Ítem
Acceso Abierto

Comparación de algoritmos de segmentación heurística de imágenes de angiografía coronaria utilizando técnicas no supervisadas

Mostrar el registro sencillo de la publicación
dc.contributor.advisorCaicedo Rodríguez, Pablo Eduardo
dc.creatorRamírez Millán, Nicolás
dc.creator.degreeMagíster en Ingeniería Biomédica
dc.creator.degreeLevelMaestría
dc.date.accessioned2026-07-21T23:15:50Z
dc.date.available2026-07-21T23:15:50Z
dc.date.created2026-06-12
dc.descriptionEl presente trabajo aborda el problema de la segmentación no supervisada de vasos coronarios en imágenes de angiografía, un desafío debido al ruido, los artefactos y la complejidad de las estructuras vasculares. Se comparan dos enfoques de segmentación, W-Net y Segment Anything Model (SAM), y se propone un modelo híbrido que combina las capacidades especializadas de W-Net con la precisión de segmentación de SAM. La estrategia desarrollada utiliza un mapa generado por W-Net para guiar la selección de las segmentaciones producidas por SAM. Los resultados muestran que, aunque el modelo híbrido presenta una menor similitud con las pseudo-etiquetas de entrenamiento, genera segmentaciones visualmente más coherentes y anatómicamente más precisas, evidenciando su potencial para mejorar la segmentación no supervisada de imágenes de angiografía coronaria.
dc.description.abstractThis paper addresses the problem of unsupervised segmentation of coronary vessels in angiography images, a challenge posed by noise, artifacts, and the complexity of vascular structures. Two segmentation approaches, W-Net and the Segment Anything Model (SAM), are compared, and a hybrid model is proposed that combines the specialized capabilities of W-Net with the segmentation accuracy of SAM. The developed strategy uses a map generated by W-Net to guide the selection of segmentations produced by SAM. The results show that, although the hybrid model exhibits less similarity to the training pseudo-labels, it generates visually more coherent and anatomically more accurate segmentations, demonstrating its potential to improve unsupervised segmentation of coronary angiography images.
dc.format.extent70 pp
dc.format.mimetypeapplication/pdf
dc.identifier.urihttps://repository.urosario.edu.co/handle/10336/48082
dc.language.isospa
dc.publisherUniversidad del Rosario
dc.publisherEscuela Colombiana de Ingeniería Julio Garavito
dc.publisher.departmentEscuela de Medicina y Ciencias de la Saludspa
dc.publisher.programMaestría en Ingeniería Biomédicaspa
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 International*
dc.rights.accesRightsinfo:eu-repo/semantics/openAccess
dc.rights.accesoAbierto (Texto Completo)
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.source.bibliographicCitationJ. El-Taraboulsi, C. P. Cabrera, C. Roney, and N. Aung, “Deep neural network architectures for cardiac image segmentation,” Artificial Intelligence in the Life Sciences, vol. 4, p. 100083, Dec. 2023, doi: 10.1016/j.ailsci.2023.100083.
dc.source.bibliographicCitationC. Bian et al., “Uncertainty-aware domain alignment for anatomical structure segmentation,” Med. Image Anal., vol. 64, p. 101732, Aug. 2020, doi: 10.1016/j.media.2020.101732
dc.source.bibliographicCitationX. Xia and B. Kulis, “W-Net: A Deep Model for Fully Unsupervised Image Segmentation,” Nov. 2017
dc.source.bibliographicCitationA. Medellin, D. Grabowsky, D. Mikulski, and R. Langari, “Sam-Sam - A Novel Approach to Hyperspectral Image Semantic Segmentation,” in 2023 13th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS), IEEE, Oct. 2023, pp. 1–5. doi: 10.1109/WHISPERS61460.2023.10431253.
dc.source.bibliographicCitationW. Chen et al., “Prostate Segmentation using 2D Bridged U-net,” in 2019 International Joint Conference on Neural Networks (IJCNN), IEEE, Jul. 2019, pp. 1–7. doi: 10.1109/IJCNN.2019.8851908.
dc.source.bibliographicCitationY. Zhang, Z. Shen, and R. Jiao, “Segment anything model for medical image segmentation: Current applications and future directions,” Comput. Biol. Med., vol. 171, p. 108238, Mar. 2024, doi: 10.1016/j.compbiomed.2024.108238.
