MEGANet-W: A Wavelet-Driven Edge-Guided Attention Framework for Weak Boundary Polyp Detection
This work addresses automated polyp detection for early colorectal cancer screening, offering incremental improvements in boundary accuracy for medical image segmentation.
The paper tackled the problem of weak and low contrast boundaries in colorectal polyp segmentation by proposing MEGANet-W, a wavelet-driven edge-guided attention framework, which improved mIoU by up to 2.3% and mDice by 1.2% on five public datasets without adding learnable parameters.
Colorectal polyp segmentation is critical for early detection of colorectal cancer, yet weak and low contrast boundaries significantly limit automated accuracy. Existing deep models either blur fine edge details or rely on handcrafted filters that perform poorly under variable imaging conditions. We propose MEGANet-W, a Wavelet Driven Edge Guided Attention Network that injects directional, parameter free Haar wavelet edge maps into each decoder stage to recalibrate semantic features. The key novelties of MEGANet-W include a two-level Haar wavelet head for multi-orientation edge extraction; and Wavelet Edge Guided Attention (W-EGA) modules that fuse wavelet cues with boundary and input branches. On five public polyp datasets, MEGANet-W consistently outperforms existing methods, improving mIoU by up to 2.3% and mDice by 1.2%, while introducing no additional learnable parameters. This approach improves reliability in difficult cases and offers a robust solution for medical image segmentation tasks requiring precise boundary detection.