International Conference on Information and Knowledge Technology

فارسی

Home / شانزدهمین کنفرانس بین المللی فناوری اطلاعات و دانش

A hybrid CNN–transformer framework for retinal disease classification

Authors :

Hanie Zomorrodi¹ Hassan Khotanlou²

1- دانشگاه بوعلی سینا 2- دانشگاه بوعلی سینا

Keywords :

convolutional neural network،retina،transformer encoder

Abstract :

Accurate diagnosis of retinal diseases is essential for preventing visual impairment and blindness. In this study, we propose a deep learning-based framework for automatic multi-class classification of retinal images that can detect 20 ocular diseases at once. The approach starts with preprocessing and improving fundus images, followed by data augmentation to boost the model’s generalization and strength. We extract features using a combined EfficientNet-ConvNeXt framework, which captures both local details and global context. Next, we refine the extracted features with a Transformer Encoder to model relationships across the entire retinal image. Finally, an MLP classifies the input. Experimental results show that our method achieves a Model Score of 0.903, surpassing earlier methods. These findings confirm that combining feature representations from the EfficientNet-ConvNeXt architecture with Transformer-based modeling significantly enhances the accuracy of retinal disease classification.

List of archived papers

GNN-based Topology Feature Extraction for Adaptive 6G Network Slicing

Amirmasoud Sepehrian - Siavash Khorsandi

IT-based and Non-IT-based methods to separate and collect waste

Hoda Harati - Farzad Haghighi-Rad - Reza Yousefi Zenouz

بررسی روش‌ها، مجموعه‌های داده و معیارهای ارزیابی در حوزه‌ی پرسش از متون درون تصویر

کبری فرشیدی - حسن ختن‌لو - محرم منصوری زاده - الهام علی قارداش

AI-based Secure Intrusion Detection Framework for Digital Twin-enabled Critical Infrastructure

Tanisha Patel - Nilesh Kumar Jadav - Tejal Rathod - Sudeep Tanwar - Deepak Garg - Hossein Shahinzadeh

بهبود دقت و کارایی در شبکه‌های عصبی کانولوشنی با استفاده از روش‌های محاسبات تقریبی

محمدرضا رفیعی نژاد - محمدرضا بینش مروستی - سید امیر اصغری

Comparative Study of Deep Reinforcement Learning and Genetic Algorithm Approaches for IoT Machine Learning Job Deployment in Fog Computing

Amir Moazeni - Omid Bushehrian

شناسایی جایگاه مالونیلاسیون در پروتئین‌ها با بهره‌گیری از استخراج ویژگی و تکنیک‌های پردازش زبان طبیعی

حنانه رجبیون - محمد قاسم زاده - وحید رنجبر بافقی

Design and modeling of a waiter robot

Amin Mohammadnejad - Hami Tourajizadeh

Designing an AI-assisted toolbox for fitness activity recognition based on deep CNN

Ali Bidaran - Dr Saeed Sharifian

Beyond One-Hot: CatBoost for Heating and Cooling Load Prediction

Shayan Naghizadeh - Mohammad Saeed Rajabi - Ehsan Nazerfard

more

Samin Hamayesh - Version 42.5.2