dc.source.bibliographicCitationT.-Y. Hsia, E. A. Peck, and J. V. Conte, “Coronary Artery Disease,” in The Johns Hopkins Manual of Cardiac Surgical Care, Elsevier, 2008, pp. 61–93. doi: 10.1016/B978-0-323-01810-4.10003-3.
dc.source.bibliographicCitationB. A. Stark et al., “Global, Regional, and National Burden of Cardiovascular Diseases and Risk Factors in 204 Countries and Territories, 1990-2023,” JACC, Sep. 2025, doi: 10.1016/j.jacc.2025.08.015.
dc.source.bibliographicCitationS. Lee, M. Lee, and M. Kang, “Poisson–Gaussian Noise Analysis and Estimation for Low-Dose X-ray Images in the NSCT Domain,” Sensors, vol. 18, no. 4, p. 1019, Mar. 2018, doi: 10.3390/s18041019.
dc.source.bibliographicCitationLUQMAN and FAKHRUD DIN, “A Hyper-Heuristic Based Strategy for Image Segmentation Using Multilevel Thresholding,” Journal of Quality Measurement and Analysis, vol. 21, no. 2, pp. 123–152, Jun. 2025, doi: 10.17576/jqma.2102.2025.10.
dc.source.bibliographicCitationY. Sato et al., “Tissue classification based on 3D local intensity structures for volume rendering,” IEEE Trans. Vis. Comput. Graph., vol. 6, no. 2, pp. 160–180, 2000, doi: 10.1109/2945.856997.
dc.source.bibliographicCitationT. Jerman, F. Pernus, B. Likar, and Z. Spiclin, “Enhancement of Vascular Structures in <?Pub _newline ?> 3D and 2D Angiographic Images,” IEEE Trans. Med. Imaging, vol. 35, no. 9, pp. 2107–2118, Sep. 2016, doi: 10.1109/TMI.2016.2550102
dc.source.bibliographicCitationD. Indra, T. Hasanuddin, R. Satra, and N. R. Wibowo, “Eggs Detection Using Otsu Thresholding Method,” in 2018 2nd East Indonesia Conference on Computer and 65 Information Technology (EIConCIT), IEEE, Nov. 2018, pp. 10–13. doi: 10.1109/EIConCIT.2018.8878517
dc.source.bibliographicCitationChen Yu, Chen Dian-ren, Li Yang, and Chen Lei, “Otsu’s thresholding method based on gray level-gradient two-dimensional histogram,” in 2010 2nd International Asia Conference on Informatics in Control, Automation and Robotics (CAR 2010), IEEE, Mar. 2010, pp. 282–285. doi: 10.1109/CAR.2010.5456687.
dc.source.bibliographicCitationD. Shi, R. Liu, L. Tao, and C. Yuan, “Heuristic Dropout: An Efficient Regularization Method for Medical Image Segmentation Models,” in ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, May 2022, pp. 1101–1105. doi: 10.1109/ICASSP43922.2022.9747409.
dc.source.bibliographicCitationX. Wang, T. Duan, G. Ouyang, W. Hao, L. Mu, and X. Zhang, “Comprehensive review of segment anything model across multiple domains,” Digit. Signal Process., vol. 167, p. 105459, Dec. 2025, doi: 10.1016/j.dsp.2025.105459.
dc.source.bibliographicCitationZ. Wang, Z. Hou, S. Hou, and D. Cao, “Generalizable Digital Rock Image Segmentation under Limited Data with the Segment Anything Model,” Intelligent Geoengineering, Dec. 2025, doi: 10.1016/j.ige.2025.12.001
dc.source.bibliographicCitationH. Rezatofighi, N. Tsoi, J. Gwak, A. Sadeghian, I. Reid, and S. Savarese, “Generalized Intersection Over Union: A Metric and a Loss for Bounding Box Regression,” in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jun. 2019, pp. 658–666. doi: 10.1109/CVPR.2019.00075.
dc.source.bibliographicCitationY. Song et al., “Transformer framework for depth-assisted UDA semantic segmentation,” Eng. Appl. Artif. Intell., vol. 137, p. 109206, Nov. 2024, doi: 10.1016/j.engappai.2024.109206.
dc.source.bibliographicCitationW. Dong, Z. Liang, L. Wang, G. Tian, and Q. Long, “Unsupervised domain adaptation segmentation algorithm with cross-domain data augmentation and category contrast,” Neurocomputing, vol. 623, p. 129393, Mar. 2025, doi: 10.1016/j.neucom.2025.129393.
