The experimental results on the Kvasir-SEG CVC-Clinic and ETIS datasets demonstrate that PRNet achieves Dice coefficients of 9209 9305 and 7419 respectively This new approach improves the segmentation

The Dice coefficients for PRNet on Kvasir-SEG, CVC-Clinic, and ETIS datasets were found to be 92.09%, 93.05%, and 74.19% respectively, based on the experimental results. The use of this new method has greatly enhanced the accuracy of segmentation for polyp regions of varying scales with indistinct boundaries.