Laser & Optoelectronics Progress, Volume. 60, Issue 14, 1417001(2023)

Classification of Diabetic Retinopathy with Feature Fusion Network

Shuang Zhao1, Ge Mu1, Wenhua Zhao2、*, and Zhiqing Ma2
Author Affiliations
  • 1Laboratory Management Office, Shandong University of Traditional Chinese Medicine, Jinan 250355, Shandong, China
  • 2College of Intelligence and Information Engineering, Shandong University of Traditional Chinese Medicine, Jinan 250355, Shandong, China
  • show less
    Figures & Tables(11)
    Five categories in the APTOS 2019 dataset. (a) No DR; (b) Mild DR; (c) Moderate DR; (d) Severe DR; (e) Proliferative DR
    Four types of images in the APTOS dataset. (a) Rectangular image, no cropping; (b) rectangular image, lossy vertical cropping; (c) square image, tight cropping; (d) rectangular image, lossy vertical and horizontal cropping
    Preprocessed images
    Overall network architecture
    Multi-scale channel attention module
    Loss curve
    Confusion matrix
    • Table 1. Distribution of APTOS 2019 dataset

      View table

      Table 1. Distribution of APTOS 2019 dataset

      LabelNumber of images
      No DR1805
      Mild DR370
      Moderate DR999
      Severe DR193
      Proliferative DR295
    • Table 2. Network architecture of EfficientNet-B0 (stage1-8)

      View table

      Table 2. Network architecture of EfficientNet-B0 (stage1-8)

      StageOperateork×kOutput sizeNumber of layers
      1Conv3×3256×256×32×1
      2MBConvBlock13×3128×128×16×1
      3MBConvBlock63×3128×128×24×2
      4MBConvBlock65×564×64×40×2
      5MBConvBlock63×332×32×80×3
      6MBConvBlock65×516×16×112×3
      7MBConvBlock65×516×16×192×4
      8MBConvBlock63×38×8×320×1
    • Table 3. Ablation study

      View table

      Table 3. Ablation study

      ModelAccuracyRecallPrecision
      EfficientNet-B083.9262.3378.94
      EfficientNet-B0+Dilated convolution84.7464.6877.08
      Proposed model85.2568.4182.20
    • Table 4. Result comparison with existing researches

      View table

      Table 4. Result comparison with existing researches

      ModelAccuracyRecallPrecision
      ResNet-503473.778
      SE-Net3472.873
      SE-ResNet-503476.144
      Model in Ref.[2380.0680.0081.00
      Model in Ref.[2583.09
      Proposed model85.2568.4182.20
    Tools

    Get Citation

    Copy Citation Text

    Shuang Zhao, Ge Mu, Wenhua Zhao, Zhiqing Ma. Classification of Diabetic Retinopathy with Feature Fusion Network[J]. Laser & Optoelectronics Progress, 2023, 60(14): 1417001

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Medical Optics and Biotechnology

    Received: Aug. 29, 2022

    Accepted: Sep. 23, 2022

    Published Online: Jul. 17, 2023

    The Author Email: Zhao Wenhua (zhaowh0621@163.com)

    DOI:10.3788/LOP222415

    Topics