Enhancing Medical Image Segmentation through Negative Sample Integration: A Study on Kvasir-SEG and Augmented Datasets
Imon Mahbub, Ahmed Zoha Zim, Talha Imran, K.M Fatin Mesbah, Md Jobayer, Md. Mehedi Hasan Shawon
This study compares U-Net, SegNet, and ResNet for colorectal polyp segmentation, finding U-Net performs best with negative samples included.
bioRxiv · Oct 14, 2025 · PreprintBioengineering