High-Resolution Boundary Detection for Medical Image Segmentation with Piece-Wise Two-Sample T-Test Augmented Loss
This addresses the limitation of existing loss functions for high-resolution medical applications like automated diagnoses, though it appears incremental as it modifies loss functions rather than introducing a new paradigm.
The paper tackled the problem of poor boundary detection in medical image segmentation by developing a novel loss function, the piece-wise two-sample t-test augmented (PTA) loss, which improved boundary detection power compared to benchmark losses.
Deep learning methods have contributed substantially to the rapid advancement of medical image segmentation, the quality of which relies on the suitable design of loss functions. Popular loss functions, including the cross-entropy and dice losses, often fall short of boundary detection, thereby limiting high-resolution downstream applications such as automated diagnoses and procedures. We developed a novel loss function that is tailored to reflect the boundary information to enhance the boundary detection. As the contrast between segmentation and background regions along the classification boundary naturally induces heterogeneity over the pixels, we propose the piece-wise two-sample t-test augmented (PTA) loss that is infused with the statistical test for such heterogeneity. We demonstrate the improved boundary detection power of the PTA loss compared to benchmark losses without a t-test component.