Segmentação de Acidente Vascular Cerebral Hemorrágico via Extração Hierárquica de Características e Contornos Ativos
Resumo
O acidente vascular cerebral (AVC) hemorrágico figura entre as principais causas de mortalidade e incapacitação no mundo, exigindo diagnóstico preciso e célere. Este estudo propõe um método híbrido de segmentação que combina Redes Neurais Convolucionais (RNC) com Contornos Ativos para a delimitação de lesões hemorrágicas em exames de tomografia computadorizada (TC) do crânio. A abordagem realiza extração hierárquica de características por meio de uma RNC multiescalar baseada em blocos Cross Stage Partial (CPS) e incorpora mapas de ativação contextual para guiar a energia interna e a inicialização do contorno ativo, promovendo segmentações morfologicamente coerentes. A avaliação quantitativa em bases públicas e privadas demonstrou a robustez e precisão do modelo, especificidade de 99,96%, precisão de 96,41% e sensibilidade de 92,02%. Em comparação com métodos do estado da arte, a arquitetura proposta apresentou desempenho competitivo, com maior consistência anatômica e menor dependência de intervenção manual, demonstrando seu potencial como ferramenta de apoio ao diagnóstico clínico.Referências
World Health Organization (WHO), “Stroke (cerebrovascular accident),” [link], 2025, accessed: 2025-07-23.
S. Yalçın and H. Vural, “Brain stroke classification and segmentation using encoder-decoder based deep convolutional neural networks,” computers in biology and Medicine, vol. 149, p. 105941, 2022.
N. Schmitt, Y. Mokli, C. Weyland, S. Gerry, C. Herweh, P. Ringleb, and S. Nagel, “Automated detection and segmentation of intracranial hemorrhage suspect hyperdensities in non-contrast-enhanced ct scans of acute stroke patients,” European radiology, vol. 32, no. 4, pp. 2246–2254, 2022.
L. Liu, S. Chen, F. Zhang, F.-X. Wu, Y. Pan, and J. Wang, “Deep convolutional neural network for automatically segmenting acute ischemic stroke lesion in multi-modality mri,” Neural Computing and Applications, vol. 32, no. 11, pp. 6545–6558, 2020.
M. T. Guimarães, A. G. Medeiros, J. S. Almeida, M. F. y Martin, R. Damaševičius, R. Maskeliūnas, C. L. C. Mattos, and P. P. Rebouças Filho, “An optimized approach to Huntington’s disease detecting via audio signals processing with dimensionality reduction,” in 2020 International Joint Conference on Neural Networks (IJCNN). IEEE, 2020, pp. 1–8.
M. B. Muhammad and M. Yeasin, “Eigen-cam: Class activation map using principal components,” in 2020 international joint conference on neural networks (IJCNN). IEEE, 2020, pp. 1–7.
M. Hssayeni, “Computed tomography images for intracranial hemorrhage detection and segmentation (version 1.0.0),” DOI: 10.13026/w8q8-ky94, 2019, rRID:SCR 007345.
A. Goldberger, L. Amaral, L. Glass, J. Hausdorff, P. C. Ivanov, R. Mark, and H. E. Stanley, “Physiobank, physiotoolkit, and physionet: Components of a new research resource for complex physiologic signals,” Circulation [Online], vol. 101, no. 23, pp. e215–e220, 2000.
B. Wu, “Brain Hemorrhage Segmentation Dataset (BHSD),” 2023, accessed: 2025-09-07. [Online]. Available: [link]
E. d. S. Rebouças, A. M. Braga, R. M. Sarmento, R. C. Marques, and P. P. Rebouças Filho, “Level set based on brain radiological densities for stroke segmentation in ct images,” in 2017 IEEE 30th International Symposium on Computer-Based Medical Systems. IEEE, 2017, pp. 391–396.
E. de Souza Rebouças, F. N. S. De Medeiros, R. C. P. Marques, J. V. S. Chagas, M. T. Guimarães, L. O. Santos, A. G. Medeiros, and S. A. Peixoto, “Level set approach based on parzen window and floor of log for edge computing object segmentation in digital images,” Applied Soft Computing, vol. 105, p. 107273, 2021.
A. M. Braga, R. C. Marques, F. N. Medeiros, J. F. R. Neto, A. G. Bianchi, C. M. Carneiro, and D. M. Ushizima, “Hierarchical median narrow band for level set segmentation of cervical cell nuclei,” Measurement, vol. 176, p. 109232, 2021.
S. P. P. d. Silva, R. F. Ivo, C. B. Barroso, J. C. N. Fernandes, T. F. Portela, A. G. Medeiros, P. H. F. d. Sousa, H. Song, and P. P. Rebouças Filho, “Context-driven active contour (cdac): A novel medical image segmentation method based on active contour and contextual understanding,” Sensors, vol. 25, no. 9, p. 2864, 2025.
