Application of Deep Learning Models for Semantic Segmentation of Colonoscopy Images

• Rubens M. G. Aguiar CEIA
• Michel H. Scheeren UTFPR
• Sandro L. de Araujo Junior UTFPR
• Eduardo Mendes UTFPR
• Pedro L. de Paula Filho UTFPR
• Ricardo A. P. Franco UFG

DOI: https://doi.org/10.5753/sbcas.2024.2257

Abstract

Most cases of colorectal cancer originate from colorectal polyps. An increase in the polyp detection rate reduces the risk of developing this pathology. Therefore, the implementation of neural networks to detect and segment such polyps represents a strategy that generates impactful results. Therefore, there arises a need to compare the different models available in the literature and identify the best options for polyp segmentation. Indeed, from the Kvasir dataset, models with potential were selected, and their performance training and evaluations were carried out. Effectively, 8 models were trained, among which the most efficient architectures for the segmentation of polyps in the dataset were identified, for example, the ESFPNet model achieved the highest DICE (0.9008).

References

Chang, Q., Ahmad, D., Toth, J., Bascom, R., and Higgins, W. E. (2023). Esfpnet: Efficient deep learning architecture for real-time lesion segmentation in autofluorescence bronchoscopic video. In Medical Imaging 2023, Progress in Biomedical Optics and Imaging Proceedings of SPIE, United States. SPIE.

Duc, N. T., Oanh, N. T., Thuy, N. T., Triet, T. M., and Dinh, V. S. (2022). Colonformer: An efficient transformer based method for colon polyp segmentation. IEEE Access, 10:80575–80586.

Dumitru, R.-G., Peteleaza, D., and Craciun, C. (2023). Using duck-net for polyp image segmentation. Scientific Reports.

Faceli, K., Lorena, A. C., Gama, J., and Carvalho, A. C. P. L. F. D. (2011). Inteligência Artificial: uma abordagem de aprendizado de máquina. LTC, Rio de Janeiro, RJ.

Guo, Y., Bernal, J., and Matuszewski, B. J. (2020). Polyp segmentation with fully convolutional deep neural networks extended evaluation study. Journal of Imaging.

Guo, Y., Liu, Y., Georgiou, T., and Lew, M. S. (2017). A review of semantic segmentation using deep neural networks. International Journal of Multimedia Information Retrieval.

Gupta, M. and Mishra, A. (2024). A systematic review of deep learning based image segmentation to detect polyp. Artificial Intelligence Review.

Hossain, M., Karuniawati, H., Jairoun, A., Urbi, Z., Ooi, J., John, A., Lim, Y., Kibria, K., Mohiuddin, A., Ming, L., Goh, K., and Hadi, M. (2022). Colorectal cancer: A review of carcinogenesis, global epidemiology, current challenges, risk factors, preventive and treatment strategies. Cancers (Basel).

Huang, C.-H., Wu, H.-Y., and Lin, Y.-L. S. (2021). Hardnet-mseg: A simple encoderdecoder polyp segmentation neural network that achieves over 0.9 mean dice and 86 fps. ArXiv, abs/2101.07172.

Jha, D., Smedsrud, P. H., Riegler, M. A., Halvorsen, P., de Lange, T., Johansen, D., and Johansen, H. D. (2020). Kvasir-seg: A segmented polyp dataset.

Krishnendu, S., Geetha, S., and Gopakumar, G. (2020). A review on polyp detection and segmentation in colonoscopy images using deep learning. International Journal of Engineering Research & Technology.

Liao, T.-Y., Yang, C.-H., Lo, Y.-W., Lai, K.-Y., Shen, P.-H., and Lin, Y.-L. (2022). Hardnet-dfus: An enhanced harmonically-connected network for diabetic foot ulcer image segmentation and colonoscopy polyp segmentation.

Liu, X., Song, L., Liu, S., and Zhang, Y. (2021). A review of deep-learning-based medical image segmentation methods. Sustainability, 13(3).

Marques, A. F., Marques, K. F., dos Santos Beraldo, M. N. M., Lima, T. B., Sassaki, L. Y., and Beraldo, R. F. (2023). Inteligência artificial na colonoscopia no rastreio do câncer colorretal: revisão de literatura. Brazilian Journal of Health Review.

Rahman, M. M. and Marculescu, R. (2023). Medical image segmentation via cascaded attention decoding. In 2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pages 6211–6220.

Silva, J., Histace, A., Romain, O., Dray, X., and Granado, B. (2014). Toward embedded detection of polyps in wce images for early diagnosis of colorectal cancer. International Journal of Computer Assisted Radiology and Surgery.

Tomar, N. K., Jha, D., Bagci, U., and Ali, S. (2022). Tganet: Text-guided attention for improved polyp segmentation. In Medical Image Computing and Computer Assisted Intervention – MICCAI 2022, pages 151–160, Cham. Springer Nature Switzerland.

Trinh, Q.-H. (2023). c. In 2023 IEEE 36th International Symposium on Computer-Based Medical Systems (CBMS), pages 742–747, Los Alamitos, CA, USA. IEEE Computer Society.

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A., Kaiser, L., and Polosukhin, I. (2017). Attention is all you need. In Guyon, I., Luxburg, U. V., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., and Garnett, R., editors, Advances in Neural Information Processing Systems, volume 30. Curran Associates, Inc.

Vázquez, D., Bernal, J., Sánchez, F. J., Fernández-Esparrach, G., López, A. M., Romero, A., Drozdzal, M., and Courville, A. C. (2016). A benchmark for endoluminal scene segmentation of colonoscopy images. Journal of Healthcare Engineering, 2017.

Wang, J., Huang, Q., Tang, F., Meng, J., Su, J., and Song, S. (2022). Stepwise feature fusion: Local guides global. In Medical Image Computing and Computer Assisted Intervention – MICCAI 2022, pages 110–120, Cham. Springer Nature Switzerland.