SegKAN: High-Resolution Medical Image Segmentation with Long-Distance Dependencies

TL;DR

SegKAN is a novel model for high-resolution medical image segmentation that enhances vessel integrity by improving embedding modules and capturing long-distance dependencies, leading to better segmentation accuracy.

Contribution

The paper introduces a new convolutional structure and a method to transform spatial relationships into temporal ones to improve vessel segmentation in noisy CT images.

Findings

Dice score improved by 1.78% over state-of-the-art

Effective noise smoothing and gradient stability in embeddings

Enhanced capture of long-distance dependencies in segmentation

Abstract

Hepatic vessels in computed tomography scans often suffer from image fragmentation and noise interference, making it difficult to maintain vessel integrity and posing significant challenges for vessel segmentation. To address this issue, we propose an innovative model: SegKAN. First, we improve the conventional embedding module by adopting a novel convolutional network structure for image embedding, which smooths out image noise and prevents issues such as gradient explosion in subsequent stages. Next, we transform the spatial relationships between Patch blocks into temporal relationships to solve the problem of capturing positional relationships between Patch blocks in traditional Vision Transformer models. We conducted experiments on a Hepatic vessel dataset, and compared to the existing state-of-the-art model, the Dice score improved by 1.78%. These results demonstrate that the proposed new structure effectively enhances the segmentation performance of high-resolution extended objects. Code will be available at https://github.com/goblin327/SegKAN