Skip to main content
Have a personal or library account? Click to login
Integrated Deep Learning Framework for Retinal Disease Detection Using Advanced CNN and Imaging Analysis Cover

Integrated Deep Learning Framework for Retinal Disease Detection Using Advanced CNN and Imaging Analysis

International Journal on Smart Sensing and Intelligent Systems
Volume 19 (2026): Issue 1 (January 2026)
By:  and    
Open Access
|Jul 2026
Figures & tables

Figures & Tables

Figure 1:

TL-Pretr-CNN framework for automated retinal disease detection and classification. CNNs, convolutional neural networks; DR, diabetic retinopathy; Grad-CAM, gradient-weighted class activation mapping; TL-Pretr-CNN, transfer learning with pre-trained CNN models.

Figure 2:

TLPretrCNN workflow for multimodal retinal diagnosis. CNNs, convolutional neural networks; DR, diabetic retinopathy; GAP, global average pooling; TL-Pretr-CNN, transfer learning with pre-trained CNN models.

Figure 3:

Progressive feature evolution across TLPretrCNN residual stages. CNNs, convolutional neural networks; TLPretrCNN, transfer learning with pre-trained CNN models.

Figure 4:

Multi-class GradCAM visualization for three diagnostic categories (A) Healthy, (B) Mild pathology, (C) Moderate/severe pathology. GradCAM, gradient-weighted class activation mapping.

Figure 5:

CDA comparison across disease severity levels of DR. CDA, clinical diagnostic accuracy; DR, diabetic retinopathy; Grad-CAM, gradient-weighted class activation mapping; IDRiD-Net, improved deep retinal intelligence detector network; TL-Pretr-CNN, transfer learning with pre-trained CNN models.

Figure 6:

LLP evaluation plot. Grad-CAM, gradient-weighted class activation mapping; IDRiD-Net, improved deep retinal intelligence detector network; TL-Pretr-CNN, transfer learning with pre-trained CNN models.

Figure 7:

SGC line chart across disease grades. DR, diabetic retinopathy; SGC, severity grading concordance.

Figure 8:

ECS evaluation plot. ECS, explainability confidence score; Grad-CAM, gradient-weighted class activation mapping; IDRiD-Net, improved deep retinal intelligence detector network; TL-Pretr-CNN, transfer learning with pre-trained CNN models.

Algorithm 1: TL-Pretr-CNN—Transfer Learning Framework for Multimodal Medical Image Analysis
Input: Dataset D = {If, IMRI}, Pre-trained CNN Model M (e.g., ResNet50, EfficientNet), Labels Y
Output: Predicted Labels y^ \hat y , Grad-CAM heatmaps, evaluation metrics
Step 1: Data Preparation
  For each Img in D:
    Preproc (resize, norm)
  Augment (rot, flip)
Step 2: Load Pre-trained CNN Model
  Load M (ImageNet): M = fInet
  Freeze base layers: θbasefrozen {\boldsymbol {\theta}}_{{\boldsymbol {base}}}^{{\boldsymbol {frozen}}}
  Add custom head: θtop
Step 3: Transfer Learning – Two-Phase Training
   θ=θbasefrozen∪θtop {\boldsymbol {\theta}} = {\boldsymbol {\theta}}_{{\boldsymbol {base}}}^{{\boldsymbol {frozen}}} \cup {{\boldsymbol {\theta}}_{{\boldsymbol {top}}}}
  For each epoch in training_epochs:
   For each mini-batch (Ib, Yb) in the training set:
   Forward Pass:
    F = FInet (Ib) # Feature extraction
     y^=Ffine tune(F)# Classification head \matrix{{\hat y = {F_{fine\,tune(F)}}} \hfill & {\# \,{\rm{Classification}}\,{\rm{head}}} \hfill \cr}
    Compute Loss: L(y^,yb) {\cal L}(\hat y,{y_b})
    Backward Pass:
      update θtop (fine-tune top layers)
Step 4: Deep Feature Extraction with GAP
  For each input I_pre in the test set:
    Fi = CNN_intermediate_layer(Ipre)
zi=GAP(Fi)=1H′W′∑h=1H′∑w=1W′Fih,w,: {z_i} = GAP({F_i}) = {1 \over {H{'}W{'}}}\sum\limits_{h = 1}^{H{'}} {\sum\limits_{w = 1}^{W{'}} {{F_i}\left[ {h,w,:} \right]}}
    #Apply Global Average Pooling:
Step 5: Model Evaluation
    Predict on test set: y^test=model(Itest) {\hat y_{test}} = {\rm{model}}({I_{test}})
      Compute metrics:
        Accuracy = correct/total
        AUC = area under ROC curve
        Sensitivity = TP / (TP + FN)
Step 6: Explainability via Grad-CAM
    For y^ \hat y in Y^_test \hat Y\_{\rm{test}} :
      Gen GCAM from the final conv
      Overlay GCAM on the original image
Step 7: Output Integration
    If eval_m ≥ th:
      Intg mdl → TH_DSS
    Else:
      Retrn y^ \hat y , Ptune

Hyperparameters

HyperparameterSelected valueNotes/optimization rationale
Batch size32Balances GPU memory and gradient stability
Learning rate1 × 10−4 (cosine annealing)Smooth convergence; avoids oscillation or plateauing
OptimizerAdamW (weight decay = 0.01)Strong generalization compared to Adam
Epochs40Trade-off between performance gain and overfitting
Fine-tuning depthLast three convolutional blocksAdapts pre-trained features to retinal pathology
Data augmentationCLAHE, ±20° rotation, random crop, brightness/contrastImproves model generalization and robustness by increasing the diversity of the training data and reducing overfitting.

Data preprocessing steps in retinal disease

Original image
Iinorm I_i^{norm}
Rθ
Fp
Zs

Comparative metrics, statistical significance and inference time

ModelAccuracy (%)LLP (IoU)SGC (QWK)ECSSensitivitySpecificityAUCp-value vs TL-Pretr-CNNInference Time (ms/image)
IDRiD-Net78.3 ± 1.20.62 ± 0.030.71 ± 0.020.670.750.800.84<0.058
DeepMedic80.5 ± 1.00.66 ± 0.020.75 ± 0.020.700.780.820.86<0.0510
EfficientNet + Grad-CAM84.1 ± 0.90.71 ± 0.020.79 ± 0.010.740.810.850.89<0.0514
TL-Pretr-CNN88.6 ± 0.80.78 ± 0.010.85 ± 0.010.810.860.890.93—12
DOI: https://doi.org/10.2478/ijssis-2026-0048 | Journal eISSN: 1178-5608
Journal RSS Feed
Language: English
Submitted on: May 20, 2025
Published on: Jul 18, 2026
Published by: Macquarie University, Australia
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year
Keywords:
retina vision net,
diabetic retinopathy,
transfer learning,
CNN,
teleophthalmology,
fundus images
Related subjects:
Engineering,
Introductions and overviews,
Engineering, other

© 2026 G. Sudha, S. Akila, published by Macquarie University, Australia
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Volume 19 (2026): Issue 1 (January 2026)
Previous article