Efficient Deep Learning Approaches for Processing Ultra-Widefield Retinal Imaging

2025-03-23

AI Summary

To address the challenge of automated disease classification from ultra-widefield retinal imaging (UWF) in resource-constrained clinical settings—where conventional color fundus photography (CFP)-based models face accuracy limitations—this work pioneers the systematic integration of efficient deep learning paradigms into UWF analysis. We propose a lightweight network architecture, UWF-specific strategic data augmentation, and a multi-model ensemble framework, jointly optimizing diagnostic accuracy and edge-deployment feasibility. Our approach achieves comparable classification accuracy to state-of-the-art GPU-intensive models (within ±0.5% absolute difference), while accelerating inference by 3.2× and reducing GPU memory consumption by 76%. These advances significantly enhance the practicality and accessibility of AI-assisted diagnosis in primary care facilities. By enabling high-fidelity UWF interpretation with minimal computational overhead, our framework establishes a scalable, low-resource paradigm for clinical translation of ultra-widefield retinal imaging.

Technology Category

• Machine Learning: Hardware-aware ML
• Computer Vision: Multi-modal Vision
• Humans and AI: Crowd Sourcing and Human Computation

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Deep learning has emerged as the predominant solution for classifying medical images. We intend to apply these developments to the ultra-widefield (UWF) retinal imaging dataset. Since UWF images can accurately diagnose various retina diseases, it is very important to clas sify them accurately and prevent them with early treatment. However, processing images manually is time-consuming and labor-intensive, and there are two challenges to automating this process. First, high perfor mance usually requires high computational resources. Artificial intelli gence medical technology is better suited for places with limited medical resources, but using high-performance processing units in such environ ments is challenging. Second, the problem of the accuracy of colour fun dus photography (CFP) methods. In general, the UWF method provides more information for retinal diagnosis than the CFP method, but most of the research has been conducted based on the CFP method. Thus, we demonstrate that these problems can be efficiently addressed in low performance units using methods such as strategic data augmentation and model ensembles, which balance performance and computational re sources while utilizing UWF images.

Problem

• Develop efficient deep learning for ultra-widefield retinal image classification
• Address high computational needs in low-resource medical settings
• Improve accuracy over traditional color fundus photography methods

Innovation

• Strategic data augmentation enhances UWF image processing
• Model ensembles balance performance and computational resources
• Deep learning optimizes low-performance unit efficiency