Application of geostatistical functions and deep features to kidney biopsy images to differentiate focal segmental glomerulosclerosis from minimal change disease

• Justino Duarte Santos UFPI
• Romuere R. V. Silva UFPI
• Rodrigo M. S. Veras UFPI

DOI: https://doi.org/10.5753/sibgrapi.est.2020.12984

Resumo

Chronic kidney diseases arise from acute or intermittent pathologies that have not been adequately treated, such as minimal change disease (MCD) and focal segmental glomerulosclerosis (FSGS). The accurate identification of these two diseases is of paramount importance, because their treatments and prognoses are different. Thus, we propose a method that is capable of differentiating MCD from FSGS based on images from pathological examinations. In the proposed method, we use four pre-trained convolutional neural networks and geostatistical functions to extract image features. Of the 8,720 extracted features, we selected 94 based on mutual information criteria, and in the classification step, we used a random forest classifier. The proposed method obtained an accuracy of 94.3% and Kappa index of 87.9%, a level that is regarded as “almost perfect”, confirming that our method is very promising.

Referências

D. M. d. N. Costa, L. M. Valente, P. A. d. C. Gouveia, F. W. Sarinho, G. V. Fernandes, M. A. G. d. M. Cavalcante, C. B. L. d. Oliveira, C. d. A. J. d. Vasconcelos, and E. S. C. Sarinho, "Comparative analysis Fig. 4. Origin of the 94 attributes with best mutual information for the v16+xce+rsnet+geo vector. The bar at the bottom depicts the source of the characteristics at each position. This article has proposed a computer vision method to differentiate between MCD and FSGS in microscopic images. The tests performed to define the final approach included feature extraction based on traditional texture features, i.e. Haralick features, geostatistical functions, and deep features from pretrained CNNs. These features were then selected using two criteria: mutual information and ANOVA-F. We then used these as input for three supervised classifiers: SVM, KNN, and RF. The results indicate that the use of a concatenated feature vector of VGG-16, Xception, ResNet50, and geostatistics gave the best scores, followed by feature selection based on mutual information and classification using a random forest classifier. This model gave results that indicate near-perfect agreement with the pathologist’s diagnosis. The characterization of the image using a single type of

Centers for Disease Control and Prevention, "Chronic kidney disease in the united states, 2019," 2019, gA: US Department of Health and Human Services, Centers for Disease Control and Prevention.

L. R. Moura, M. F. Franco, and G. M. Kirsztajn, "Minimal change disease and focal segmental glomerulosclerosis in adults: response to steroids and risk of renal failure," Brazilian Journal of Nephrology, vol. 37, no. 4, pp. 475–480, 2015.

Y. Zhao, E. F. Black, L. Marini, K. McHenry, N. Kenyon, R. Patil, A. Balla, and A. Bartholomew, "Automatic glomerulus extraction in whole slide images towards computer aided diagnosis," in 12th Interna- tional Conference on e-Science (e-Science). IEEE, 2016, pp. 165–174.

P. Sarder, B. Ginley, and J. E. Tomaszewski, "Automated renal histopathology: digital extraction and quantification of renal pathology," in Medical Imaging 2016: Digital Pathology, vol. 9791. International Society for Optics and Photonics, 2016, p. 97910F.

B. Ginley, J. E. Tomaszewski, and P. Sarder, "Automatic computational labeling of glomerular textural boundaries," in Medical Imaging 2017: Digital Pathology, vol. 10140. International Society for Optics and Photonics, 2017, p. 101400G.

G. O. Barros, B. Navarro, A. Duarte, and W. L. Dos-Santos, "Pathospotter-k: A computational tool for the automatic identification of glomerular lesions in histological images of kidneys," Scientific reports, vol. 7, p. 46769, 2017.

D. W. Aha, D. Kibler, and M. K. Albert, "Instance-based learning algorithms," Machine Learning, vol. 6, no. 1, pp. 37–66, Jan 1991.

I. C. d. Araújo, L. Schnitman, A. A. Duarte, and W. L. dos Santos, "Au- tomated detection of segmental glomerulosclerosis in kidney histopathol- ogy," in XIII Brazilian Congress on Computational Intelligence, 2017.

C. Cortes and V. Vapnik, "Support-vector networks," Machine Learning, vol. 20, no. 3, pp. 273–297, Sep 1995.

B. G. Ginley, J. E. Tomaszewski, K.-Y. Jen, A. Fogo, S. Jain, and P. Sarder, "Computational analysis of the structural progression of in diabetic nephropathy," in Proceedings of SPIE human glomeruli Medical Imaging, vol. 10581, 2018, pp. 105 810A–1–105 810A–6.

