همایش بین المللی هوش مصنوعی

صفحه اصلی / دومین همایش بین المللی هوش مصنوعی

AI-CADx in Retinal Disease: Integrating Explainability, Privacy, and Scalability for Next-Generation Telemedicine

نویسندگان :

Mohammad Shojaeinia¹ Hamid Moghaddasi²

1- دانشگاه علوم پزشکی شهید بهشتی 2- دانشگاه علوم پزشکی شهید بهشتی

کلمات کلیدی :

Artificial intelligence،computer-aided diagnosis،ophthalmology،deep learning،explainable AI (XAI)،multimodal fusion،privacy-preserving computation

چکیده :

The evolution of Artificial Intelligence–based Computer-Aided Diagnostic (AI-CADx) systems is exerting a profound influence on contemporary medical diagnostics, shifting their role from auxiliary support tools for clinical specialists to scalable, population-level screening solutions. This transition has been particularly consequential in telemedicine environments and resource-constrained healthcare settings. The present text analyzes the developmental trajectory of AI-CADx systems, with emphasis on advances in model architectures, implementation methodologies, and real-world clinical integration, using retinal disease assessment as a representative application domain. Recent research demonstrates that key innovations are progressively overcoming long-standing barriers to clinical adoption. The fusion of multimodal imaging data with Explainable AI (XAI) methodologies, such as Grad-CAM++, substantially enhances model interpretability by making decision pathways more transparent, thereby strengthening clinician confidence in automated outputs. In parallel, emerging privacy-preserving computational approaches, including encryption, support secure processing of sensitive patient data while maintaining compliance with confidentiality requirements. The effectiveness of these synergistic strategies is illustrated by systems such as MadhuNetri, which report high sensitivity and specificity in identifying referable diabetic retinopathy and exemplify the potential of AI-driven platforms to alleviate workforce shortages and broaden access to high-quality ophthalmic care. Despite these advances, the routine clinical deployment of AI-CADx systems continues to face significant challenges. Variability in regulatory requirements, unresolved issues related to algorithmic bias, and the need for comprehensive, human-centered governance frameworks remain critical considerations. Addressing these factors is essential to ensuring the long-term transparency, fairness, and reliability of AI-enabled diagnostic technologies.

لیست مقالات

لیست مقالات بایگانی شده

Fibroglandular Tissue Classification in Breast MRI: A Comparative Study of Automated Decision Strategies

Meysam Khalaj - Arvin Arian - Ala Torabi - Nasrin Ahmadinejad - Masoumeh Gity - Seyedeh Nooshin Miratashi Yazdi - Mohammad Pooya Afshari - Melika Sadeghi Tabrizi - Hamid Soltanian-Zadeh