Prostate histopathology image classification and retrieval using weakly-supervised multimodal fusion and representation learning

dc.contributor.advisorGonzález Osorio, Fabio Augustospa
dc.contributor.authorLara Ramírez, Juan Sebastiánspa
dc.contributor.researchgroupMindLabspa
dc.date.accessioned2020-12-07T22:48:51Zspa
dc.date.available2020-12-07T22:48:51Zspa
dc.date.issued2020-10-30spa
dc.description.abstractThis thesis presents an information fusion strategy for the automatic classification and retrieval of prostate histopathology whole-slide images (WSIs) that incorporates novel machine learning components from deep learning and kernel methods. Its main purpose is to enhance the representation of the WSIs using additional text content extracted from diagnosis reports. This is achieved using the multimodal latent semantic alignment (M-LSA) model, which employs a weakly-multimodal-supervised methodology that incorporates text information during the model training to enrich the representation of the WSIs with complementary semantic information. Besides, M-LSA does not require the text data during the prediction phase, which makes it suitable for realistic scenarios where a pathologist may only have the image data. The experimental evaluation demonstrates that the weakly-supervised multimodal enhancement has a significant improvement in the performance during classification and retrieval, further, the proposed model outperforms the state--of--the--art unimodal and multimodal baselines in automatic prostate cancer assessment.spa
dc.description.abstractEsta tesis presenta una estrategia de fusión de información para la clasificación y recuperación automática de imágenes de histopatología de próstata incorporando novedosos compenentes de aprendizaje de máquina y aprendizaje profundo. El propósito de la estrategia es mejorar la representación de las imágenes con contenido textual adicional que es extraído de reportes de diagnóstico. Para lograr esto, se propone el modelo multimodal latent semantic alignment (M-LSA), el cual emplea una metodología de supervisión multimodal débil que incorpora información textual durante el entrenamiento para enriquecer la representación de las imágenes con información semántica complementaria. Adicionalmente, M-LSA no requiere la modalidad textual durante la fase de predicción, por lo que el modelo es apropiado para escenarios más realistas donde un patólogo puede tener sólo las imágenes. La evaluación experimental muestra que el enriquecimiento por supervisión débil multimodal presenta una mejora significativa en el despempeño durante clasificación y recuperación, además, el método propuesto supera otros enfoques unimodales y multimodales en el estado del arte del análisis automático de cáncer de próstata.spa
dc.description.additionalResearch Area: Machine Learningspa
dc.description.degreelevelMaestríaspa
dc.format.extent55spa
dc.format.mimetypeapplication/pdfspa
dc.identifier.citationLara Ramirez, J. S. (2020). Prostate Histopathology Image Classification and Retrieval using Weakly-Supervised Multimodal Fusion and Representation Learning. Universidad Nacional de Colombia.spa
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/78687
dc.language.isoengspa
dc.publisher.branchUniversidad Nacional de Colombia - Sede Bogotáspa
dc.publisher.programBogotá - Ingeniería - Maestría en Ingeniería - Ingeniería de Sistemas y Computaciónspa
dc.relation.referencesWCRF, “Worldwide cancer data. Global cancer statistics for the most common cancers,” 2018spa
dc.relation.referencesPCEC, “Gleason Score, Prostate Cancer Grading & Prognostic Scoring,” 2020spa
dc.relation.referencesJ. C. Caicedo, J. A. Vanegas, F. P´aez, and F. A. Gonz´alez, “Histology image search using multimodal fusion,” Journal of Biomedical Informatics, vol. 51, pp. 114–128, 2014spa
dc.relation.referencesP. W. Hamilton, P. Bankhead, Y. Wang, R. Hutchinson, D. Kieran, D. G. McArt, J. James, and M. Salto-Tellez, “Digital pathology and image analysis in tissue biomarker research,” Methods, vol. 70, no. 1, pp. 59–73, 2014spa
dc.relation.referencesS. Al-Janabi, A. Huisman, and P. J. Van Diest, “Digital pathology: current status and future perspectives,” Histopathology, vol. 61, no. 1, pp. 1–9, 2012spa
