Dellazoppa, Lucca (2020) Un enfoque de aprendizaje profundo para la cuantificación de la función cardíaca / A deep learning approach for cardiac function quantification. Maestría en Ingeniería, Universidad Nacional de Cuyo, Instituto Balseiro.
![[img]](/style/images/fileicons/application_pdf.png)
| PDF (Tesis) Disponible bajo licencia Creative Commons: Reconocimiento - No comercial - Compartir igual. Español 7Mb |
Resumen en español
Las enfermedades cardiovasculares son la primera causa de muerte a nivel mundial. Es por esto que las tareas de detección temprana, diagnóstico y pronóstico resultan de vital importancia para las personas que padecen o corren riesgo de padecer estas patolog ías. Dos indicadores ampliamente utilizados para cuantificar la función y estructura cardiaca son la fracción de eyección (EF) y la masa del miocardio correspondiente al ventrículo izquierdo (LVM), cuyo cálculo requiere identificar y delinear de forma precisa los contornos de distintos tejidos del corazón en estudios de imágenes médicas. Si bien existen métodos semiautomáticos para realizar esta tarea, no suelen tener la precisi ón alcanzada mediante la demarcación manual, la cual es comúnmente considerada tediosa, repetitiva y puede demandar alrededor de 20 minutos por paciente para un médico experto. En este contexto, se propone trasladar al ámbito clínico una aplicación que permita cuantificar la función cardiaca a partir de la estimación de la EF y LVM de manera automática en estudios de resonancia magnética nuclear; con precisiones y errores comparables a los obtenidos por especialistas. Con ese objetivo se diseñó y desarrolló la herramienta de código abierto CardIAc, la cual incorpora dos modelos de redes neuronales profundas que se entrenaron para detectar las estructuras cardiacas de interés. De esta manera, se presenta un enfoque automático que logra reducir el tiempo de los métodos tradicionales a 2 min. (CPU) y 30 seg. (GPU), con errores en la cuantificación de EF (0.02±4.6[ %]) que se encuentran en el rango inter-observador actual. Asimismo, este enfoque contempla la posibilidad de realizar una corrección manual de los tejidos identificados en la etapa de detección de estructuras. La herramienta provista se desarrolló como una extensión del software de visualización y procesado de imágenes médicas 3DSlicer, integrándose así con el resto de utilidades nativas.
Resumen en inglés
Cardiovascular diseases are the leading cause of death worldwide. That is why the tasks of early detection, diagnosis and prognosis are of paramount importance for people who suer or are at risk of suering from these pathologies. Two descriptors widely used for the quantification of cardiac function and structure are the ejection fraction (EF) and the left ventricular myocardial mass (LVM), whose calculation requires identifying and precisely delineating the contours of dierent heart tissues in medical imagingbased studies. Although there are semi-automatic methods to perform this task, they usually do not achieve the precision obtained through manual demarcation, which is commonly considered tedious and repetitive, requiring up to 20 minutes per patient for an expert. In this context, we aim to provide the clinical eld with an application that allows the automatic quantication of cardiac function (EF and LVM) in nuclear magnetic resonance imaging studies, with accuracies and errors similar to those obtained by specialists. With this objective, the open source tool CardIAc was designed and developed, which incorporates two models of deep neural networks that were trained to identify the cardiac structures of interest. In this way, we present an automatic approach that manages to reduce the time of traditional methods to 2 min. (CPU) and 30 sec. (GPU), with an observed error in EF quantication (0.02±4.6 [%]) within the current inter-observer range. Besides, the capability of making manual corrections to the identified tissues is covered in this approach. The tool provided was developed as an extension of the 3DSlicer medical image visualization and processing software, thus integrating with the rest of native utilities.
