Tumor and Polyp Segmentation in Medical Images Using Deep Learning

We built a tumor and polyp segmentation tool integrating a U-Net inspired architecture and deep learning. The primary objective was to lessen the burden of manual segmentation in medical image review.   From the experiments we conducted, we demonstrated that our model accurately completed segmentation and defined features for accurate clinical judgement and diagnosis.   We demonstrated that a modified U-Net architecture learns the ability to perform segmentation tasks in medical imaging. We have demonstrated that this approach has the potential to be a strong tool in medical imaging and segmentation, particularly with the optimization of large image segmentation in a clinical setting.

[1] Y. Zhang, H. Li, and M. Sun, “Explainable Deep Learning for Polyp Segmentation using Grad-CAM,” 2025.   [2] T. Chen, S. Liu, and X. Wang, “Dilated U-Net for Small Object Medical Segmentation,” Computers in Biology and Medicine, 2024.   [3] J. Ma, Y. He, F. Li, et al., “Medical Image Segmentation using Deep Learning: A Survey,” Artificial Intelligence Review, 2023.   [4] P. Bilic, P. F. Christ, E. Vorontsov, et al., “The Liver Tumor Segmentation Benchmark (LiTS),” Medical Image Analysis, 2023.   [5] X. Huang, L. Wang, and J. Zhang, “Hybrid CNN-Transformer Network for Tumor Segmentation,” Biomedical Signal Processing and Control, 2023.   [6] A. Hatamizadeh, Y. Tang, D. Nath, et al., “UNETR: Transformers for 3D Medical Image Segmentation,” WACV, 2022.   [7] S. Minaee, Y. Boykov, F. Porikli, et al., “Image Segmentation using Deep Learning: A Survey,” IEEE TPAMI, 2022.   [8] D. Jha, S. Ali, K. Emanuelsen, et al., “A Comparative Study on Colorectal Polyp Segmentation using Deep Learning,” IEEE Access, 2021.   [9] S. Ali, D. Jha, N. Ghatwary, et al., “An Objective Comparison of Deep Learning-based Polyp Segmentation Methods,” Scientific Reports, 2021.   [10] H. Cao, Y. Wang, J. Chen, et al., “Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation,” 2021.