EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

HIRD-Net: An Explainable CNN-Based Framework with Attention Mechanism for Diabetic Retinopathy Diagnosis Using CLAHE-D-DoG Enhanced Fundus Images

Muhammad Hassaan Ashraf iD, Muhammad Nabeel Mehmood iD, Musharif Ahmed iD, Dildar Hussain iD, Jawad Khan iD, Younhyun Jung iD, Mohammed Zakariah iD, Deema Mohammed AlSekait iD

Life · 08.09.2025 · 15 (9) · 1411

DOI: 10.3390/life15091411 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Life, online erschienen am 08.09.2025. Verfasst von Muhammad Hassaan Ashraf und 7 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Computing, Riphah International University, Islamabad 46000, Pakistan. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia, King Saud University, Riyadh, Saudi Arabia.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Diabetic Retinopathy (DR) is a leading cause of vision impairment globally, underscoring the need for accurate and early diagnosis to prevent disease progression. Although fundus imaging serves as a cornerstone of Computer-Aided Diagnosis (CAD) systems, several challenges persist, including lesion scale variability, blurry morphological patterns, inter-class imbalance, limited labeled datasets, and computational inefficiencies. To address these issues, this study proposes an end-to-end diagnostic framework that integrates an enhanced preprocessing pipeline with a novel deep learning architecture, Hierarchical-Inception-Residual-Dense Network (HIRD-Net). The preprocessing stage combines Contrast Limited Adaptive Histogram Equalization (CLAHE) with Dilated Difference of Gaussian (D-DoG) filtering to improve image contrast and highlight fine-grained retinal structures. HIRD-Net features a hierarchical feature fusion stem alongside multiscale, multilevel inception-residual-dense blocks for robust representation learning. The Squeeze-and-Excitation Channel Attention (SECA) is introduced before each Global Average Pooling (GAP) layer to refine the Feature Maps (FMs). It further incorporates four GAP layers for multi-scale semantic aggregation, employs the Hard-Swish activation to enhance gradient flow, and utilizes the Focal Loss function to mitigate class imbalance issues. Experimental results on the IDRiD-APTOS2019, DDR, and EyePACS datasets demonstrate that the proposed framework achieves 93.46%, 82.45% and 79.94% overall classification accuracy using only 4.8 million parameters, highlighting its strong generalization capability and computational efficiency. Furthermore, to ensure transparent predictions, an Explainable AI (XAI) approach known as Gradient-weighted Class Activation Mapping (Grad-CAM) is employed to visualize HIRD-Net’s decision-making process.

Schlagwörter

Stichwörter der Autor:innen

Computer-Aided DiagnosisConvolutional Neural NetworkDifference of GaussianGrad-CAMHard-Swishdeep learning in medical imagingdiabetic retinopathy diagnosisexplainable AIfundus image enhancement

Autor:innen und Institutionen

  • Faculty of Computing, Riphah International University, Islamabad 46000, Pakistan.
  • Department of Artificial Intelligence and Data Science, Sejong University, Seoul 05006, Republic of Korea.
  • School of Computing, Gachon University, Seongnam 13120, Republic of Korea.
  • Department of Computer Science and Engineering, College of Applied Studies, King Saud University, P.O. Box 22459, Riyadh 11495, Saudi Arabia.
  • Department of Information Technology, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, P.O. Box 84428, Riyadh 11671, Saudi Arabia.

ORCID-Profile: Muhammad Hassaan Ashraf, Muhammad Nabeel Mehmood, Musharif Ahmed, Dildar Hussain, Jawad Khan, Younhyun Jung, Mohammed Zakariah, Deema Mohammed AlSekait

Förderung

  • Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia
  • King Saud University, Riyadh, Saudi Arabia

Zitieren

APA

Ashraf, M. H., Mehmood, M. N., Ahmed, M., Hussain, D., Khan, J., Jung, Y., Zakariah, M., & AlSekait, D. M. (2025). HIRD-Net: An Explainable CNN-Based Framework with Attention Mechanism for Diabetic Retinopathy Diagnosis Using CLAHE-D-DoG Enhanced Fundus Images. Life, 15(9), 1411. https://doi.org/10.3390/life15091411

Vancouver

Ashraf MH, Mehmood MN, Ahmed M, Hussain D, Khan J, Jung Y, et al. HIRD-Net: An Explainable CNN-Based Framework with Attention Mechanism for Diabetic Retinopathy Diagnosis Using CLAHE-D-DoG Enhanced Fundus Images. Life. 2025;15(9):1411. doi:10.3390/life15091411

BibTeX
@article{Ashraf2025,
  author = {Ashraf, Muhammad Hassaan and Mehmood, Muhammad Nabeel and Ahmed, Musharif and Hussain, Dildar and Khan, Jawad and Jung, Younhyun and Zakariah, Mohammed and AlSekait, Deema Mohammed},
  title = {HIRD-Net: An Explainable CNN-Based Framework with Attention Mechanism for Diabetic Retinopathy Diagnosis Using CLAHE-D-DoG Enhanced Fundus Images},
  journal = {Life},
  year = {2025},
  volume = {15},
  number = {9},
  pages = {1411},
  doi = {10.3390/life15091411}
}
Kostenlos starten →