Classificação de Graus de Osteoartrite de Joelho em Raio-X via EfficientNetV2 com Atenção por Canal e Otimização Bayesiana
Resumo
A osteoartrite de joelho (OAJ) é uma das principais causas de dor e incapacidade funcional, sendo geralmente classificada via escala de KellgrenLawrence (KL). A diferenciação entre graus adjacentes dessa escala representa um desafio pela similaridade das alterações articulares. O método proposto neste trabalho visa classificar da severidade da OAJ utilizando a arquitetura EfficientNetV2B0 combinada com um mecanismo de atenção por canal e otimização Bayesiana para ajuste de hiperparâmetros. Os resultados obtidos demonstram desempenho promissor, alcançando acurácia de 77,58%, F1-score de 76,98% e coeficiente Kappa de 70,48%. Os resultados do método indicam potencial para auxiliar especialistas no diagnóstico e na avaliação da OAJ.Referências
Abdusalomov, A., Mirzakhalilov, S., Umirzakova, S., Ismailov, O., Sultanov, D., Nasimov, R., and Cho, Y. I. (2025). Lightweight early detection of knee osteoarthritis in athletes. Scientific Reports, 15(1):31413.
Akiba, T., Sano, S., Yanase, T., Ohta, T., and Koyama, M. (2019). Optuna: A next-generation hyperparameter optimization framework. In Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining, pages 2623–2631.
Beyaz, S., Yayli, S. B., and Kılıç, K. (2025). From variability to consistency: building a kellgren-lawrence gonarthrosis dataset. Journal of Orthopaedic Surgery and Research, 20(1):1–9.
Brejnebøl, M. W., Lenskjold, A., Ziegeler, K., Ruitenbeek, H., Müller, F. C., Nybing, J. U., Visser, J. J., Schiphouwer, L. M., Jasper, J., Bashian, B., et al. (2024). Interobserver agreement and performance of concurrent ai assistance for radiographic evaluation of knee osteoarthritis. Radiology, 312(1):e233341.
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., and Fei-Fei, L. (2009). Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, pages 248–255. Ieee.
Diniz, J. O., Ribeiro, N. P., Junior, D. A. D., da Cruz, L. B., de Carvalho Filho, A. O., Gomes Jr, D. L., Silva, A. C., and de Paiva, A. C. (2024). Efficientxyz-deepfeatures: seleção de esquema de cor e arquitetura deep features na classificação de câncer de cólon em imagens histopatológicas. In Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS), pages 82–93. SBC.
Fei, M., Lu, S., Chung, J. H., Hassan, S., Elsissy, J., and Schneiderman, B. A. (2024). Diagnosing the severity of knee osteoarthritis using regression scores from artificial intelligence convolution neural networks. Orthopedics, 47(5):e247–e254.
Gornale, S. S., Patravali, P. U., and Hiremath, P. S. (2020). A comprehensive digital knee x-ray image dataset for the assessment of osteoarthritis. JSM Biomed. Imag. Data Pap, 6:1012.
Hassan, E. and Ghadiri, H. (2025). Advancing brain tumor classification: A robust framework using efficientnetv2 transfer learning and statistical analysis. Computers in Biology and Medicine, 185:109542.
Jain, A. K., Murty, M. N., and Flynn, P. J. (1999). Data clustering: a review. ACM computing surveys (CSUR), 31(3):264–323.
Kellgren, J. and Lawrence, J. (1957). Radiological assessment of rheumatoid arthritis. Annals of the rheumatic diseases, 16(4):485.
Kumari, L. V. R., Jagruti, K., Chandra, G. R., Reddy, M. S., and Bhadramma, B. (2024). Transfer learning based efficientnet for knee osteoarthritis classification. Traitement du Signal, 41(2).
Langworthy, M., Dasa, V., and Spitzer, A. I. (2024). Knee osteoarthritis: disease burden, available treatments, and emerging options. Therapeutic Advances in Musculoskeletal Disease, 16:1759720X241273009.
Lee, D. W., Song, D. S., Han, H.-S., and Ro, D. H. (2024a). Accurate, automated classification of radiographic knee osteoarthritis severity using a novel method of deep learning: Plug-in modules. Knee Surgery & Related Research, 36(1):24.
