Glomeruli Biyopsi Görüntülerinin Görme Dönüştürücüleri ve İstatistiksel Özellik Artırma Kullanılarak Hibrit Özellik Tabanlı Sınıflandırılması

Abstract

Glomeruloskleroz gibi anormalliklerin tanımlanması, böbrek hastalıklarının tanısında kullanılan glomeruli biyopsi çalışmasının en önemli yönlerinden biridir. Glomeruli biyopsi görüntülerini Normal ve Sklerozlu kategorilerine sınıflandırmak amacıyla, bu çalışma hibrit bir sınıflandırma sistemi uygular. Kaggle'dan elde edilen veri seti, herhangi bir ek eğitim gerektirmeden özellik çıkarma amacıyla Vision Transformers (ViTs) ile işlendi. Daha spesifik olmak gerekirse, ilk olarak eğitilen Vision Transformer modelinin baş katmanından bin derin özellik çıkarıldı. Sınıflandırmanın etkinliğini artırmak için, ortalama, minimum, maksimum, entropi, basıklık, çarpıklık ve ortalama karekökü içeren on iki istatistiksel özellik hesaplandı ve alınan derin özelliklere eklendi. Bu, 1.012 özellik içeren hibrit bir gösterimle sonuçlandı. Sonraki adımda, görüntü sınıflandırması amacıyla geleneksel makine öğrenimi sınıflandırıcıları kullanıldı. Bu sınıflandırıcıların performansının değerlendirilmesi ve karşılaştırılması, istatistiksel özelliklerin kullanılmasıyla elde edilen iyileştirmeye özel bir vurgu yapılarak gerçekleştirildi. Deneylerin bulguları, geliştirilen hibrit modelin doğruluk ve dayanıklılık açısından temel derin özelliklerden daha iyi performans gösterdiğini göstermektedir. Bu, hibrit modelin glomeruli biyopsi görüntülerinin sınıflandırılması için umut verici bir teknik olduğunu göstermektedir.

Keywords

• Glomeruli Biyopsisi
• Görüntü Sınıflandırması
• Görme Dönüştürücüler
• İstatistiksel Özellikler
• Hibrit Model

Hybrid Feature-Based Classification of Glomeruli Biopsy Images Using Vision Transformers and Statistical Feature Augmentation

Abstract

The identification of abnormalities such as glomerulosclerosis is one of the most important aspects of the glomeruli biopsy study that is used in the diagnosis of kidney illnesses. For the purpose of classifying glomeruli biopsy images into Normal and Sclerosed categories, this work implements a hybrid classification system. The dataset, which was obtained from Kaggle, was processed with Vision Transformers (ViTs) for the purpose of feature extraction without any additional training being required. To be more specific, one thousand deep features were extracted from the head layer of the Vision Transformer model that had been first trained. In order to improve the effectiveness of classification, twelve statistical characteristics, which included mean, minimum, maximum, entropy, kurtosis, skewness, and root mean square, were computed and added to the deep features that were retrieved. This resulted in a hybrid representation that contained 1,012 features. In the subsequent step, traditional machine learning classifiers were utilized for the purpose of image classification. Evaluation and comparison of the performance of these classifiers were carried out, with a particular emphasis placed on the enhancement that was accomplished by using statistical characteristics. The findings of the experiments show that the hybrid model that was developed performs better than the baseline deep features in terms of accuracy and resilience. This indicates that the hybrid model is a promising technique for the classification of glomeruli biopsy images.

Keywords

• Glomeruli Biopsy
• Image Classification
• Vision Transformers
• Statistical Features
• Hybrid Model

References

1. Abdel-Nabi H, Ali M, Awajan A, Daoud M, Alazrai R, Suganthan PN, et al. A comprehensive review of the deep learning-based tumor analysis approaches in histopathological images: segmentation, classification and multi-learning tasks. Cluster Comput. 2023;26(5):3145-85. doi:10.1007/s10586-023-03769-4.
2. Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. Proc Int Conf Learn Representations (ICLR). 2015. Available from: https://arxiv.org/abs/1409.1556.
3. Litjens G, et al. A survey on deep learning in medical image analysis. Med Image Anal. 2017;42:60-88. doi:10.1016/j.media.2017.07.005.
4. Ronneberger O, Fischer P, Brox T. U-Net: Convolutional networks for biomedical image segmentation. Proc MICCAI. 2015;234-41. doi:10.1007/978-3-319-24574-4_28.
5. Ioffe S, Szegedy C. Batch normalization: Accelerating deep network training by reducing internal covariate shift. Proc Int Conf Mach Learn (ICML). 2015;448-56. Available from: https://arxiv.org/abs/1502.03167.
6. Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Adv Neural Inf Process Syst (NIPS). 2012;1097-105. doi:10.1145/2999134.2999273.
7. Szegedy C, et al. Going deeper with convolutions. Proc IEEE Conf Comput Vis Pattern Recognit (CVPR). 2015;1-9. doi:10.1109/CVPR.2015.7298594.
8. Dosovitskiy A, et al. An image is worth 16x16 words: Transformers for image recognition at scale. Proc Int Conf Learn Representations (ICLR). 2021. Available from: https://arxiv.org/abs/2103.14030.