dc.relation.referencesD. Komura and S. Ishikawa, “Machine Learning Methods for Histopathological Image Analysis,” Computational and Structural Biotechnology Journal, vol. 16, pp. 34–42, 2018.spa
dc.relation.referencesD. Komura and S. Ishikawa, “Machine Learning Methods for Histopathological Image Analysis,” Computational and Structural Biotechnology Journal, vol. 16, pp. 34–42, 2018.spa
dc.relation.referencesD. Wang, A. Khosla, R. Gargeya, H. Irshad, and A. H. Beck, “Deep Learning for Identifying Metastatic Breast Cancer,” in Internacional Conference on Learning Representations, pp. 1–6, 2016.spa
dc.relation.referencesO. Jiménez del Toro and et al., “Deep Multimodal Case - Based Retrieval for Large Histopathology Datasets,” in MICCAI 2017 workshop on Patch-based image analysis, vol. 2, pp. 149–157, 2017.spa
dc.relation.referencesJ. Arevalo and et al., “Gated multimodal networks,” Neural Computing and Applications, vol. 1, 2020.spa
dc.relation.referencesY. Cao, S. Steffey, H. Jianbiao, D. Xiao, C. Tao, P. Chen, and H. M¨uller, “Medical Image Retrieval: A Multimodal Approach,” Cancer Informatics, vol. 13, pp. 125–136, 2014.spa
dc.relation.referencesA. Mourao, F. Martins, and J. a. Magalhaes, “Multimodal medical information retrieval with unsupervised rank fusion,” Computerized Medical Imaging and Graphics, vol. 39, pp. 35–45, 2015.spa
dc.relation.referencesA. G. Seco de Herrera, R. Schaer, D. Markonis, and H. Müller, “Comparing fusion techniques for the ImageCLEF 2013 medical case retrieval task,” Computerized Medical Imaging and Graphics, vol. 39, no. Medical visual information analysis and retrieval, pp. 46–54, 2015.spa
dc.relation.referencesM. N. Gurcan, L. E. Boucheron, A. Can, A. Madabhushi, N. M. Rajpoot, and B. Yener, “Histopathological Image Analysis: A Review,” IEEE Reviews in Biomedical Engineering, vol. 2, pp. 147–171, 2009.spa
dc.relation.referencesC. Mosquera-Lopez, S. Agaian, A. Velez-Hoyos, and I. Thompson, “Computer-Aided Prostate Cancer Diagnosis from Digitized Histopathology: A Review on Texture-Based Systems,” IEEE Reviews in Biomedical Engineering, vol. 8, pp. 98–113, 2015.spa
dc.relation.referencesJ. Ren, I. Hacihaliloglu, E. A. Singer, D. J. Foran, and X. Qi, “Unsupervised Domain Adaptation for Classification of Histopathology Whole-Slide Images,” Frontiers in Bioengineering and Biotechnology, vol. 7, 2019.spa
dc.relation.referencesO. Eminaga, M. Abbas, C. Kunder, A. M. Loening, J. Shen, D. Brooks, C. P. Langlotz, and D. L. Rubin, “Plexus Convolutional Neural Network ( PlexusNet ): A novel neural network architecture for histologic image analysis,” arXiv:1908.09067 [q-bio.QM], 2019.spa
dc.relation.referencesH. Xu, S. Park, and T. H. Hwang, “Computerized Classification of Prostate Cancer Gleason Scores from Whole Slide Images,” IEEE/ACM Transactions on Computational Biology and Bioinformatics, vol. PP, no. X, pp. 1–1, 2019.spa
dc.relation.referencesE. Esteban, A. Colomer, V. Naranjo, C. D. Vera, and U. D. Valencia, “GranulometryBased Descriptor for Pathological Tissue Discrimination in Histopathological Images,” in IEEE International Conference on Image Processing (ICIP), pp. 1413–1417, IEEE, 2018.spa
dc.relation.referencesO. Jiménez del Toro, M. Atzori, S. Otálora, M. Andersson, K. Eurén, M. Hedlund, P. Ronnquist, and H. Müller, “Convolutional neural networks for an automatic classification of prostate tissue slides with high-grade Gleason score,” in SPIE Medical Imaging, pp. 1–9, 2017.spa
dc.relation.referencesS. Bhattacharjee, H. G. Park, C. H. Kim, D. Prakash, N. Madusanka, J. H. So, N. H. Cho, and H. K. Choi, “Quantitative analysis of benign and malignant tumors in histopathology: Predicting prostate cancer grading using SVM,” Applied Sciences (Switzerland), vol. 9, no. 15, 2019.spa
dc.relation.referencesD. Karimi, G. Nir, L. Fazli, P. C. Black, L. Goldenberg, and S. E. Salcudean, “Deep Learning-Based Gleason Grading of Prostate Cancer from Histopathology Images - Role of Multiscale Decision Aggregation and Data Augmentation,” IEEE Journal of Biomedical and Health Informatics, pp. 1–14, 2019.spa
dc.relation.referencesD. Wang, D. J. Foran, J. Ren, H. Zhong, I. Y. Kim, and X. Qi, “Exploring automatic prostate histopathology image gleason grading via local structure modeling,” in International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, vol. 2015-Novem, pp. 2649–2652, 2015.spa
