: 2022. Non-maximum Suppression (NMS). https://towardsdatascience.com/non-maximum-suppression-nms-93ce178e177c), (Erişim tarihi: Ekim 2022).
• [23] Bochkovskiy, A., Wang, C., Liao, H., M., 2020. YOLOv4: Optimal Speed and Accuracy of Object Detection. ArXiv, abs/2004.10934.
• [24] Taşyürek, M., Öztürk, C., 2022. DDL: Çoklu kapı numarası tespit etme ve kümeleme için derin öğrenme tabanlı yeni bir yaklaşım. Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, 37 (2), 843-856.
• [25] Alan, A., Karabatak, M., 2020. Veri Seti - Sınıflandırma İlişkisinde Performansa Etki Eden Faktörlerin Değerlendirilmesi . Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 32 (2) , 531-540 .
• [26] Khagi, B., Kwon, G.-R., 2020. 3D CNN Design for the Classification of Alzheimer’s Disease Using Brain MRI and PET. IEEE Access, 8, 217830-217847, 2020, doi: 10.1109/ACCESS.2020.3040486.
• [27] Nguyen, A., Yosinski, J., Clune, J., 2015. Deep neural networks are easily fooled: High confidence predictions for unrecognizable images, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 427 436.
• [28] Mzoughi, H., Njeh, I., Wali, A., Slima, M., B., BenHamida, A., Mhiri, C., Mahfoudhe, K., B., 2020. Deep multi-scale 3D Convolutional Neural Network (CNN) for MRI gliomas brain tumor classification. Journal of Digital Imaging, 33(4):903-915. doi: 10.1007/s10278-020-00347-9. PMID: 32440926; PMCID: PMC7522155.
• [29] Zhou, M., Chaudhury, B., Hall, L., O., Goldgof, D., B., Gillies, R., J., Gatenby, R., A., 2017. Identifying spatial imaging biomarkers of glioblastoma multiforme for survival group prediction. Journal of Magnetic Resonance Imaging, 46(1):115–123. doi: 10.1002/jmri.25497.
• [30] Rao, C., Liu, Y., 2020. Three-dimensional convolutional neural network (3D-CNN) for heterogeneous material homogenization. ArXiv, abs/2002.07600.
• [31] Haehn, D., Rannou, N., Ahtam, B., Grant, E., Pienaar, R., 2014. Neuroimaging in the Browser using the X Toolkit. Frontiers in Neuroinformatics.