Unsupervised Reconstruction of Abdominal Muscle Contours in CT Images based on Statistical Shape Modeling
Resumo
A análise da composição corporal por tomografia computadorizada (TC) em pacientes oncológicos exige a diferenciação precisa entre tecidos adiposos e musculares, frequentemente dificultada por contornos incompletos. Este trabalho propõe um método não supervisionado para reconstrução desses contornos utilizando Modelos Estatísticos de Forma (SSM) e Modelos de Distribuição de Pontos (PDM), com classificação via Quociente Isoperimétrico (IPQ). Resultados experimentais em segmentação de tecido adiposo subcutâneo alcançaram um índice de Jaccard (IoU) médio de 93,4%, validando a eficácia e o baixo custo computacional da técnica.
Palavras-chave:
Segmentação de Imagens, Modelo Estatístico de Forma, Tomografia Computadorizada, Composição Corporal
Referências
Aubrey, J., Esfandiari, N., Baracos, V. E., Buteau, F. A., Frenette, J., Putman, C. T., and Mazurak, V. C. (2014). Measurement of skeletal muscle radiation attenuation and basis of its biological variation. Acta Physiologica (Oxford, England), 210(3):489–497.
Blåsjö, V. (2005). The isoperimetric problem. The American Mathematical Monthly, 112(6):526–566.
Bradshaw, P. T., Ibrahim, J. G., Stevens, J., Cleveland, R., Gould, M. K., Olshan, A. F., and Troester, M. A. (2019). Association of sarcopenia and body composition with recurrence and mortality in patients with nonmetastatic breast cancer. JAMA oncology, 5(2):178–186.
Caan, B. J., Cespedes Feliciano, E. M., Prado, C. M., Alexeeff, S., Kroenke, C. H., Bradshaw, P., Quesenberry, C. P., Weltzien, E. K., Castillo, A. L., Olobatuyi, T. A., and Chen, W. Y. (2018). Association of muscle and adiposity measured by computed tomography with survival in patients with nonmetastatic breast cancer. JAMA Oncology, 4(6):798–804.
Castiglione, J., Somasundaram, E., Gilligan, L. A., Trout, A. T., and Brady, S. (2021). Automated segmentation of abdominal skeletal muscle on pediatric ct scans using deep learning. Radiology: Artificial Intelligence, 3(2):e200130. PMID: 33937859; PMCID: PMC8043356.
Cootes, T. F., Taylor, C. J., Cooper, D. H., and Graham, J. (1995). Active shape modelstheir training and application. Computer Vision and Image Understanding, 61(1):38–59.
Hounsfield, G. N. (1980). Computed medical imaging technique: The EMI scanner. British Journal of Radiology, 53(628):446–457.
Kamiya, N., Zhou, X., Chen, H., Hara, T., Yokoyama, R., Kanematsu, M., Hoshi, H., and Fujita, H. (2010). A model based method for recognizing psoas major muscles in torso ct images. In Medical Imaging 2010: Computer-Aided Diagnosis, volume 7624. SPIE.
Lamecker, H., SeebaSS, M., Hege, H.-C., and Deuflhard, P. (2004). A 3d statistical shape model of the pelvic bone for segmentation. In Medical Imaging 2004: Image Processing, volume 5370. SPIE.
Mourtzakis, M., Prado, C. M., Lieffers, J. R., Reiman, T., McCargar, L. J., and Baracos, V. E. (2008). A practical and precise approach to quantification of body composition in cancer patients using computed tomography images acquired during routine care. Applied Physiology, Nutrition, and Metabolism, 33(5):997–1006. PMID: 18923576.
Pedrini, H. and Schwartz, W. R. (2007). Morfologia Matemática, Análise de Imagens Digitais, volume 1. Cengage Learning.
Rasoulian, A., Rohling, R., and Abolmaesumi, P. (2013). A statistical multi-vertebrae shape plus pose model for segmentation of ct images. In Medical Imaging 2013: Image Processing, volume 8669. SPIE.
Ronneberger, O., Fischer, P., and Brox, T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation. Springer International Publishing.
TomoVision (2021). sliceOmatic. Versão 5.0, Magog, Canadá.
