KAC-Unet: A Medical Image Segmentation With the Adaptive Group Strategy and Kolmogorov-Arnold Network

In the field of deep learning-based medical image segmentation, convolutional neural networks (CNNs) extract image features by combining linear convolutional layers with nonlinear activation functions. However, excessive stacking of linear layers in the network limits the model’s ability to capture fine-grained details. In addition, the feature distribution imbalance caused by the traditional fixed grouping strategy (FGS) can affect the deep model’s capacity to perceive the overall structure of the image. To address these challenges, we propose a medical image segmentation framework, called Kolmogorov-Arnold Network with the adaptive group strategy and contextual Transformer based on Unet (KAC-Unet). First, we propose the adaptive group strategy (AGS) to balance the grouping of different input channels, alleviating the performance degradation caused by differences in group information. Then, we propose the Shift Tokenized Kolmogorov-Arnold Network (KAN) Block to capture complex features in medical images through flexible nonlinear transformations and shift operations. Extensive experiments are conducted on three medical image segmentation datasets. The results demonstrate the effectiveness and superiority of our proposed method compared with state-of-the-art algorithms.

I. Introduction

Medical image segmentation is significant for modern clinical diagnosis and therapeutic intervention [1], [2], [3]. The accuracy and efficiency of segmentation substantially elevate the quality of clinical decision-making and significantly optimize patient treatment results. This technology empowers clinicians to more precisely identify and localize lesion areas facilitating the development of highly personalized for patients [4], [5], [6], [7], [8].