Segmentación de núcleos celulares en imágenes histopalógicas mediante aprendizaje profundo. / Nuclei segmentation in histopathological images using deep learning.

Ziemecki Burgos, Andrés A. (2019) Segmentación de núcleos celulares en imágenes histopalógicas mediante aprendizaje profundo. / Nuclei segmentation in histopathological images using deep learning. Integration Project in Mechanical Engineering, Universidad Nacional de Cuyo, Instituto Balseiro.

PDF (Tesis)

Abstract in Spanish

El cáncer es una de las enfermedades mas frecuentes y graves que causa una de las mayores muertes al año. El diagnostico precoz y la discriminación de esta enfermedad son clínicamente importantes. Por otro lado, el análisis de imágenes y diagnostico asistido por computadora se han vuelto cada vez mas accesibles para los especialistas, con los que les permiten realizar diagnósticos de manera mas eciente. Pero, en muchos métodos computacionales, la parte del proceso que conlleva la detección, segmentación y clasificación de células en imágenes histopatológicas sigue siendo un problema principal. Dado que la detección celular y la segmentación son críticas para los diagnósticos del cáncer, en respuesta, en este trabajo se tuvo como objetivo proveer el soporte tecnológico para los especialistas que les permita coadyuvar a la detección y clasificación del cáncer. El mismo se realizo mediante un enfoque de segmentación celular útil de imágenes histopatológicas haciendo uso de técnicas de aprendizaje profundo prominentes como las redes neuronales convolucionales. Los resultados revelaron que, para un tratamiento de imágenes adecuado y una arquitectura de red neuronal convolucional apropiada, los algoritmos de aprendizaje profundo se desempeñaron mejor que los métodos convencionales en la segmentación celular llegando a obtener resultados con una precisión, a través del índice de Dice, de 88,2 %. Además, el método propuesto del presente trabajo resulto tener un mejor desempeño que los presentados por la bibliografía y por publicaciones que han hecho uso de dichas técnicas de aprendizaje profundo empleando enfoques similares al nuestro.

Abstract in English

Cancer is one of the most frequent and serious diseases that causes one of the biggest deaths a year. Early diagnosis and discrimination of this disease are clinically important. On the other hand, image analysis and computer-aided diagnosis have become increasingly accessible to specialists, allowing them to make diagnoses more efficiently. But, in many computational methods, the part of the process that involves the detection, segmentation and classication of cells in histopathological images remains a major problem. Given that cell detection and segmentation are critical for cancer diagnoses, in response, the objective of this study was to provide technological support for specialists that would allow them to contribute to the detection and classication of cancer. It was performed using a useful cell segmentation approach of histopathological images through the use of prominent deep learning techniques such as convolutional neural networks. The results revealed that, for an adequate image processing and an appropriate convolutional neural network architecture, the deep learning algorithms performed better than the conventional methods in cell segmentation reaching results with a precision, through the Dice index, of 88.2 %. In addition, the proposed method of this work turned out to have a better performance than those presented by the bibliography and by publications that have made use of such deep learning techniques using approaches similar to ours.

