LHU-Net: A Light Hybrid U-Net for Cost-Efficient, High-Performance Volumetric Medical Image Segmentation

arXiv:2404.05102v1 Announce Type: cross
Abstract: As a result of the rise of Transformer architectures in medical image analysis, specifically in the domain of medical image segmentation, a multitude of hybrid models have been created that merge the advantages of Convolutional Neural Networks (CNNs) and Transformers. These hybrid models have achieved notable success by significantly improving segmentation accuracy. Yet, this progress often comes at the cost of increased model complexity, both in terms of parameters and computational demand. Moreover, many of …

abstract advantages analysis architectures arxiv cnns convolutional neural networks cost cs.cv cs.lg domain eess.iv hybrid image light medical merge networks neural networks performance segmentation transformer transformers type