Comparative Study of the Combo Loss Function for Deep Endometriosis Segmentation in Magnetic Resonance Imaging Using DeepLabv3
Abstract
Endometriosis is an inflammatory disease that affects 10% of the female population globally, causing symptoms ranging from bleeding and pain to infertility. Diagnosing endometriosis is a time-consuming and costly process for clinicians. Diagnosis by imaging tests is becoming the standard, with magnetic resonance imaging (MRI) standing out in this field. Using image processing techniques combined with machine learning, it is possible to build algorithms to aid in the location of lesions, functioning as tools to assist the physician. This work seeks to conduct a study on different strategies for segmenting these lesions in MRI images.References
Allaire, C., Bedaiwy, M. A., and Yong, P. J. (2023). Diagnosis and management of endometriosis. Cmaj, 195(10):E363–E371.
Asgari Taghanaki, S., Zheng, Y., Zhou, S. K., Georgescu, B., Sharma, P., Xu, D., Comaniciu, D., and Hamarneh, G. (2018). Combo loss: Handling input and output imbalance in multi-organ segmentation. arXiv e-prints, pages arXiv–1805.
Bellelis, P., Dias Jr, J. A., Podgaec, S., Gonzales, M., Baracat, E. C., and Abrão, M. S. (2010). Aspectos epidemiológicos e clínicos da endometriose pélvica: uma série de casos. Revista da associação médica brasileira, 56:467–471.
Chen, L.-C., Papandreou, G., Schroff, F., and Adam, H. (2017). Rethinking atrous convolution for semantic image segmentation.
De Corte, P., Klinghardt, M., von Stockum, S., and Heinemann, K. (2025). Time to diagnose endometriosis: Current status, challenges and regional characteristics—a systematic literature review. BJOG: An International Journal of Obstetrics & Gynaecology, 132(2):118–130.
Figueredo, W. K., da Silva, I. F., Diniz, J. O., Silva, A. C., de Paiva, A. C., Salomão, A. C. B., and de Oliveira, M. A. (2023). Abordagem computacional baseada em deep learning para o diagnóstico de endometriose profunda através de imagens de ressonância magnética. In Anais do XXIII Simpósio Brasileiro de Computação Aplicada à Saúde, pages 138–149. SBC.
He, K., Zhang, X., Ren, S., and Sun, J. (2015). Deep residual learning for image recognition.
Howard, A., Sandler, M., Chu, G., Chen, L.-C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., Vasudevan, V., Le, Q. V., and Adam, H. (2019). Searching for mobilenetv3.
Leyland, N., Casper, R., Laberge, P., Singh, S. S., Allen, L., Arendas, K., Leyland, N., Allaire, C., Awadalla, A., Best, C., et al. (2010). Endometriosis: diagnosis and management. Journal of Endometriosis, 2(3):107–134.
Liang, X., Alpuing Radilla, L. A., Khalaj, K., Dawoodally, H., Mokashi, C., Guan, X., Roberts, K. E., Sheth, S. A., Tammisetti, V. S., and Giancardo, L. (2025). A multi-modal pelvic mri dataset for deep learning-based pelvic organ segmentation in endometriosis. Scientific Data, 12(1):1292.
Moassefi, M., Faghani, S., Colak, C., Sheedy, S. P., Andrieu, P. L. C., Wang, S. S., McPhedran, R. L., Flicek, K. T., Suman, G., Takahashi, H., et al. (2025). Advancing endometriosis detection in daily practice: a deep learning-enhanced multi-sequence mri analytical model. Abdominal Radiology, pages 1–12.
Nácul, A. P. and Spritzer, P. M. (2010). Aspectos atuais do diagnóstico e tratamento da endometriose. Revista Brasileira de ginecologia e obstetrícia, 32:298–307.
Oktay, O., Schlemper, J., Folgoc, L. L., Lee, M., Heinrich, M., Misawa, K., Mori, K., McDonagh, S., Hammerla, N. Y., Kainz, B., Glocker, B., and Rueckert, D. (2018). Attention u-net: Learning where to look for the pancreas.
Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18, pages 234–241. Springer.
Si, H., Shi, Z., Hu, X., Wang, Y., and Yang, C. (2020). Image semantic segmentation based on improved deeplab v3 model. International Journal of Modelling, Identification and Control, 36(2):116–125.
Tsang, S.-H. Review: DeepLabv3-Atrous Separable Convolution (Semantic Segmentation) — sh-tsang.medium.com. [Accessed 13-01-2026].