Item Type:Thesis (Integration Project in Mechanical Engineering)
Keywords:Artificial intelligence; Inteligencia artificial; Images; Imágenes; Automation; Automatización; [Deep learning; Aprendizaje profundo]
References:[1] Murray, C. J., Lopez, A. D. Mortality by cause for eight regions of the world: Global burden of disease study. The Lancet, 349 (9061), 1269-1276, mayo 1997. URL 1 [2] Gurcan, M., Boucheron, L., Can, A., Madabhushi, A., Rajpoot, N., Yener, B. Histopathological image analysis: A review. IEEE Reviews in Biomedical Engineering, 2, 147-171, 2009. URL 1, 2 [3] Wienert, S., Heim, D., Saeger, K., Stenzinger, A., Beil, M., Hufnagl, P., et al. Detection and segmentation of cell nuclei in virtual microscopy images: A minimummodel approach. Scientic Reports, 2 (1), jul. 2012. URL 1 [4] Demir, C., Yener, B. Automated cancer diagnosis based on histopathological images: A systematic survey, 11 2004. 1 [5] Jothi, J. A. A., Rajam, V. M. A. A survey on automated cancer diagnosis from histopathology images. Articial Intelligence Review, 48 (1), 31-81, jul. 2016. URL 1 [6] Veta, M., Pluim, J. P. W., van Diest, P. J., Viergever, M. A. Breast cancer histopathology image analysis: A review. IEEE Transactions on Biomedical Engineering, 61 (5), 1400-1411, mayo 2014. URL 2014.2303852. 1, 2 [7] Veta, M., van Diest, P. J., Kornegoor, R., Huisman, A., Viergever, M. A., Pluim, J. P. W. Automatic nuclei segmentation in h&e stained breast cancer histopathology images. PLoS ONE, 8 (7), e70221, jul. 2013. URL 2, 3 [8] Dundar, M. M., Badve, S., Bilgin, G., Raykar, V., Jain, R., Sertel, O., et al. Computerized classication of intraductal breast lesions using histopathological images. IEEE Transactions on Biomedical Engineering, 58 (7), 1977-1984, jul. 2011. URL 2 [9] Otsu, N. A threshold selection method from gray-level histograms. IEEE Transactions on Systems, Man, and Cybernetics, 9 (1), 62-66, ene. 1979. URL 3 [10] Filipczuk, P., Kowal, M., Obuchowicz, A. Automatic breast cancer diagnosis based on k-means clustering and adaptive thresholding hybrid segmentation. En: Advances in Intelligent and Soft Computing, pags. 295-302. Springer Berlin Heidelberg, 2011. URL 3 [11] Rother, C., Kolmogorov, V., Blake, A. "GrabCut". En: ACM SIGGRAPH 2004 Papers on - SIGGRAPH 04. ACM Press, 2004. URL 1186562.1015720. 3 [12] Al-Kofahi, Y., Lassoued, W., Lee, W., Roysam, B. Improved automatic detection and segmentation of cell nuclei in histopathology images. IEEE Transactions on Biomedical Engineering, 57 (4), 841-852, abr. 2010. URL 3 [13] Mouelhi, A., Sayadi, M., Fnaiech, F., Mrad, K., Romdhane, K. B. Automatic image segmentation of nuclear stained breast tissue sections using color active contour model and an improved watershed method. Biomedical Signal Processing and Control, 8 (5), 421-436, sep. 2013. URL 4 [14] Xu, J., Xiang, L., Liu, Q., Gilmore, H., Wu, J., Tang, J., et al. Stacked sparse autoencoder (SSAE) for nuclei detection on breast cancer histopathology images. IEEE Transactions on Medical Imaging, 35 (1), 119-130, ene. 2016. URL [15] Sirinukunwattana, K., Raza, S. E. A., Tsang, Y.-W., Snead, D. R. J., Cree, I. A., Rajpoot, N. M. Locality sensitive deep learning for detection and classication of nuclei in routine colon cancer histology images. IEEE Transactions on Medical Imaging, 35 (5), 1196-1206, mayo 2016. URL 4 [16] Deng, L. Deep learning: Methods and applications. Foundations and Trends R in Signal Processing, 7 (3-4), 197-387, 2014. URL 4 [17] Curiale, A. H., Colavecchia, F. D., Kaluza, P., Isoardi, R. A., Mato, G. Automatic myocardial segmentation by using a deep learning network in cardiac MRI. En: 2017 XLIII Latin American Computer Conference (CLEI). IEEE, 2017. URL 4, 28, 34 [18] Hatipoglu, N., Bilgin, G. Cell segmentation in histopathological images with deep learning algorithms by utilizing spatial relationships. Medical & Biological Engineering & Computing, 55 (10), 1829-1848, feb. 2017. URL https: // 33 [19] Janowczyk, A., Madabhushi, A. Deep learning for digital pathology image analysis: A comprehensive tutorial with selected use cases. Journal of Pathology Informa-tics, 7 (1), 29, 2016. URL 29,44, 48 [20] Yuxin Cui, Z. L. Z. X. J. H., Guiying Zhang. A deel pearning algorithm for one-step contour aware nuclei segmentation of histopathological images. IEEE Department of Computer Science and Engineering, University of South Carolina, Columbia, SC 29208, 2018. 4, 33, 34 [21] Samarasinghe, S. Neural network for applied sciences and engineering: From fundamentals to complex pattern recognition. tomo 65, pags. 11-27. 