dc.source.bibliographicCitationS. Li, X. Zhao, and J. Cao, “An UDA bearing fault diagnosis method based on synergistic optimization of unbalanced data,” Mech. Syst. Signal Process., vol. 243, p. 113700, Jan. 2026, doi: 10.1016/j.ymssp.2025.113700.
dc.source.bibliographicCitationS. Soobia.et.al., “Analysis of Software Development Methodologies,” International Journal of Computing and Digital Systems, vol. 8, no. 5, pp. 445–460, Jan. 2019, doi: 10.12785/ijcds/080502
dc.source.bibliographicCitationC. Schröer, F. Kruse, and J. M. Gómez, “A Systematic Literature Review on Applying CRISP-DM Process Model,” Procedia Comput. Sci., vol. 181, pp. 526–534, 2021, doi: 10.1016/j.procs.2021.01.199.
dc.source.bibliographicCitationQ. Yang, H. Yi, L. Yi, M. Liu, and X. Chen, “Accurate segmentation and labeling of coronary artery segments in X-ray angiography with an improved UNet-based cGAN architecture,” Biomed. Signal Process. Control, vol. 112, p. 108812, Feb. 2026, doi: 10.1016/j.bspc.2025.108812.
dc.source.bibliographicCitationH. He, A. Banerjee, R. P. Choudhury, and V. Grau, “Deep learning based coronary vessels segmentation in X-ray angiography using temporal information,” Med. Image Anal., vol. 102, p. 103496, May 2025, doi: 10.1016/j.media.2025.103496.
dc.source.bibliographicCitationD. Lesage, E. D. Angelini, I. Bloch, and G. Funka-Lea, “A review of 3D vessel lumen segmentation techniques: Models, features and extraction schemes,” Med. Image Anal., vol. 13, no. 6, pp. 819–845, Dec. 2009, doi: 10.1016/j.media.2009.07.011.
dc.source.bibliographicCitationH. Xiang, Q. Zou, M. A. Nawaz, X. Huang, F. Zhang, and H. Yu, “Deep learning for image inpainting: A survey,” Pattern Recognit., vol. 134, p. 109046, Feb. 2023, doi: 10.1016/j.patcog.2022.109046.
dc.source.bibliographicCitationJ. Yang and N. I. R. Ruhaiyem, “Review of Deep Learning-Based Image Inpainting Techniques,” IEEE Access, vol. 12, pp. 138441–138482, 2024, doi: 10.1109/ACCESS.2024.3461782.
dc.source.bibliographicCitationP. M. Szczypiński, A. Klepaczko, and R. Olbrycht, “VesselKnife – software for the analysis of tubular structures in biomedical images,” Comput. Methods Programs Biomed., vol. 268, p. 108817, Aug. 2025, doi: 10.1016/j.cmpb.2025.108817.
dc.source.bibliographicCitationA. Arino, N. Yamamoto, K. Sasaki, R. Kimura, and T. Aizawa, “Quantitative evaluation of fatty change in rotator cuff muscles using the Otsu thresholding method,” J. Shoulder Elbow Surg., vol. 35, no. 1, pp. 200–208, Jan. 2026, doi: 10.1016/j.jse.2025.04.026.
dc.source.bibliographicCitationJ. Huang et al., “Deep learning for automated mandibular canal segmentation in CBCT scans,” BMC Oral Health, vol. 25, no. 1, p. 1699, Oct. 2025, doi: 10.1186/s12903-025-07098-5
dc.source.bibliographicCitationH. RostomMohamed and A. A. Mohammed Baqer Al-Shammaa, “Contour with Connected Components and Mesh Surface for Skin Image Segmentation,” Int. J. Appl. Inf. Syst., vol. 6, no. 2, pp. 48–52, Sep. 2013, doi: 10.5120/ijais13-451023.
dc.source.bibliographicCitationY. Wu, Q. Xu, Z. Zou, and H. Zhang, “Conditional generative model with skip- connection structure for low-light image enhancement,” Digit. Signal Process., vol. 151, p. 104556, Aug. 2024, doi: 10.1016/j.dsp.2024.104556.