Lee, K.-H., Lee, R.-W., Yun, J.-S., Kim, M.-S., and Choi, H.-S. (2024b). Automated diagnosis of knee osteoarthritis using resnet101 on a deep: Phi: leveraging a no-code ai platform for efficient and accurate medical image analysis. Diagnostics, 14(21):2451.
Li, E., Tan, J., Xu, K., Pan, Y., and Xu, P. (2024). Global burden and socioeconomic impact of knee osteoarthritis: a comprehensive analysis. Frontiers in Medicine, 11:1323091.
Momenpour, T. and Abu Mallouh, A. (2025). Optimizing cnn-based diagnosis of knee osteoarthritis: Enhancing model accuracy with cleanlab relabeling. Diagnostics, 15(11):1332.
Polesel, A., Ramponi, G., and Mathews, V. J. (2000). Image enhancement via adaptive unsharp masking. IEEE transactions on image processing, 9(3):505–510.
Rani, S., Memoria, M., Almogren, A., Bharany, S., Joshi, K., Altameem, A., Rehman, A. U., and Hamam, H. (2024). Deep learning to combat knee osteoarthritis and severity assessment by using cnn-based classification. BMC Musculoskeletal Disorders, 25(1):817.
Silvério, A. C. M. and Machado, A. M. C. (2025). Classificaçao da osteoartrite de joelho em imagens de raio-x por meio de ensemble learning. In Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS), pages 973–978. SBC.
Tan, M. and Le, Q. (2021). Efficientnetv2: Smaller models and faster training. In International conference on machine learning, pages 10096–10106. PMLR.
Touahema, S., Zaimi, I., Zrira, N., Ngote, M. N., Doulhousne, H., and Aouial, M. (2024). Medknee: A new deep learning-based software for automated prediction of radiographic knee osteoarthritis. Diagnostics, 14(10):993.
Wang, Q., Wu, B., Zhu, P., Li, P., Zuo, W., and Hu, Q. (2020). Eca-net: Efficient channel attention for deep convolutional neural networks. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11534–11542.
Zuiderveld, K. (1994). Contrast limited adaptive histogram equalization. In Graphics gems IV, pages 474–485. Academic Press Professional, Inc.
Akiba, T., Sano, S., Yanase, T., Ohta, T., and Koyama, M. (2019). Optuna: A next-generation hyperparameter optimization framework. In Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining, pages 2623–2631.
Beyaz, S., Yayli, S. B., and Kılıç, K. (2025). From variability to consistency: building a kellgren-lawrence gonarthrosis dataset. Journal of Orthopaedic Surgery and Research, 20(1):1–9.
Brejnebøl, M. W., Lenskjold, A., Ziegeler, K., Ruitenbeek, H., Müller, F. C., Nybing, J. U., Visser, J. J., Schiphouwer, L. M., Jasper, J., Bashian, B., et al. (2024). Interobserver agreement and performance of concurrent ai assistance for radiographic evaluation of knee osteoarthritis. Radiology, 312(1):e233341.
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., and Fei-Fei, L. (2009). Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, pages 248–255. Ieee.
Diniz, J. O., Ribeiro, N. P., Junior, D. A. D., da Cruz, L. B., de Carvalho Filho, A. O., Gomes Jr, D. L., Silva, A. C., and de Paiva, A. C. (2024). Efficientxyz-deepfeatures: seleção de esquema de cor e arquitetura deep features na classificação de câncer de cólon em imagens histopatológicas. In Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS), pages 82–93. SBC.
Fei, M., Lu, S., Chung, J. H., Hassan, S., Elsissy, J., and Schneiderman, B. A. (2024). Diagnosing the severity of knee osteoarthritis using regression scores from artificial intelligence convolution neural networks. Orthopedics, 47(5):e247–e254.
Gornale, S. S., Patravali, P. U., and Hiremath, P. S. (2020). A comprehensive digital knee x-ray image dataset for the assessment of osteoarthritis. JSM Biomed. Imag. Data Pap, 6:1012.