| Tipo de objeto: | Tesis (Maestría en Ingeniería) |
|---|---|
| Palabras Clave: | [Cardiac quantification; Cuantificación cardíaca; Cardiac segmentation; Segmentación cardíaca; Convolutional networks; Redes convolucionales] |
| Referencias: | [1] H. Ritchie, "Causes of death," Our World in Data, 2018. https:// ourworldindata.org/causes-of-death. [2] World Health Organization. Office of Library and Health Literature Services, "Cardiovascular diseases (cvds)," 2017. https://www.who.int/news-room/factsheets/ detail/cardiovascular-diseases-(cvds). [3] M. Cikes and S. D. Solomon, "Beyond ejection fraction: an integrative approach for assessment of cardiac structure and function in heart failure," European Heart Journal, vol. 37, no. 21, p. 1642 1650, 2015. [4] J. Román, J. Candell-Riera, R. Arnold, P. Sanchez, S. Aguadé-Bruix, J. Bermejo, A. Revilla, A. Villa, H. Cuellar, C. Hernandez, and F. Fernández-Avilés, "Análisis cuantitativo de la función ventricular izquierda como herramienta para la investigaci ón clínica. fundamentos y metodología," Revista Española De Cardiología, vol. 62, pp. 535 551, 05 2009. [5] T. H. Marwick, "Ejection fraction pros and cons," Journal of the American College of Cardiology, vol. 72, no. 19, p. 2360 2379, 2018. [6] S. E. Petersen, N. Aung, M. M. Sanghvi, F. Zemrak, K. Fung, J. M. Paiva, J. M. Francis, M. Y. Khanji, E. Lukaschuk, A. M. Lee, and et al., "Reference ranges for cardiac structure and function using cardiovascular magnetic resonance (cmr) in caucasians from the uk biobank population cohort," Journal of Cardiovascular Magnetic Resonance, vol. 19, no. 1, 2017. [7] A. R. Patel and C. M. Kramer, "Role of cardiac magnetic resonance in the diagnosis and prognosis of nonischemic cardiomyopathy," Journal of the American College of Cardiology: Cardiovascular Imaging, vol. 10, 10 2017. [8] A. Rosset, L. Spadola, and O. Ratib, "Osirix: An open-source software for navigating in multidimensional dicom images," Journal of Digital Imaging, vol. 17, no. 3, p. 205 216, 2004. [9] E. Heiberg, J. Sjögren, M. Ugander, M. Carlsson, H. Engblom, and H. Arheden, "Design and validation of segment - freely available software for cardiovascular image analysis," BMC Medical Imaging, vol. 10, no. 1, 2010. [10] W. Bai, M. Sinclair, G. Tarroni, O. Oktay, M. Rajchl, G. Vaillant, A. M. Lee, N. Aung, E. Lukaschuk, M. M. Sanghvi, and et al., "Automated cardiovascular magnetic resonance image analysis with fully convolutional networks," Journal of Cardiovascular Magnetic Resonance, vol. 20, no. 1, 2018. [11] O. Ronneberger, P. Fischer, and T. Brox, "U-net: Convolutional networks for biomedical image segmentation," Lecture Notes in Computer Science Medical Image Computing and Computer-Assisted Intervention " MICCAI 2015, p. 234 241, 2015. [12] A. H. Curiale, F. D. Colavecchia, P. Kaluza, R. A. Isoardi, and G. Mato, "Automatic myocardial segmentation by using a deep learning network in cardiac mri," in 2017 XLIII Latin American Computer Conference (CLEI), pp. 1 6, Sept. 2017. [13] A. Curiale, F. Colavecchia, and G. Mato, "Automatic quantification of the lv function and mass: A deep learning approach for cardiovascular mri," Computer Methods and Programs in Biomedicine, vol. 169, pp. 37 50, 02 2019. [14] "Human anatomy planes." https://www.anatomynote.com/human-anatomy/ gross-view-of-human-body/human-anatomy-planes/. Accessed: 13-10-2020. [15] C. Mitchell, P. S. Rahko, L. A. Blauwet, B. Canaday, J. A. Finstuen, M. C. Foster, K. Horton, K. O. Ogunyankin, R. A. Palma, E. J. Velazquez, and et al., "Guidelines for performing a comprehensive transthoracic echocardiographic examination in adults: Recommendations from the american society of echocardiography," Journal of the American Society of Echocardiography, vol. 32, no. 1, p. 1 64, 2019. [16] "Diagram of the human heart." https://commons.wikimedia.org/w/index.php? curid=11429447. Accessed: 13-10-2020. [17] G. J. Tortora and B. H. Derrickson, Principles of Anatomy and Physiology. Wiley, 12 ed., 2008. [18] A. Frangi, W. Niessen, and M. Viergever, "Three-dimensional modeling for functional analysis of cardiac images: A review," IEEE transactions on medical imaging, vol. 20, pp. 2 25, 02 2001. [19] S. E. Petersen, M. Y. Khanji, S. Plein, P. Lancellotti, and C. Bucciarelli-Ducci, "European association of cardiovascular imaging expert consensus paper: a comprehensive review of cardiovascular magnetic resonance normal values of cardiac chamber size and aortic root in adults and recommendations for grading severity," European Heart Journal - Cardiovascular Imaging, 09 2019. [20] A. G. Gheorghe, A. Fuchs, C. Jacobsen, K. F. Kofoed, R. Møgelvang, N. Lynnerup, and