Aprendizado Profundo na Classificação de Lesões Crescentes Glomerulares: modelos e condições

• Joacy Mesquita da Silva UEFS
• Michele Fulvia Angelo UEFS
• Washington L. C. dos Santos FIOCRUZ
• Angelo C. Loula UEFS

DOI: https://doi.org/10.5753/sibgrapi.est.2021.20031

Abstract

Glomeruli are structures in the kidneys, responsible for filtering the blood, that can be affected by several lesions, such as the glomerular crescent, which is characterized by abnormal cell proliferation. In this work, different models and conditions for the application of deep learning are to evaluated in the task of classifying glomerular crescent histopathological images. The pre-trained networks Xception, InceptionV3, MobileNet, VGG16 and ResNet50 were compared, by applying to the classification of images with crescent vs normal glomeruli. Comparing the accuracy, precision, recall and f1-score of the models, the ResNet50 showed significantly better performance than the other networks, in all measures. The application of data augmentation did not result in a significant improvement in the results in this case. In an experiment of classification of crescent vs non-crescent glomeruli, adding images of three other lesions to the database, the application of Focal Loss presented greater accuracy and precision.

References

P. Chagas, L. Souza, I. Araujo, N. Aldeman, A. Duarte, M. Angelo, W. L. dos Santos, and L. Oliveira, Classification of glomerular hypercellularity using convolutional features and support vector machine, Artificial Intelligence in Medicine, vol. 103, p. 101808, Mar. 2020.

G. Bueno, M. M. Fernandez-Carrobles, L. Gonzalez-Lopez, and O. Deniz, Glomerulosclerosis identification in whole slide images using semantic segmentation, Computer Methods and Programs in Biomedicine, vol. 184, p. 105273, Feb. 2020.

A. B. Fogo and M. Kashgarian, Diagnostic atlas of renal pathology, 3rd ed. Philadelphia, PA: Elsevier, 2017.

L. Anguiano, R. Kain, and H.-J. Anders, The glomerular crescent: triggers, evolution, resolution, and implications for therapy, Current Opinion in Nephrology and Hypertension, vol. 29, no. 3, p. 302, 2020.

G. O. Barros, B. Navarro, A. Duarte, and W. L. C. dos Santos, PathoSpotter-k: A computational tool for the automatic identification of glomerular lesions in histological images of kidneys, Scientific Reports, vol. 7, no. 1, Apr. 2017.

J. N. Marsh, M. K. Matlock, S. Kudose, T.-C. Liu, T. S. Stappenbeck, J. P. Gaut, and S. J. Swamidass, Deep learning global glomerulosclerosis in transplant kidney frozen sections, IEEE Transactions on Medical Imaging, vol. 37, no. 12, pp. 27182728, Dec. 2018.

B. Ginley, B. Lutnick, K.-Y. Jen, A. B. Fogo, S. Jain, A. Rosenberg, V. Walavalkar, G. Wilding, J. E. Tomaszewski, R. Yacoub, G. M. Rossi, and P. Sarder, Computational segmentation and classification of diabetic glomerulosclerosis, Journal of the American Society of Nephrology, vol. 30, no. 10, pp. 19531967, Sep. 2019.

E. Uchino, K. Suzuki, N. Sato, R. Kojima, Y. Tamada, S. Hiragi, H. Yokoi, N. Yugami, S. Minamiguchi, H. Haga, M. Yanagita, and Y. Okuno, Classification of glomerular pathological findings using deep learning and nephrologistAI collective intelligence approach, International Journal of Medical Informatics, vol. 141, p. 104231, Sep. 2020.

P. Chagas, L. Souza, I. Pontes, R. Calumby, M. Angelo, A. Duarte, W. dos Santos, and L. Oliveira, Deep-learning-based membranous nephropathy classification and monte-carlo dropout uncertainty estimation, in Anais do XXI Simposio Brasileiro de Computac ao Aplicada à Saude. Porto Alegre, RS, Brasil: SBC, 2021, pp. 257268.

J. M. C. Rehem, W. L. C. dos Santos, A. A. Duarte, L. R. de Oliveira, and M. F. Angelo, Automatic glomerulus detection in renal histological images, in Medical Imaging 2021: Digital Pathology, J. E. Tomaszewski and A. D. Ward, Eds. SPIE, Feb. 2021.

J. M. Johnson and T. M. Khoshgoftaar, Survey on deep learning with class imbalance, Journal of Big Data, vol. 6, no. 1, pp. 154, 2019.

T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollar, Focal loss for dense object detection, in Proceedings of the IEEE International Conference on Computer Vision (ICCV), Oct 2017.

X. Zhu and A. B. Goldberg, Introduction to semi-supervised learning, Synthesis Lectures on Artificial Intelligence and Machine Learning, vol. 3, no. 1, pp. 1130, Jan. 2009.

P. Linardatos, V. Papastefanopoulos, and S. Kotsiantis, Explainable AI: A Review of Machine Learning Interpretability Methods, Entropy, vol. 23, no. 1, p. 18, Dec. 2020.