Classification of Alzheimer's disease with EfficientNet B3

Abstract

Alzheimer's disease (AD) is a condition that manifests as a loss of consciousness and cognitive dysfunction, eventually leaving the individual incapable of performing basic functions. The process culminates in death. The brain anomalies caused by the disease can be monitored using magnetic resonance imaging (MRI). This study aims to facilitate the clinical diagnosis of AD and proposes a hybrid model to classify the stages of the disease. The magnetic resonance images used in the study were obtained from the Kaggle database and include the classes non-demented, very mild dementia, mild dementia, and moderate dementia. Background removal was applied to the images, which were then segmented using the k-means clustering method. By combining EfficientNet B3 and the Gray Level Co-Occurrence Matrix (GLCM) feature extractor, this hybrid model was trained to perform the classification task. The model was trained five times, and experimental results were recorded. In training, the batch size was set to 18, the number of epochs was 20, and the learning rate was set to 0.0001. Experimental results showed an average training accuracy of 99.99% and a testing accuracy of 99.67%. Additional performance metrics, such as precision, recall, and F1-score, are also reported.

Keywords

• Alzheimer
• Machine learning
• EffcientNet B3
• Gray Level Co-Occurence Matrix
• Image segmentation

Classification of Alzheimer's disease with EfficientNet B3

Abstract

Alzheimer hastalığı (AH), bilinç kaybı ve bilişsel işlev bozukluğu olarak ortaya çıkan ve sonunda bireyi temel işlevleri yerine getiremez hale getiren bir durumdur. Süreç ölümle sonuçlanır. Hastalığın neden olduğu beyin anomalileri manyetik rezonans görüntüleme (MRI) kullanılarak izlenebilir. Bu çalışma, AD'nin klinik teşhisini kolaylaştırmayı amaçlamakta ve hastalığın evrelerini sınıflandırmak için hibrit bir model önermektedir. Çalışmada kullanılan manyetik rezonans görüntüleri Kaggle veri tabanından elde edilmiştir ve bunama olmayan, çok hafif bunama, hafif bunama ve orta derecede bunama sınıflarını içermektedir. Görüntülere arka plan kaldırma işlemi uygulanmış ve ardından k-ortalamalar kümeleme yöntemi kullanılarak segmentlere ayrılmıştır. EfficientNet B3 ve Gri Seviye Eş Oluşum Matrisi (GLCM) özellik çıkarıcısını birleştiren bu hibrit model, sınıflandırma görevini yerine getirmek üzere eğitilmiştir. Model beş kez eğitilmiş ve deneysel sonuçlar kaydedilmiştir. Eğitimde, yığın boyutu 18, epok sayısı 20 ve öğrenme oranı 0.0001 olarak ayarlanmıştır. Deneysel sonuçlar ortalama %99,99 eğitim doğruluğu ve %99,67 test doğruluğu göstermiştir. Kesinlik, geri çağırma ve F1-skoru gibi ek performans ölçümleri de raporlanmıştır.

Keywords

• Alzheimer
• Machine learning
• Gray Level Co-Occurence Matrix
• Image segmentation

References

1. M. Ü. Öziç and S. Özşen, "Üç Boyutlu T1 Ağırlıklı Manyetik Rezonans Görüntülerinde Ön İşleme Yöntemleri," Avrupa Bilim ve Teknoloji Dergisi, no. 19, pp. 227-240, 2020.
2. E. Gülay and S. İçer, “Evaluation of Lung Size in Patients with Pneumonia and Healthy Individuals”, Avrupa Bilim Ve Teknoloji Dergisi,no.özel sayı,pp. 304-309, 2020.
3. A. Burns and S.Iliffe, “Alzheimer's disease”, British Medical Journal, Vol.6, No.8,pp. 338, b158, 2014.
4. K. Keski̇n and F. Tokat, “ALZHEİMER VE FİZİKSEL AKTİVİTE”, Fiziksel Aktivite ve Sağlık, pp. 285-293, 2023.
5. S. Pala, “Alzheimer hastalığının erken teşhisi için biyobelirteçlere dayalı stratejik yol haritası derleme çeviri çalışması”, Tıbbi Politika Yazısı, 2021.
6. P. S. Sisodia, G. K. Ameta, Y. Kumar et al. “A Review of Deep Transfer Learning Approaches for Class-Wise Prediction of Alzheimer’s Disease Using MRI Images”, Archives of Computational Methods in Engineering, vol. 30, pp. 2409–2429, 2023.
7. V. Sanjay and P. Swarnalatha, “A Concatenated Deep Feature Extraction Architecture For Multi-Class Alzheimer Disease Prediction”, Journal of Advanced Research in Applied Sciences and Engineering Technology, vol. 33, pp. 102-121, 2023.
8. H. Sharen, B. Dhanush et al., “Efficient Diagnosis of Alzheimer’s Disease Using EfficientNet in Neuroimaging”, Lecture Notes in Electrical Engineering, vol. 914, 2022.