dc.source.bibliographicCitationB. Walter, “Analysis of convolutional neural network image classifiers in a hierarchical max-pooling model with additional local pooling,” J. Stat. Plan. Inference, vol. 224, pp. 109–126, May 2023, doi: 10.1016/j.jspi.2022.11.001.
dc.source.bibliographicCitationW. Zhang et al., “Focal difficult-to-predict pixels dice loss for mitigating data imbalance in medical image segmentation,” Expert Syst. Appl., vol. 290, p. 128518, Sep. 2025, doi: 10.1016/j.eswa.2025.128518.
dc.source.bibliographicCitationQ. Li et al., “Low-dose computed tomography image reconstruction via a multistage convolutional neural network with autoencoder perceptual loss network,” Quant. Imaging Med. Surg., vol. 12, no. 3, pp. 1929–1957, Mar. 2022, doi: 10.21037/qims-21- 465.
dc.source.bibliographicCitationS. Shihui, W. Yilin, S. Xiaofeng, C. Huilin, and W. Biao, “Intelligent prediction of favorable shale oil areas based on deep learning and batch normalization,” 67 Geoenergy Science and Engineering, vol. 252, p. 213955, Sep. 2025, doi: 10.1016/j.geoen.2025.213955.
dc.source.bibliographicCitationY. Xu and H. Zhang, “Convergence of deep ReLU networks,” Neurocomputing, vol. 571, p. 127174, Feb. 2024, doi: 10.1016/j.neucom.2023.127174.
dc.source.bibliographicCitationJ. Li, Y. Cheng, Y. Lu, Z. Xia, Y. Mo, and G. Huang, “From ReLU to GeMU: Activation functions in the lens of cone projection,” Neural Networks, vol. 190, p. 107654, Oct. 2025, doi: 10.1016/j.neunet.2025.107654.
dc.source.bibliographicCitationH. Salem, A. E. Kabeel, E. M. S. El-Said, and O. M. Elzeki, “Predictive modelling for solar power-driven hybrid desalination system using artificial neural network regression with Adam optimization,” Desalination, vol. 522, p. 115411, Jan. 2022, doi: 10.1016/j.desal.2021.115411.
dc.source.bibliographicCitationZ. Khan, H. Liu, Y. Shen, and X. Zeng, “Deep learning improved YOLOv8 algorithm: Real-time precise instance segmentation of crown region orchard canopies in natural environment,” Comput. Electron. Agric., vol. 224, p. 109168, Sep. 2024, doi: 10.1016/j.compag.2024.109168
dc.source.bibliographicCitationF. Binsar and Sasmoko, “Integrating chain of thought and explainable AI in BERT- based deep learning for interpretable medical diagnosis,” Procedia Comput. Sci., vol. 269, pp. 389–399, 2025, doi: 10.1016/j.procs.2025.08.291.
dc.source.bibliographicCitationY. Zhuo, Z. Xu, D. Zhou, P. Xu, and Y. Tian, “RankSAM: Lightweight adapters and prompt generation in zero-shot semantic segmentation,” Neurocomputing, vol. 670, p. 132594, Mar. 2026, doi: 10.1016/j.neucom.2025.132594.
dc.source.bibliographicCitationM. Marjani, M. Mahdianpari, D. J. Varon, and F. Mohammadimanesh, “The integration of vision transformers and SAM for automated methane super-emitter detection using TROPOMI data,” J. Environ. Manage., vol. 393, p. 127034, Oct. 2025, doi: 10.1016/j.jenvman.2025.127034
dc.source.bibliographicCitationA. Carraro, M. Sozzi, and F. Marinello, “The Segment Anything Model (SAM) for accelerating the smart farming revolution,” Smart Agricultural Technology, vol. 6, p. 100367, Dec. 2023, doi: 10.1016/j.atech.2023.100367.
dc.source.bibliographicCitationY. Huang et al., “Segment anything model for medical images?,” Med. Image Anal., vol. 92, p. 103061, Feb. 2024, doi: 10.1016/j.media.2023.103061.
dc.source.bibliographicCitationJ. Wu et al., “Medical SAM adapter: Adapting segment anything model for medical image segmentation,” Med. Image Anal., vol. 102, p. 103547, May 2025, doi: 10.1016/j.media.2025.103547.
dc.source.bibliographicCitation2014 International Conference on Computer Vision Theory and Applications. IEEE, 2014.