9. Liu X, et al. Swin transformer: Hierarchical vision transformer using shifted windows. Proc ICCV. 2021;10012-22. doi:10.1109/ICCV48922.2021.00985.
10. Sriwastawa A, Arul Jothi JA. Vision transformer and its variants for image classification in digital breast cancer histopathology: A comparative study. Multimed Tools Appl. 2024;83(13):39731-53. doi:10.1007/s11042-023-15564-x.
11. Kaur B, Goyal B, Dogra A. A hybrid feature-based model development for computer-aided diagnosis of lung cancer. Proc 2023 10th Int Conf Comput Sustainable Global Dev (INDIACom). 2023;1031-6. doi:10.1109/INDIACom59655.2023.10250090.
12. Dong G, Liu H, editors. Feature engineering for machine learning and data analytics. CRC Press; 2018. ISBN: 9781498760078.
13. Gupta S, Gupta S. Feature extraction and feature selection procedures for medical image analysis. In: Computer-Assisted Analysis for Digital Medicinal Imagery. IGI Global; 2025;(221)80. doi:10.4018/978-1-7998-3654-6.ch011.
14. Ozdemir B, Aslan E, Pacal I. Attention enhanced InceptionNeXt based hybrid deep learning model for lung cancer detection. IEEE Access. 2025. doi:10.1109/ACCESS.2025.1234567.
15. Kaggle. Glomeruli biopsy image dataset. Available from URL: https://www.kaggle.com/datasets/sachinkumarsaxena/glomeruli-biopsy-dataset, Last Access Date: 17.12.2024.
16. Cortes C, Vapnik V. Support-vector networks. Mach Learn. 1995;20(3):273-97. doi:10.1007/BF00994018.
17. Fogaing IM, Abdo A, Ballis-Berthiot P, Adrian-Felix S, Olagne J, Merieux R, et al. Detection and classification of glomerular lesions in kidney graft biopsies using a 2-stage deep learning approach. Medicine. 2025;104(7):e41560. doi:10.1097/MD.0000000000041560.
18. Celard P, Iglesias EL, Sorribes-Fdez JM, Romero R, Vieira AS, Borrajo L. A survey on deep learning applied to medical images: from simple artificial neural networks to generative models. Neural Comput Appl. 2023;35(3):2291-323. doi:10.1007/s00542-022-06634-x.
19. Zhang Z, et al. Swin Transformer for histopathological image analysis. Biomed Signal Process Control. 2022;72:103265. doi:10.1016/j.bspc.2021.103265.
20. Nguyen DK, Assran M, Jain U, Oswald MR, Snoek CG, Chen X. An image is worth more than 16x16 patches: Exploring transformers on individual pixels. arXiv preprint arXiv:2406.09415. 2024. Available from: https://arxiv.org/abs/2406.09415.
21. Yadav SP, Yadav S. Image fusion using hybrid methods in multimodality medical images. Med Biol Eng Comput. 2020;58(4):669-87. doi:10.1007/s11517-020-02266-3.
22. Cootes TF, Taylor CJ. Anatomical statistical models and their role in feature extraction. Br J Radiol. 2004;77(Suppl 2):S133-9. doi:10.1259/bjr/24653832.
23. Zheng A, Casari A. Feature engineering for machine learning: principles and techniques for data scientists. O’Reilly Media, Inc.; 2018. ISBN: 978-1491953243.
24. Ravi S, Ramachandran S. Hybrid deep learning models for medical diagnosis. J Artif Intell Soft Comput Res. 2021;11(1):45-60. doi:10.22055/jaiscr.2021.33498.1321.
25. Breiman L. Random forests. Mach Learn. 2001;45(1):5-32. doi:10.1023/A:1010933404324.
26. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Proc IEEE CVPR. 2016;770-8. doi:10.1109/CVPR.2016.90.
27. Tian, C., Chen, Y., Liu, Y., Wang, X., Lv, Q., Li, Y., ... & Li, W. Accurate classification of glomerular diseases by hyperspectral imaging and transformer. Computer Methods and Programs in Biomedicine, 2024;254, 108285.
28. Liu, Y. A hybrid CNN-transXNet approach for advanced glomerular segmentation in renal histology imaging. International Journal of Computational Intelligence Systems, 2024;17(1), 126.
29. Yin, Y., Tang, Z., & Weng, H. Application of visual transformer in renal image analysis. BioMedical Engineering OnLine, 2024;23(1), 27.
30. Santos, J. D., de MS Veras, R., Silva, R. R., Aldeman, N. L., Araújo, F. H., Duarte, A. A., & Tavares, J. M. R. A hybrid of deep and textural features to differentiate glomerulosclerosis and minimal change disease from glomerulus biopsy images. Biomedical Signal Processing and Control, 2021;70, 103020.
31. Tian, R., Liu, D., Bai, Y., Jin, Y., Wan, G., & Guo, Y. Swin-MSP: A shifted windows masked spectral pretraining model for hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing. 2024.
32. Yin, Y., Tang, Z., & Weng, H. Application of visual transformer in renal image analysis. BioMedical Engineering OnLine, 2024;23(1), 27.
33. Liu, Y. A hybrid CNN-transXNet approach for advanced glomerular segmentation in renal histology imaging. International Journal of Computational Intelligence Systems, 2024;17(1), 126.
34. Santos, J. D., de MS Veras, R., Silva, R. R., Aldeman, N. L., Araújo, F. H., Duarte, A. A., & Tavares, J. M. R. A hybrid of deep and textural features to differentiate glomerulosclerosis and minimal change disease from glomerulus biopsy images. Biomedical Signal Processing and Control, 2021;70, 103020.