S. M. Sheehan and R. Korstanje, "Automatic glomerular identification and quantification of histological phenotypes using image analysis and machine learning," American Journal of Physiology-Renal Physiology, vol. 315, no. 6, pp. F1644–F1651, 2018.

G. A. Noordmans, C. R. Caputo, Y. Huang, S. M. Sheehan, M. Bulthuis, P. Heeringa, J.-L. Hillebrands, H. van Goor, and R. Korstanje, "Genetic analysis of mesangial matrix expansion in aging mice and identification of far2 as a candidate gene," Journal of the American Society of Nephrology, vol. 24, no. 12, pp. 1995–2001, 2013.

T. K. Ho, "Random decision forests," in Proceedings of 3rd International Conference on Document Analysis and Recognition, vol. 1, Aug 1995, pp. 278–282.

J. N. Marsh, M. K. Matlock, S. Kudose, T. 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. 2718–2728, Dec 2018.

K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," arXiv preprint arXiv:1409.1556, 2014.

P. Chagas, L. Souza, I. Araújo, N. Aldeman, A. Duarte, M. Angelo, W. L. dos Santos, and L. Oliveira, "Classification of glomerular hy- percellularity using convolutional features and support vector machine," arXiv preprint arXiv:1907.00028, 2019.

S. Kannan, L. A. Morgan, B. Liang, M. G. Cheung, C. Q. Lin, D. Mun, R. G. Nader, M. E. Belghasem, J. M. Henderson, J. M. Francis et al., "Segmentation of glomeruli within trichrome images using deep learning," Kidney international reports, vol. 4, no. 7, pp. 955–962, 2019.

N. P. Pavinkurve, K. Natarajan, and A. J. Perotte, "Deep vision: learning to identify renal disease with neural networks," Kidney International Reports, vol. 4, no. 7, p. 914, 2019.

R. M. Haralick, K. Shanmugam et al., "Textural features for image classification," IEEE Transactions on systems, man, and cybernetics, vol. 3, no. 6, pp. 610–621, 1973.

F. Chollet, "Xception: Deep learning with depthwise separable convo- lutions," arXiv preprint, pp. 1610–02 357, 2017.

K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2015, pp. 770–778.

T. Ojala, M. Pietik¨ainen, and D. Harwood, "A comparative study of texture measures with classification based on featured distributions," Pattern Recognition, vol. 29, no. 1, pp. 51 – 59, 1996.

E. Isaaks, R. Srivastava, and K. (Firm), An introduction to Applied Geostatistics. Oxford University Press, 1989.

A. C. Silva, P. C. P. Carvalho, and M. Gattass, "Analysis of spatial variability using geostatistical functions for diagnosis of lung nodule in computerized tomography images," Pattern Analysis and Applications, vol. 7, no. 3, pp. 227–234, 2004.

J. A. de Sousa, A. C. de Paiva, J. D. S. de Almeida, A. C. Silva, G. B. Junior, and M. Gattass, "Texture based on geostatistic for glaucoma diagnosis from fundus eye image," Multimedia Tools and Applications, vol. 76, no. 18, pp. 19 173–19 190, 2017.

N. Tajbakhsh, J. Y. Shin, S. R. Gurudu, R. T. Hurst, C. B. Kendall, M. B. Gotway, and J. Liang, "Convolutional neural networks for medical image analysis: Full training or fine tuning?" IEEE transactions on medical imaging, vol. 35, no. 5, pp. 1299–1312, 2016.

G. Litjens, T. Kooi, B. E. Bejnordi, A. A. A. Setio, F. Ciompi, M. Ghafoorian, J. A. van der Laak, B. van Ginneken, and C. I. Sánchez, "A survey on deep learning in medical image analysis," Medical Image Analysis, vol. 42, pp. 60 – 88, 2017.

S. J. Pan, Q. Yang et al., "A survey on transfer learning," IEEE Transactions on knowledge and data engineering, vol. 22, no. 10, pp. 1345–1359, 2010.

O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A. Khosla, M. Bernstein, A. C. Berg, and L. Fei- Fei, "Imagenet large scale visual recognition challenge," International Journal of Computer Vision, vol. 115, no. 3, pp. 211–252, Dec 2015.

T. M. Cover and J. A. Thomas, "Entropy, relative entropy and mutual information," Elements of information theory, vol. 2, pp. 1–55, 1991.

J. Cohen, "A coefficient of agreement for nominal scales," Educational and psychological measurement, vol. 20, no. 1, pp. 37–46, 1960. of primary and secondary glomerulopathies in the northeast of Brazil: data from the Pernambuco registry of glomerulopathies-repeg," Brazilian Journal of Nephrology, vol. 39, no. 1, pp. 29–35, 2017.