CRA-Net: Transformer guided category-relation attention network for diabetic retinopathy grading

被引:2
|
作者
Zang, Feng [1 ]
Ma, Hui [1 ]
机构
[1] Heilongjiang Univ, Sch Elect Engn, Harbin 150080, Peoples R China
关键词
Diabetic retinopathy; Fundus image; Attention block; Transformer;
D O I
10.1016/j.compbiomed.2024.107993
中图分类号
Q [生物科学];
学科分类号
07 ; 0710 ; 09 ;
摘要
Automated grading of diabetic retinopathy (DR) is an important means for assisting clinical diagnosis and preventing further retinal damage. However, imbalances and similarities between categories in the DR dataset make it highly challenging to accurately grade the severity of the condition. Furthermore, DR images encompass various lesions, and the pathological relationship information among these lesions can be easily overlooked. For instance, under different severity levels, the varying contributions of different lesions to accurate model grading differ significantly. To address the aforementioned issues, we design a transformer guided category -relation attention network (CRA-Net). Specifically, we propose a novel category attention block that enhances feature information within the class from the perspective of DR image categories, thereby alleviating class imbalance problems. Additionally, we design a lesion relation attention block that captures relationships between lesions by incorporating attention mechanisms in two primary aspects: capsule attention models the relative importance of different lesions, allowing the model to focus on more "informative"ones. Spatial attention captures the global position relationship between lesion features under transformer guidance, facilitating more accurate localization of lesions. Experimental and ablation studies on two datasets DDR and APTOS 2019 demonstrate the effectiveness of CRA-Net and obtain competitive performance.
引用
收藏
页数:11
相关论文
共 30 条
  • [21] DRG-NET: A graph neural network for computer-aided grading of diabetic retinopathy
    Amritha Abdul Salam
    Manjunatha Mahadevappa
    Asha Das
    Madhu S. Nair
    [J]. Signal, Image and Video Processing, 2022, 16 : 1869 - 1875
  • [22] DRG-NET: A graph neural network for computer-aided grading of diabetic retinopathy
    Salam, Amritha Abdul
    Mahadevappa, Manjunatha
    Das, Asha
    Nair, Madhu S.
    [J]. SIGNAL IMAGE AND VIDEO PROCESSING, 2022, 16 (07) : 1869 - 1875
  • [23] LCAM-Net: Local Context Attention Network for Diabetic Retinopathy Severity Classification
    Alvarado-Carrillo, Dora E.
    Ovalle-Magallanes, Emmanuel
    Dalmau-Cedeno, Oscar S.
    [J]. PATTERN RECOGNITION, MCPR 2024, 2024, 14755 : 355 - 367
  • [24] Fine-grained attention & knowledge-based collaborative network for diabetic retinopathy grading
    Tian, Miao
    Wang, Hongqiu
    Sun, Yingxue
    Wu, Shaozhi
    Tang, Qingqing
    Zhang, Meixia
    [J]. HELIYON, 2023, 9 (07)
  • [25] Resampling-based cost loss attention network for explainable imbalanced diabetic retinopathy grading
    Li, Haiyan
    Dong, Xiaofang
    Shen, Wei
    Ge, Fuhua
    Li, Hongsong
    [J]. COMPUTERS IN BIOLOGY AND MEDICINE, 2022, 149
  • [26] Multi-lesion segmentation guided deep attention network for automated detection of diabetic retinopathy
    Li, Feng
    Sheng, Xinyu
    Wei, Hao
    Tang, Shiqing
    Zou, Haidong
    [J]. Computers in Biology and Medicine, 2024, 183
  • [27] Arithmetic Optimization for Coinciding Diabetic Retinopathy and Diabetic Macular Edema Grading based on Self-Attention Convolutional Neural Network
    Devi, T. M.
    Karthikeyan, P.
    [J]. BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2024, 95
  • [28] DRG-Net: Diabetic Retinopathy Grading Network Using Graph Learning with Extreme Gradient Boosting Classifier
    Poranki, Venkata Kotam Raju
    Rao, B. Srinivasa
    [J]. Informatica (Slovenia), 2024, 48 (02): : 171 - 184
  • [29] CSANet: a lightweight channel and spatial attention neural network for grading diabetic retinopathy with optical coherence tomography angiography
    Ma, Fei
    Liu, Xiao
    Wang, Shengbo
    Li, Sien
    Dai, Cuixia
    Meng, Jing
    [J]. QUANTITATIVE IMAGING IN MEDICINE AND SURGERY, 2024, 14 (02) : 1820 - 1834
  • [30] DAHT-Net: Deformable Attention-Guided Hierarchical Transformer Network Based on Remote Sensing Image Change Detection
    Shi, Gang
    Mei, Yunfei
    Wang, Xiaoli
    Yang, Qingwen
    [J]. IEEE ACCESS, 2023, 11 : 103033 - 103043