Exploring the Effectiveness of Deep Features from Domain-Specific Foundation Models in Retinal Image Synthesis
By: Zuzanna Skorniewska, Bartlomiej W. Papiez
Potential Business Impact:
Makes fake eye pictures look more real.
The adoption of neural network models in medical imaging has been constrained by strict privacy regulations, limited data availability, high acquisition costs, and demographic biases. Deep generative models offer a promising solution by generating synthetic data that bypasses privacy concerns and addresses fairness by producing samples for under-represented groups. However, unlike natural images, medical imaging requires validation not only for fidelity (e.g., Fr\'echet Inception Score) but also for morphological and clinical accuracy. This is particularly true for colour fundus retinal imaging, which requires precise replication of the retinal vascular network, including vessel topology, continuity, and thickness. In this study, we in-vestigated whether a distance-based loss function based on deep activation layers of a large foundational model trained on large corpus of domain data, colour fundus imaging, offers advantages over a perceptual loss and edge-detection based loss functions. Our extensive validation pipeline, based on both domain-free and domain specific tasks, suggests that domain-specific deep features do not improve autoen-coder image generation. Conversely, our findings highlight the effectiveness of con-ventional edge detection filters in improving the sharpness of vascular structures in synthetic samples.
Similar Papers
When Do Domain-Specific Foundation Models Justify Their Cost? A Systematic Evaluation Across Retinal Imaging Tasks
Image and Video Processing
Smaller computer models see eye diseases better.
FundusGAN: A Hierarchical Feature-Aware Generative Framework for High-Fidelity Fundus Image Generation
Image and Video Processing
Creates realistic eye pictures for better eye disease diagnosis.
Functional Localization Enforced Deep Anomaly Detection Using Fundus Images
CV and Pattern Recognition
Finds eye diseases in pictures better.