arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2404.16346eess.IVcs.AIcs.CV
收藏

具有全局推理的轻量级视网膜层分割

Light-weight Retinal Layer Segmentation with Global Reasoning

School of Mechanical Engineering, Shandong University(山东大学机械工程学院);Joint SDU-NTU Centre for Artificial Intelligence Research (C-FAIR)(山东大学-南洋理工大学人工智能联合研究中心);Fondazione Bruno Kessler(布鲁诺·凯塞勒基金会)

Xiang He, Weiye Song, Yiming Wang, Fabio Poiesi, Ji Yi, Manishi Desai, Quanqing Xu, Kongzheng Yang, Yi Wan

AI总结:

提出轻量级视网膜层分割网络LightReSeg,通过编码器多尺度特征提取与Transformer实现全局推理,解码器采用多尺度非对称注意力模块保留语义信息,仅3.3M参数即超越105.7M的TransUnet。

摘要:

使用医学图像(如光学相干断层扫描(OCT)图像)进行自动视网膜层分割,是诊断眼科疾病的重要工具。然而,由于图像中存在低对比度和血流噪声,实现准确分割具有挑战性。此外,算法应具备轻量级特性,以便在实际临床应用中部署。因此,需要设计一种高性能的轻量级网络用于视网膜层分割。本文提出LightReSeg用于视网膜层分割,可应用于OCT图像。具体而言,该方法采用编码器-解码器结构,编码器部分采用多尺度特征提取和Transformer块,以充分利用所有尺度特征图的语义信息,并使特征具备更好的全局推理能力;解码器部分,我们设计了多尺度非对称注意力(MAA)模块,以保留每个编码器尺度的语义信息。实验表明,与具有105.7M参数的当前最先进方法TransUnet相比,我们的方法仅用3.3M参数,在我们收集的数据集和另外两个公开数据集上均实现了更好的分割性能。

英文摘要:

Automatic retinal layer segmentation with medical images, such as optical coherence tomography (OCT) images, serves as an important tool for diagnosing ophthalmic diseases. However, it is challenging to achieve accurate segmentation due to low contrast and blood flow noises presented in the images. In addition, the algorithm should be light-weight to be deployed for practical clinical applications. Therefore, it is desired to design a light-weight network with high performance for retinal layer segmentation. In this paper, we propose LightReSeg for retinal layer segmentation which can be applied to OCT images. Specifically, our approach follows an encoder-decoder structure, where the encoder part employs multi-scale feature extraction and a Transformer block for fully exploiting the semantic information of feature maps at all scales and making the features have better global reasoning capabilities, while the decoder part, we design a multi-scale asymmetric attention (MAA) module for preserving the semantic information at each encoder scale. The experiments show that our approach achieves a better segmentation performance compared to the current state-of-the-art method TransUnet with 105.7M parameters on both our collected dataset and two other public datasets, with only 3.3M parameters.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