9. M. Sethi, S. Ahuja, et al., "An Intelligent Framework for Alzheimer's disease Classification Using EfficientNet Transfer Learning Model," 2022 International Conference on Emerging Smart Computing and Informatics (ESCI), pp. 1-4,2022
10. M. Mujahid, A. Rehman, et al., “An Efficient Ensemble Approach for Alzheimer's Disease Detection Using an Adaptive Synthetic Technique and Deep Learning”, Diagnostics, vol. 13, no. 15: 2489.
11. Y. F. Pranata, R. Magdalena and N. K. C Pratiwi, “Optimizer analysis on efficient-net architecture for Alzheimer’s classification based on magnetic resonance imaging (MRI)”, The 3rd International Conference on Engineering Technology and Innovative Researches, vol. 2482, Issue 1, 2023.
12. Uraninjo., Augmented Alzheimer MRI Dataset, Kaggle, 2022. [Date of access: June, 2024]. Available online https://www.kaggle.com/datasets/uraninjo/augmented-alzheimer-mri-dataset.
13. Gaussian Smoothing, School of Informatics, University of Edinburgh, [Online]. Available: https://homepages.inf.ed.ac.uk/rbf/HIPR2/gsmooth.htm. [Date of access: Aug. 15, 2024]
14. J. Wang and J. Chen, “Subpixel edge detection algorithm based on improved Gaussian fitting and Canny operatör”, Academic Journal of Computing & Information Science, vol. 5, Issue 7: 33-39, 2022.
15. K. Ramesh et al., “A Review of Medical Image Segmentation Algorithms”, EAI Endorsed Trans Perv Health Tech, vol. 7, no. 27, p. e6, 2021.
16. M. Tuceryan and A. K. Jain, “Texture Analysis”, Handbook of Pattern Recognition and Computer Vision, pp. 207-248,1998.
17. N. Iqbal, R. Mumtaz, et al., “Gray level co-occurrence matrix (GLCM) texture based crop classification using low altitude remote sensing platforms”, PeerJ. Computer science,7, e536, 2021.
18. GLCM Equations (2011) GLCM Equations Gray Level Co-occurence Matrix equations. 2011. [19 July 2024]
19. S. Abd El-Ghany, M. Elmogy, and A. A. Abd El-Aziz, "Computer-Aided Diagnosis System for Blood Diseases Using EfficientNet-B3 Based on a Dynamic Learning Algorithm" Diagnostics vol.13, no. 3: 404,2023.
20. A. Nafea, et al., “A Deep Learning Algorithm for Lung Cancer Detection Using EfficientNet-B3”, Wasit Journal of Computer and Mathematics Science, vol. 2, no.4, pp. 68-76, 2023.
21. A. Batool and Y. Byun,“Lightweight EfficientNetB3 Model Based on Depthwise Separable Convolutions for Enhancing Classification of Leukemia White Blood Cell Images”,IEEE Access.vol. 11, pp. 37203-37215, 2023.
22. X. Ying, “An Overview of Overfitting and its Solutions”, Journal of Physics: Conference Series, vol.1168, Issue 2, pp. 022022, 2019.