P. P. Rebouças Filho, A. C. da Silva Barros, J. S. Almeida, J. Rodrigues, and V. H. C. de Albuquerque, “A new effective and powerful medical image segmentation algorithm based on optimum path snakes,” Applied Soft Computing, vol. 76, pp. 649–670, 2019.
S. Yalçın and H. Vural, “Brain stroke classification and segmentation using encoder-decoder based deep convolutional neural networks,” computers in biology and Medicine, vol. 149, p. 105941, 2022.
N. Schmitt, Y. Mokli, C. Weyland, S. Gerry, C. Herweh, P. Ringleb, and S. Nagel, “Automated detection and segmentation of intracranial hemorrhage suspect hyperdensities in non-contrast-enhanced ct scans of acute stroke patients,” European radiology, vol. 32, no. 4, pp. 2246–2254, 2022.
L. Liu, S. Chen, F. Zhang, F.-X. Wu, Y. Pan, and J. Wang, “Deep convolutional neural network for automatically segmenting acute ischemic stroke lesion in multi-modality mri,” Neural Computing and Applications, vol. 32, no. 11, pp. 6545–6558, 2020.
M. T. Guimarães, A. G. Medeiros, J. S. Almeida, M. F. y Martin, R. Damaševičius, R. Maskeliūnas, C. L. C. Mattos, and P. P. Rebouças Filho, “An optimized approach to Huntington’s disease detecting via audio signals processing with dimensionality reduction,” in 2020 International Joint Conference on Neural Networks (IJCNN). IEEE, 2020, pp. 1–8.
M. B. Muhammad and M. Yeasin, “Eigen-cam: Class activation map using principal components,” in 2020 international joint conference on neural networks (IJCNN). IEEE, 2020, pp. 1–7.
M. Hssayeni, “Computed tomography images for intracranial hemorrhage detection and segmentation (version 1.0.0),” DOI: 10.13026/w8q8-ky94, 2019, rRID:SCR 007345.
A. Goldberger, L. Amaral, L. Glass, J. Hausdorff, P. C. Ivanov, R. Mark, and H. E. Stanley, “Physiobank, physiotoolkit, and physionet: Components of a new research resource for complex physiologic signals,” Circulation [Online], vol. 101, no. 23, pp. e215–e220, 2000.
B. Wu, “Brain Hemorrhage Segmentation Dataset (BHSD),” 2023, accessed: 2025-09-07. [Online]. Available: [link]
E. d. S. Rebouças, A. M. Braga, R. M. Sarmento, R. C. Marques, and P. P. Rebouças Filho, “Level set based on brain radiological densities for stroke segmentation in ct images,” in 2017 IEEE 30th International Symposium on Computer-Based Medical Systems. IEEE, 2017, pp. 391–396.
E. de Souza Rebouças, F. N. S. De Medeiros, R. C. P. Marques, J. V. S. Chagas, M. T. Guimarães, L. O. Santos, A. G. Medeiros, and S. A. Peixoto, “Level set approach based on parzen window and floor of log for edge computing object segmentation in digital images,” Applied Soft Computing, vol. 105, p. 107273, 2021.
A. M. Braga, R. C. Marques, F. N. Medeiros, J. F. R. Neto, A. G. Bianchi, C. M. Carneiro, and D. M. Ushizima, “Hierarchical median narrow band for level set segmentation of cervical cell nuclei,” Measurement, vol. 176, p. 109232, 2021.
S. P. P. d. Silva, R. F. Ivo, C. B. Barroso, J. C. N. Fernandes, T. F. Portela, A. G. Medeiros, P. H. F. d. Sousa, H. Song, and P. P. Rebouças Filho, “Context-driven active contour (cdac): A novel medical image segmentation method based on active contour and contextual understanding,” Sensors, vol. 25, no. 9, p. 2864, 2025.
P. P. Rebouças Filho, A. C. da Silva Barros, J. S. Almeida, J. Rodrigues, and V. H. C. de Albuquerque, “A new effective and powerful medical image segmentation algorithm based on optimum path snakes,” Applied Soft Computing, vol. 76, pp. 649–670, 2019.
Publicado
30/09/2025
Como Citar
OHATA, Elene F.; MOREIRA, Hector L. M.; MEDEIROS, Aldísio G.; SILVA, Suane P. P. da; SANTOS, José Daniel de A.; ROCHA, Atslands R. da; REBOUÇAS FILHO, Pedro P..
Segmentação de Acidente Vascular Cerebral Hemorrágico via Extração Hierárquica de Características e Contornos Ativos. In: WORKSHOP DE TRABALHOS EM ANDAMENTO - CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI), 38. , 2025, Salvador/BA.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 97-102.