dc.source.bibliographicCitationK. S. Ni and T. Q. Nguyen, “Adaptable K-nearest neighbor for image interpolation,” in 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, IEEE, Mar. 2008, pp. 1297–1300. doi: 10.1109/ICASSP.2008.4517855.
dc.source.bibliographicCitationG. Tziafas and H. Kasaei, “Early or Late Fusion Matters: Efficient RGB-D Fusion in Vision Transformers for 3D Object Recognition,” in 2023 IEEE/RSJ International 68 Conference on Intelligent Robots and Systems (IROS), IEEE, Oct. 2023, pp. 9558– 9565. doi: 10.1109/IROS55552.2023.10341422.
dc.source.bibliographicCitationK. Gadzicki, R. Khamsehashari, and C. Zetzsche, “Early vs Late Fusion in Multimodal Convolutional Neural Networks,” in 2020 IEEE 23rd International Conference on Information Fusion (FUSION), IEEE, Jul. 2020, pp. 1–6. doi: 10.23919/FUSION45008.2020.9190246.
dc.source.bibliographicCitationO. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” 2015, pp. 234–241. doi: 10.1007/978-3-319-24574- 4_28.
dc.source.bibliographicCitationS. He, Y. Feng, P. E. Grant, and Y. Ou, “Segmentation ability map: Interpret deep features for medical image segmentation,” Med. Image Anal., vol. 84, p. 102726, Feb. 2023, doi: 10.1016/j.media.2022.102726.
dc.source.bibliographicCitationM. Alidoost et al., “Model utility of a deep learning-based segmentation is not Dice coefficient dependent: A case study in volumetric brain blood vessel segmentation,” Intell. Based. Med., vol. 7, p. 100092, 2023, doi: 10.1016/j.ibmed.2023.100092.
dc.source.bibliographicCitationK. V. Houde, P. M. Kamble, and R. S. Hegadi, “A Comparative Analysis of IoU-Based Losses with YOLOv8: Fruit Tree Detection from UAV Images,” Procedia Comput. Sci., vol. 260, pp. 898–905, 2025, doi: 10.1016/j.procs.2025.03.272.
dc.source.bibliographicCitationT. Rädsch et al., “Quality Assured: Rethinking Annotation Strategies in Imaging AI,” Jul. 2024.
dc.source.bibliographicCitationA. A. Taha and A. Hanbury, “Metrics for evaluating 3D medical image segmentation: analysis, selection, and tool,” BMC Med. Imaging, vol. 15, no. 1, p. 29, Dec. 2015, doi: 10.1186/s12880-015-0068-x.
dc.source.bibliographicCitationS. Ghosal and P. Shah, “A deep-learning toolkit for visualization and interpretation of segmented medical images,” Cell Reports Methods, vol. 1, no. 7, p. 100107, Nov. 2021, doi: 10.1016/j.crmeth.2021.100107.
dc.source.bibliographicCitationZ. Li, D. Huo, M. Meurer, M. Panesso, W.-G. Drossel, and T. Bergs, “Enhancing Tool Wear Segmentation with LoRA-SAM and Point Prompts,” Procedia CIRP, vol. 134, pp. 705–710, 2025, doi: 10.1016/j.procir.2025.02.184
dc.source.instnameinstname:Universidad del Rosario
dc.source.reponamereponame:Repositorio Institucional EdocUR
dc.subjectSegmentación No Supervisada
dc.subjectAngiografia coronaria
dc.subjectDeep learning
dc.subjectW-Net
dc.subjectSegment Anything Mode
dc.subject.keywordUnsupervised segmentation
dc.subject.keywordCoronary angiography
dc.subject.keywordDeep learning
dc.subject.keywordW-Net
dc.subject.keywordSegment Anything Model
dc.titleComparación de algoritmos de segmentación heurística de imágenes de angiografía coronaria utilizando técnicas no supervisadas
dc.title.TranslatedTitleComparison of heuristic segmentation algorithms for coronary angiography images using unsupervised techniques
dc.typemasterThesis
dc.type.hasVersioninfo:eu-repo/semantics/acceptedVersion
dc.type.spaTrabajo de grado
local.regionesBogotá
Archivos
Bloque original
Mostrando1 - 1 de 1
Miniatura
Nombre:
Comparacion_de_algoritmos_de_segmentacion.pdf
Tamaño:
932.52 KB
Formato:
Adobe Portable Document Format
Descripción:
Descargar
Colecciones
Maestría en Ingeniería Biomédica