J. Banner, "Cardiac left ventricular myocardial tissue density, evaluated by computed tomography and autopsy," BMC Medical Imaging, vol. 19, 12 2019. [21] R. Kline, "Cybernetics, automata studies, and the dartmouth conference on artifi- cial intelligence," IEEE Annals of the History of Computing, vol. 33, no. 4, p. 5 16, 2011. [22] A. Kaplan and M. Haenlein, "Siri, siri, in my hand: Who's the fairest in the land? on the interpretations, illustrations, and implications of artificial intelligence," Business Horizons, vol. 62, no. 1, pp. 15 25, 2019. [23] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. MIT Press, 2017. [24] Y. Lecun, P. Haner, and Y. Bengio, "Object recognition with gradient-based learning," 08 2000. [25] Bai Kunlun, "A comprehensive introduction to different types of convolutions in deep learning." https://towardsdatascience.com/a-comprehensive-introductionto- different- types-of-convolutions-in-deep-learning-669281e58215. [26] D. P. Kingma and M.Welling, "Auto-Encoding Variational Bayes," in 2nd International Conference on Learning Representations, ICLR 2014, Banff, AB, Canada, April 14-16, 2014, Conference Track Proceedings, 2014. [27] W. J. Schroeder and K. M. Martin, "The visualization toolkit," Visualization Handbook, p. 593 614, 2005. [28] H. J. Johnson, M. M. McCormick, and L. Ibanez, Template:The ITK Software Guide Book 1: Introduction and Development Guidelines-Volume 1, 2015. [29] A. Bernardo, "Estimación del esfuerzo miocárdico a partir del procesado de imá- genes médicas," Maestría en Física Médica, Master's thesis, Instituto Balseiro, Diciembre 2020. [30] A. Bernardo, L. Dellazoppa, G. Mato, and A. H. Curiale, Manual de uso: CardIAc, 2020. https://cutt.ly/cardIAc-manual. [31] C. C. Imaging, "cvi42." https://www.circlecvi.com/. [32] O. Bernard, A. Lalande, C. Zotti, F. Cervenansky, X. Yang, P.-A. Heng, I. Cetin, K. Lekadir, O. Camara, M. A. G. Ballester, and et al., "Deep learning techniques for automatic mri cardiac multi-structures segmentation and diagnosis: Is the problem solved?," IEEE Transactions on Medical Imaging, vol. 37, no. 11, p. 2514 2525, 2018. [33] A. H. Curiale, L. Dellazoppa, and G. Mato, "Cardiac: Cuantificación de la función cardiaca mediante la detección automática del ventrículo izquierdo en imágenes de resonancia magnética nuclear," in Taller Argentino de Computación Científi- ca, (San Luis, Argentina), Facultad de Ciencias Físico Matemáticas y Naturales, Universidad Nacional de San Luis, 2019. [34] K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition, " in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770 778, 2016. [35] S. Ioffe and C. Szegedy, "Batch normalization: Accelerating deep network training by reducing internal covariate shift," CoRR, vol. abs/1502.03167, 2015. [36] A. Suinesiaputra, M. M. Sanghvi, N. Aung, J. M. Paiva, F. Zemrak, K. Fung, E. Lukaschuk, A. M. Lee, V. Carapella, Y. J. Kim, and et al., "Fully-automated left ventricular mass and volume mri analysis in the uk biobank population cohort: evaluation of initial results," The International Journal of Cardiovascular Imaging, vol. 34, no. 2, p. 281 291, 2017. [37] D. Dey, P. J. Slomka, P. Leeson, D. Comaniciu, S. Shrestha, P. P. Sengupta, and T. H. Marwick, "Artificial intelligence in cardiovascular imaging," Journal of the American College of Cardiology, vol. 73, no. 11, pp. 1317 1335, 2019. [38] C. Chen, C. Qin, H. Qiu, G. Tarroni, J. Duan, W. Bai, and D. Rueckert, "Deep learning for cardiac image segmentation: A review," Frontiers in Cardiovascular Medicine, vol. 7, 2020. [39] Q. Zheng, H. Delingette, N. Duchateau, and N. Ayache, "3d consistent & robust segmentation of cardiac images by deep learning with spatial propagation," CoRR, vol. abs/1804.09400, 2018. [40] R. Ranftl, K. Lasinger, D. Hafner, K. Schindler, and V. Koltun, "Towards robust monocular depth estimation: Mixing datasets for zero-shot cross-dataset transfer," IEEE Transactions on Pattern Analysis and Machine Intelligence, p. 1 1, 2020. [41] C. Sudlow, J. Gallacher, N. Allen, V. Beral, P. Burton, J. Danesh, P. Downey, P. Elliott, J. Green, M. Landray, and et al., "Uk biobank: An open access resource for identifying the causes of a wide range of complex diseases of middle and old age," PLOS Medicine, vol. 12, no. 3, 2015. |
| Materias: | Ingeniería |
| Divisiones: | Gcia. de área de Investigación y aplicaciones no nucleares > Gcia. de Física > Física médica |
| Código ID: | 980 |
| Depositado Por: | Marisa G. Velazco Aldao |
| Depositado En: | 21 Sep 2021 16:09 |
| Última Modificación: | 21 Sep 2021 16:09 |
Personal del repositorio solamente: página de control del documento