Weston, A. D., Korfiatis, P., Kline, T. L., Philbrick, K. A., Kostandy, P., Sakinis, T., Sugimoto, M., Takahashi, N., and Erickson, B. J. (2019). Automated abdominal segmentation of ct scans for body composition analysis using deep learning. Radiology, 290(3):669–679.
Blåsjö, V. (2005). The isoperimetric problem. The American Mathematical Monthly, 112(6):526–566.
Bradshaw, P. T., Ibrahim, J. G., Stevens, J., Cleveland, R., Gould, M. K., Olshan, A. F., and Troester, M. A. (2019). Association of sarcopenia and body composition with recurrence and mortality in patients with nonmetastatic breast cancer. JAMA oncology, 5(2):178–186.
Caan, B. J., Cespedes Feliciano, E. M., Prado, C. M., Alexeeff, S., Kroenke, C. H., Bradshaw, P., Quesenberry, C. P., Weltzien, E. K., Castillo, A. L., Olobatuyi, T. A., and Chen, W. Y. (2018). Association of muscle and adiposity measured by computed tomography with survival in patients with nonmetastatic breast cancer. JAMA Oncology, 4(6):798–804.
Castiglione, J., Somasundaram, E., Gilligan, L. A., Trout, A. T., and Brady, S. (2021). Automated segmentation of abdominal skeletal muscle on pediatric ct scans using deep learning. Radiology: Artificial Intelligence, 3(2):e200130. PMID: 33937859; PMCID: PMC8043356.
Cootes, T. F., Taylor, C. J., Cooper, D. H., and Graham, J. (1995). Active shape modelstheir training and application. Computer Vision and Image Understanding, 61(1):38–59.
Hounsfield, G. N. (1980). Computed medical imaging technique: The EMI scanner. British Journal of Radiology, 53(628):446–457.
Kamiya, N., Zhou, X., Chen, H., Hara, T., Yokoyama, R., Kanematsu, M., Hoshi, H., and Fujita, H. (2010). A model based method for recognizing psoas major muscles in torso ct images. In Medical Imaging 2010: Computer-Aided Diagnosis, volume 7624. SPIE.
Lamecker, H., SeebaSS, M., Hege, H.-C., and Deuflhard, P. (2004). A 3d statistical shape model of the pelvic bone for segmentation. In Medical Imaging 2004: Image Processing, volume 5370. SPIE.
Mourtzakis, M., Prado, C. M., Lieffers, J. R., Reiman, T., McCargar, L. J., and Baracos, V. E. (2008). A practical and precise approach to quantification of body composition in cancer patients using computed tomography images acquired during routine care. Applied Physiology, Nutrition, and Metabolism, 33(5):997–1006. PMID: 18923576.
Pedrini, H. and Schwartz, W. R. (2007). Morfologia Matemática, Análise de Imagens Digitais, volume 1. Cengage Learning.
Rasoulian, A., Rohling, R., and Abolmaesumi, P. (2013). A statistical multi-vertebrae shape plus pose model for segmentation of ct images. In Medical Imaging 2013: Image Processing, volume 8669. SPIE.
Ronneberger, O., Fischer, P., and Brox, T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation. Springer International Publishing.
TomoVision (2021). sliceOmatic. Versão 5.0, Magog, Canadá.
Weston, A. D., Korfiatis, P., Kline, T. L., Philbrick, K. A., Kostandy, P., Sakinis, T., Sugimoto, M., Takahashi, N., and Erickson, B. J. (2019). Automated abdominal segmentation of ct scans for body composition analysis using deep learning. Radiology, 290(3):669–679.
Publicado
19/10/2026
Como Citar
MELO, Gabriel Furtado Lins; VASCONSELOS, Paulo Matheus Barroso de; MENDONÇA, José Fortunato; ANDRADE, Guilherme Souto de; ARAUJO, Thelmo Pontes de.
Unsupervised Reconstruction of Abdominal Muscle Contours in CT Images based on Statistical Shape Modeling. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 23. , 2026, Cuiabá/MT.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 1304-1315.
ISSN 2763-9061.
DOI: https://doi.org/10.5753/eniac.2026.31062.