dc.relation.referencesP. W. Huang and C. H. Lee, “Automatic classification for pathological prostate images based on fractal analysis,” IEEE Transactions on Medical Imaging, vol. 28, no. 7, pp. 1037–1050, 2009.spa
dc.relation.referencesS. Doyle, M. Feldman, J. Tomaszewski, and A. Madabhushi, “A boosted Bayesian multiresolution classifier for prostate cancer detection from digitized needle biopsies,” IEEE Transactions on Biomedical Engineering, vol. 59, no. 5, pp. 1205–1218, 2012.spa
dc.relation.referencesP. e. a. Strom, “Artificial intelligence for diagnosis and grading of prostate cancer in biopsies: a population-based, diagnostic study.,” The Lancet. Oncology, vol. 2045, no. 19, pp. 1–11, 2020.spa
dc.relation.referencesK. e. a. Nagpal, “Development and validation of a deep learning algorithm for improving Gleason scoring of prostate cancer,” npj Digital Medicine, vol. 2, no. 1, 2019.spa
dc.relation.referencesH. Kallen, J. Molin, A. Heyden, C. Lundstr¨om, and ˚A. Kalle, “Towards grading gleason score using generically trained deep convolutional neural networks,” in IEEE 13th International Symposium on Biomedical Imaging (ISBI), pp. 1163–1167, 2016.spa
dc.relation.referencesL. Hou, D. Samaras, T. M. Kurc, Y. Gao, J. E. Davis, and J. H. Saltz, “Patch-Based Convolutional Neural Network for Whole Slide Tissue Image Classification,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2016-Decem, 2016.spa
dc.relation.referencesJ. Li, W. Li, A. Gertych, B. S. Knudsen, W. Speier, and C. W. Arnold, “An attentionbased multi-resolution model for prostate whole slide image classification and localization,” arXiv:1905.13208 [cs.CV], 2019.spa
dc.relation.referencesG. Saposnik, D. Redelmeier, C. C. Ruff, and P. N. Tobler, “Cognitive biases associated with medical decisions: a systematic review,” BMC Medical Informatics and Decision Making, vol. 16, no. 1, pp. 1–14, 2016.spa
dc.relation.referencesH. Müller and T. M. Deserno, “Content-Based Medical Image Retrieval,” in Biomedical Image Processing, ch. 19, Springer, 2011.spa
dc.relation.referencesR. Sparks and A. Madabhushi, “Out-of-Sample Extrapolation utilizing Semi-Supervised Manifold Learning (OSE-SSL): Content Based Image Retrieval for Histopathology Images.,” Scientific reports, vol. 6, p. 27306, jun 2016.spa
dc.relation.referencesA. Sridhar, S. Doyle, and A. Madabhushi, “Content-based image retrieval of digitized histopathology in boosted spectrally embedded spaces,” Journal of Pathology Informatics, vol. 6, no. 1, p. 41, 2015.spa
dc.relation.referencesA. Sridhar, S. Doyle, and A. Madabhshi, “Bossted spectral embedding (BoSE): applications to content-based image retrieval of histopathology,” in IEEE International Symposium on Biomedical Imaging, pp. 1897–1900, 2011.spa
dc.relation.referencesR. Sparks and A. Madabhushi, “Content-based image retrieval utilizing explicit shape descriptors: applications to breast MRI and prostate histopathology,” Medical Imaging 2011: Image Processing, vol. 7962, p. 79621I, 2011.spa
dc.relation.referencesN. Hegde, J. D. Hipp, Y. Liu, M. Emmert-Buck, E. Reif, D. Smilkov, M. Terry, C. J. Cai, M. B. Amin, C. H. Mermel, P. Q. Nelson, L. H. Peng, G. S. Corrado, and M. C. Stumpe, “Similar image search for histopathology: SMILY,” npj Digital Medicine, vol. 2, no. 1, pp. 1–9, 2019.spa
dc.relation.referencesR. Schaer, S. Otálora, O. Jimenez-del Toro, M. Atzori, and H. M¨uller, “Deep Learning - Based Retrieval System for Gigapixel Histopathology Cases and the Open Access Literature Roger,” Journal of Pathology Informatics, vol. 9, no. 1, 2019.spa
dc.relation.referencesR. J. Chen, M. Y. Lu, J. Wang, D. F. Williamson, S. J. Rodig, N. I. Lindeman, and F. Mahmood, “Pathomic fusion: An integrated framework for fusing histopathology and genomic features for cancer diagnosis and prognosis,” arXiv preprint arXiv:1912.08937, 2019.spa
dc.relation.referencesR. Yan, F. Ren, X. Rao, B. Shi, T. Xiang, L. Zhang, Y. Liu, J. Liang, C. Zheng, and F. Zhang, “Integration of multimodal data for breast cancer classification using a hybrid deep learning method,” in International Conference on Intelligent Computing, pp. 460–469, Springer, 2019.spa
dc.relation.referencesW.-H. Weng, Y. Cai, A. Lin, F. Tan, and P.-H. C. Chen, “Multimodal multitask representation learning for pathology biobank metadata prediction,” arXiv preprint arXiv:1909.07846, 2019.spa