Asgari Taghanaki, S., Zheng, Y., Zhou, S. K., Georgescu, B., Sharma, P., Xu, D., Comaniciu, D., and Hamarneh, G. (2018). Combo loss: Handling input and output imbalance in multi-organ segmentation. arXiv e-prints, pages arXiv–1805.
Bellelis, P., Dias Jr, J. A., Podgaec, S., Gonzales, M., Baracat, E. C., and Abrão, M. S. (2010). Aspectos epidemiológicos e clínicos da endometriose pélvica: uma série de casos. Revista da associação médica brasileira, 56:467–471.
Chen, L.-C., Papandreou, G., Schroff, F., and Adam, H. (2017). Rethinking atrous convolution for semantic image segmentation.
De Corte, P., Klinghardt, M., von Stockum, S., and Heinemann, K. (2025). Time to diagnose endometriosis: Current status, challenges and regional characteristics—a systematic literature review. BJOG: An International Journal of Obstetrics & Gynaecology, 132(2):118–130.
Figueredo, W. K., da Silva, I. F., Diniz, J. O., Silva, A. C., de Paiva, A. C., Salomão, A. C. B., and de Oliveira, M. A. (2023). Abordagem computacional baseada em deep learning para o diagnóstico de endometriose profunda através de imagens de ressonância magnética. In Anais do XXIII Simpósio Brasileiro de Computação Aplicada à Saúde, pages 138–149. SBC.
He, K., Zhang, X., Ren, S., and Sun, J. (2015). Deep residual learning for image recognition.
Howard, A., Sandler, M., Chu, G., Chen, L.-C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., Vasudevan, V., Le, Q. V., and Adam, H. (2019). Searching for mobilenetv3.
Leyland, N., Casper, R., Laberge, P., Singh, S. S., Allen, L., Arendas, K., Leyland, N., Allaire, C., Awadalla, A., Best, C., et al. (2010). Endometriosis: diagnosis and management. Journal of Endometriosis, 2(3):107–134.
Liang, X., Alpuing Radilla, L. A., Khalaj, K., Dawoodally, H., Mokashi, C., Guan, X., Roberts, K. E., Sheth, S. A., Tammisetti, V. S., and Giancardo, L. (2025). A multi-modal pelvic mri dataset for deep learning-based pelvic organ segmentation in endometriosis. Scientific Data, 12(1):1292.
Moassefi, M., Faghani, S., Colak, C., Sheedy, S. P., Andrieu, P. L. C., Wang, S. S., McPhedran, R. L., Flicek, K. T., Suman, G., Takahashi, H., et al. (2025). Advancing endometriosis detection in daily practice: a deep learning-enhanced multi-sequence mri analytical model. Abdominal Radiology, pages 1–12.
Nácul, A. P. and Spritzer, P. M. (2010). Aspectos atuais do diagnóstico e tratamento da endometriose. Revista Brasileira de ginecologia e obstetrícia, 32:298–307.
Oktay, O., Schlemper, J., Folgoc, L. L., Lee, M., Heinrich, M., Misawa, K., Mori, K., McDonagh, S., Hammerla, N. Y., Kainz, B., Glocker, B., and Rueckert, D. (2018). Attention u-net: Learning where to look for the pancreas.
Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18, pages 234–241. Springer.
Si, H., Shi, Z., Hu, X., Wang, Y., and Yang, C. (2020). Image semantic segmentation based on improved deeplab v3 model. International Journal of Modelling, Identification and Control, 36(2):116–125.
Tsang, S.-H. Review: DeepLabv3-Atrous Separable Convolution (Semantic Segmentation) — sh-tsang.medium.com. [Accessed 13-01-2026].
Published
2026-06-01
How to Cite
FARIAS, Marcos V.; FIGUEREDO, Weslley K. R.; SILVA, Aristófanes C.; PAIVA, Anselmo C. de; SALOMÃO, Alice C. C. B.; OLIVEIRA, Marco A. P. de.
Comparative Study of the Combo Loss Function for Deep Endometriosis Segmentation in Magnetic Resonance Imaging Using DeepLabv3. In: BRAZILIAN SYMPOSIUM ON COMPUTING APPLIED TO HEALTH (SBCAS), 26. , 2026, Ouro Preto/MG.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 337-348.
ISSN 2763-8952.
DOI: https://doi.org/10.5753/sbcas.2026.20836.