2006. 7 [22] MCCULLOCH, W., PITTS, W. A logical calculus of the ideas immanent in nervous activity. Bulletin of Mathematical Biology, 52 (1-2), 99-115, 1990. URL 8 [23] Rosenblatt, F. The perceptron: A probabilistic model for information storage and organization in the brain. Psychological Review, 65 (6), 386-408, 1958. URL 8 [24] John Hertz, R. G. P., Anders Krogh. Introduction to theory and neural computation. tomo 1, pags. 97-100. Santa Fe Institute Studies in the Sciences of Complexity, 1991. 9 [25] Murphy, C., Gray, P., Stewart, G. Veried perceptron convergence theorem. En: Proceedings of the 1st ACM SIGPLAN International Workshop on Machine Learning and Programming Languages - MAPL 2017. ACM Press, 2017. URL 9 [26] Widrow, B., Hoff, M. E. Adaptive switching circuits. En: 1960 IRE WESCON Convention Record, Part 4, pags. 96-104. New York: IRE, 1960. 10 [27] Samarasinghe, S. Neural network for applied sciences and engineering: From fundamentals o complex pattern recognition. tomo 65, pags. 115-138. 2006. 10 [28] Leven, S. The roots of backpropagation: From ordered derivatives to neural networks and political forecasting. Neural Networks, 9 (3), 543-544, abr. 1996. URL 10 [29] Cybenko, G. Approximation by superpositions of a sigmoidal function. Mat- hematics of Control, Signals, and Systems, 2 (4), 303-314, dic. 1989. URL 10 [30] Samarasinghe, S. Neural network for applied sciences and engineering: From fundamentals to complex pattern recognition. tomo 65, pags. 437-555. 2006. 11 [31] Rumelhart, D. E., Hinton, G. E., Williams, R. J. Learning representations by back-propagating errors. Nature, 323 (6088), 533-536, oct. 1986. URL https: // 12, 14 [32] Hinton, G. E., Osindero, S., Teh, Y.-W. A fast learning algorithm for deep belief nets. Neural Computation, 18 (7), 1527-1554, jul. 2006. URL 14 [33] Lecun, Y., Bottou, L., Bengio, Y., Haffner, P. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86 (11), 2278-2324, 1998. URL 16 [34] Ragav Venkatesan, B. L. Convolutional Neural Networks in Visual Computing: A Concise Guide. Data-Enabled Engineering, 1a ed. CRC Press, 2018. URL 18 [35] Shelhamer, E., Long, J., Darrell, T. Fully convolutional networks for semantic segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39 (4), 640-651, abr. 2017. URL 19 [36] Krizhevsky, A., Sutskever, I., Hinton, G. E. ImageNet classication with deep convolutional neural networks. Communications of the ACM, 60 (6), 84-90, mayo 2017. URL 19 [37] Goodfellow, I., Bengio, Y., Courville, A. Deep Learning. MIT Press, 2016. 20 [38] Sabour, S., Frosst, N., Hinton, G. E. Dynamic routing between capsules. En: I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, R. Garnett (eds.) Advances in Neural Information Processing Systems 30, pags. 3856-3866. Curran Associates, Inc., 2017. URL 6975-dynamic-routing-between-capsules.pdf. 20, 22, 42 [39] Payyavula, G. A collection of segmentation datasets used to achieve 10th place in dsb 2018. cell-imagesNucleiREADME:md; 82018:25 [40] Oktay, O., Ferrante, E., Kamnitsas, K., Heinrich, M., Bai, W., Caballero, J., et al. Anatomically constrained neural networks (ACNNs): Application to cardiac image enhancement and segmentation. IEEE Transactions on Medical Imaging, 37 (2), 384-395, feb. 2018. URL 26, 44 [41] Ronneberger, O., Fischer, P., Brox, T. U-net: Convolutional networks for biomedical image segmentation. En: Lecture Notes in Computer Science, pags. 234-241. Springer International Publishing, 2015. URL 26, 33, 35 [42] Hochreiter, S. The vanishing gradient problem during learning recurrent neural nets and problem solutions. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 06 (02), 107-116, abr. 1998. URL 28 [43] Hand, D., Christen, P. A note on using the f-measure for evaluating record linkage algorithms. Statistics and Computing, 28 (3), 539-547, abr. 2017. URL 28 [44] Naylor, P., Lae, M., Reyal, F., Walter, T. Nuclei segmentation in histopathology images using deep neural networks. En: 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017). IEEE, 2017. URL 33 [45] Ronneberger, O., Fischer, P., Brox, T. U-net: Convolutional networks for biomedical image segmentation. En: Lecture Notes in Computer Science, pags. 234-241. Springer International Publishing, 2015. URL 34
Subjects:Engineering > Inteligencia artificial
Divisions:Gcia. de área de Investigación y aplicaciones no nucleares > Gcia. de Física > Física médica
ID Code:834
Deposited By:Tamara Cárcamo
Deposited On:15 Mar 2021 12:26
Last Modified:15 Mar 2021 12:26

Repository Staff Only: item control page