Hassan, E. and Ghadiri, H. (2025). Advancing brain tumor classification: A robust framework using efficientnetv2 transfer learning and statistical analysis. Computers in Biology and Medicine, 185:109542.
Jain, A. K., Murty, M. N., and Flynn, P. J. (1999). Data clustering: a review. ACM computing surveys (CSUR), 31(3):264–323.
Kellgren, J. and Lawrence, J. (1957). Radiological assessment of rheumatoid arthritis. Annals of the rheumatic diseases, 16(4):485.
Kumari, L. V. R., Jagruti, K., Chandra, G. R., Reddy, M. S., and Bhadramma, B. (2024). Transfer learning based efficientnet for knee osteoarthritis classification. Traitement du Signal, 41(2).
Langworthy, M., Dasa, V., and Spitzer, A. I. (2024). Knee osteoarthritis: disease burden, available treatments, and emerging options. Therapeutic Advances in Musculoskeletal Disease, 16:1759720X241273009.
Lee, D. W., Song, D. S., Han, H.-S., and Ro, D. H. (2024a). Accurate, automated classification of radiographic knee osteoarthritis severity using a novel method of deep learning: Plug-in modules. Knee Surgery & Related Research, 36(1):24.
Lee, K.-H., Lee, R.-W., Yun, J.-S., Kim, M.-S., and Choi, H.-S. (2024b). Automated diagnosis of knee osteoarthritis using resnet101 on a deep: Phi: leveraging a no-code ai platform for efficient and accurate medical image analysis. Diagnostics, 14(21):2451.
Li, E., Tan, J., Xu, K., Pan, Y., and Xu, P. (2024). Global burden and socioeconomic impact of knee osteoarthritis: a comprehensive analysis. Frontiers in Medicine, 11:1323091.
Momenpour, T. and Abu Mallouh, A. (2025). Optimizing cnn-based diagnosis of knee osteoarthritis: Enhancing model accuracy with cleanlab relabeling. Diagnostics, 15(11):1332.
Polesel, A., Ramponi, G., and Mathews, V. J. (2000). Image enhancement via adaptive unsharp masking. IEEE transactions on image processing, 9(3):505–510.
Rani, S., Memoria, M., Almogren, A., Bharany, S., Joshi, K., Altameem, A., Rehman, A. U., and Hamam, H. (2024). Deep learning to combat knee osteoarthritis and severity assessment by using cnn-based classification. BMC Musculoskeletal Disorders, 25(1):817.
Silvério, A. C. M. and Machado, A. M. C. (2025). Classificaçao da osteoartrite de joelho em imagens de raio-x por meio de ensemble learning. In Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS), pages 973–978. SBC.
Tan, M. and Le, Q. (2021). Efficientnetv2: Smaller models and faster training. In International conference on machine learning, pages 10096–10106. PMLR.
Touahema, S., Zaimi, I., Zrira, N., Ngote, M. N., Doulhousne, H., and Aouial, M. (2024). Medknee: A new deep learning-based software for automated prediction of radiographic knee osteoarthritis. Diagnostics, 14(10):993.
Wang, Q., Wu, B., Zhu, P., Li, P., Zuo, W., and Hu, Q. (2020). Eca-net: Efficient channel attention for deep convolutional neural networks. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11534–11542.
Zuiderveld, K. (1994). Contrast limited adaptive histogram equalization. In Graphics gems IV, pages 474–485. Academic Press Professional, Inc.
Publicado
01/06/2026
Como Citar
SANTOS, Walberto M.; AMORIM, Marcos R. A.; DINIZ, João O. B.; RIBEIRO, Neilson P.; B. JÚNIOR, Geraldo; ALMEIDA, João Dallyson S..
Classificação de Graus de Osteoartrite de Joelho em Raio-X via EfficientNetV2 com Atenção por Canal e Otimização Bayesiana. In: SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO APLICADA À SAÚDE (SBCAS), 26. , 2026, Ouro Preto/MG.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 1134-1145.
ISSN 2763-8952.
DOI: https://doi.org/10.5753/sbcas.2026.21655.