dc.relation.referencesJ. A. Vanegas, J. C. Caicedo, F. A. Gonz´alez, and E. Romero, “Histology image indexing using a non-negative semantic embedding,” in MICCAI International Workshop on Medical Content-Based Retrieval for Clinical Decision Support, pp. 80–91, Springer, 2011.spa
dc.relation.referencesA. Cheerla and O. Gevaert, “Deep learning with multimodal representation for pancancer prognosis prediction,” Bioinformatics, vol. 35, no. 14, pp. i446–i454, 2019.spa
dc.relation.referencesV. H. Contreras, J. S. Lara, O. J. Perdomo, and F. A. Gonz´alez, “Supervised online matrix factorization for histopathological multimodal retrieval,” in 14th International Symposium on Medical Information Processing and Analysis, vol. 10975, p. 109750Y, International Society for Optics and Photonics, 2018.spa
dc.relation.referencesJ. S. Lara, V. H. Contreras O., S. Ot´alora, H. M¨uller, and F. A. Gonz´alez, “Multimodal latent semantic alignment for automated prostate tissue classification and retrieval,” in International Conferene on Medical Image Computing and Computer Assisted Intervention, MICCAI, 2020.spa
dc.relation.referencesH. Chang, L. Loss, and B. Parvin, “Nuclear segmentation in H&E sections via multireference graph cut (MRGC),” in International Symposium on Biomedical Imaging (ISBI), pp. 1–4, 2012.spa
dc.relation.referencesJ. A. Vanegas, Large-scale Non-linear Multimodal Semantic Embedding. Doctoral thesis, Universidad Nacional de Colombia, 2017.spa
dc.relation.referencesJ. A. Vanegas, H. J. Escalante, and F. A. Gonzalez, “Semi-supervised Online Kernel Semantic Embedding for Multi-label Annotation,” Lecture Notes in Computer Science, pp. 693–701, 2018.spa
dc.relation.referencesS. Otálora, M. Atzori, A. Khan, O. Jimenez-del Toro, V. Andrearczyk, and H. Müller, “A systematic comparison of deep learning strategies for weakly supervised gleason grading,” in Medical Imaging 2020: Digital Pathology, vol. 11320, p. 113200L, International Society for Optics and Photonics, 2020.spa
dc.relation.referencesP. A. Humphrey, “Gleason grading and prognostic factors in carcinoma of the prostate,” Modern pathology, vol. 17, no. 3, pp. 292–306, 2004.spa
dc.relation.referencesG. Sauter, S. Steurer, T. S. Clauditz, T. Krech, C. Wittmer, F. Lutz, M. Lennartz, T. Janssen, N. Hakimi, R. Simon, et al., “Clinical utility of quantitative gleason grading in prostate biopsies and prostatectomy specimens,” European urology, vol. 69, no. 4, pp. 592–598, 2016.spa
dc.rightsDerechos reservados - Universidad Nacional de Colombiaspa
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.licenseReconocimiento 4.0 Internacionalspa
dc.rights.spaAcceso abiertospa
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/spa
dc.subject.ddc610 - Medicina y salud::614 - Medicina Forense; incidencia de lesiones, heridas, enfermedades; medicina preventiva públicaspa
dc.subject.proposalFusión de Informaciónspa
dc.subject.proposalInformation Fusioneng
dc.subject.proposalHistopathology Imageseng
dc.subject.proposalImágenes de Histopatologíaspa
dc.subject.proposalAprendizaje de la Representaciónspa
dc.subject.proposalRepresentation Learningeng
dc.subject.proposalMétodos de Kernelspa
dc.subject.proposalKernel Methodseng
dc.subject.proposalSupervisión Débilspa
dc.subject.proposalWeakly-Supervisioneng
dc.subject.proposalMultimodal Learningeng
dc.subject.proposalAprendizaje Multimodalspa
dc.titleProstate histopathology image classification and retrieval using weakly-supervised multimodal fusion and representation learningspa
dc.typeTrabajo de grado - Maestríaspa
dc.type.coarhttp://purl.org/coar/resource_type/c_bdccspa
dc.type.coarversionhttp://purl.org/coar/version/c_ab4af688f83e57aaspa
dc.type.contentTextspa
dc.type.driverinfo:eu-repo/semantics/masterThesisspa
dc.type.versioninfo:eu-repo/semantics/acceptedVersionspa
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2spa

Archivos

Bloque original

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
1014278662.2020.pdf
Tamaño:
3.19 MB
Formato:
Adobe Portable Document Format

Bloque de licencias

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
license.txt
Tamaño:
3.8 KB
Formato:
Item-specific license agreed upon to submission
Descripción: