Structure-Oriented Transformer for retinal diseases grading from OCT images
Structure-Oriented Transformer for retinal diseases grading from OCT images
复制标题
用于根据 OCT 图像对视网膜疾病进行分级的面向结构的 Transformer
DOI：
10.1016/j.compbiomed.2022.106445
复制
发表时间：
2023
影响因子：
7.7
通讯作者：
Liu Jiang
中科院分区：
文献类型：
--
作者：
Shen Junyong;Hu Yan;Zhang Xiaoqing;Gong Yan;Kawasaki Ryo;Liu Jiang
视网膜疾病是导致视力暂时或永久丧失的主要原因。精确的视网膜疾病分级是早期干预或特定治疗方案的先决条件。现有的基于卷积神经网络（CNN）的工作集中在典型的局部结构，不能捕捉长程依赖。但视网膜病变分级更多地依赖于局部病变与整个视网膜的关系，这与自我注意机制是一致的。因此，本文提出了一种新的面向结构的Transformer（SoT）框架，以进一步构建临床数据集上病变和视网膜之间的关系。为了减少对数据量的依赖，我们设计了结构引导作为一个面向模型的过滤器，强调整个视网膜结构和指导关系的建设。然后，我们采用预训练的视觉Transformer，通过迁移学习有效地建模所有特征块的关系。此外，为了最大限度地利用所有输出的标记，提出了一个标记投票分类器，以获得最终的分级结果。我们对一个临床新生血管性黄斑变性（nAMD）数据集进行了广泛的实验。实验证明了SoT组件的有效性，并提高了病变与视网膜之间的关系构建能力，这优于用于nAMD分级的最新方法。在一个公开的视网膜疾病数据集上对算法进行了测试，结果表明该算法具有分类优势和良好的通用性。
Retinal diseases are the leading causes of vision temporary or permanent loss. Precise retinal disease grading is a prerequisite for early intervention or specific therapeutic schedules. Existing works based on Convolutional Neural Networks (CNN) focus on typical locality structures and cannot capture long-range dependencies. But retinal disease grading relies more on the relationship between the local lesion and the whole retina, which is consistent with the self-attention mechanism. Therefore, the paper proposes a novel Structure-Oriented Transformer (SoT) framework to further construct the relationship between lesions and retina on clinical datasets. To reduce the dependence on the amount of data, we design structure guidance as a model-oriented filter to emphasize the whole retina structure and guide relation construction. Then, we adopt the pre-trained vision transformer that efficiently models all feature patches’ relationships via transfer learning. Besides, to make the best of all output tokens, a Token vote classifier is proposed to obtain the final grading results. We conduct extensive experiments on one clinical neovascular Age-related Macular Degeneration (nAMD) dataset. The experiments demonstrate the effectiveness of SoT components and improve the ability of relation construction between lesion and retina, which outperforms the state-of-the-art methods for nAMD grading. Furthermore, we evaluate our SoT on one publicly available retinal diseases dataset, which proves our algorithm has classification superiority and good generality.
登录
查看更多内容
使用卷积神经网络通过 OCT 图像对视网膜病理进行分类
DOI：
10.1109/cosite52651.2021.9649630
发表时间：
2021
期刊：
2021 International Conference on Computer System, Information Technology, and Electrical Engineering (COSITE)
影响因子：
--
作者：
Dewi Annisa Anam;L. Novamizanti;S. Rizal
通讯作者：
S. Rizal
用于医学图像病变检测的面向交互的特征分解
DOI：
10.1007/978-3-031-16437-8_31
发表时间：
2022
期刊：
International Conference on Medical Image Computing and Computer-Assisted Intervention
影响因子：
--
作者：
Junyong Shen;Yan Hu;Xiaoqin Zhang;Zhongxi Qiu;Tingming Deng;Yanwu Xu;Jiang Liu
通讯作者：
Jiang Liu
影响因子：
3.4
作者：
Srinivasan, Pratul P.;Kim, Leo A.;Farsiu, Sina
通讯作者：
Farsiu, Sina
影响因子：
3.9
作者：
Faatz, Henrik;Farecki, Marie-Louise;Pauleikhoff, Daniel
通讯作者：
Pauleikhoff, Daniel