Ordinal Diffusion Models for Color Fundus Images

2026-02-27

AI Summary

This work addresses a key limitation in existing conditional diffusion models for diabetic retinopathy (DR) by recognizing that DR stages form an ordered continuum rather than independent categories. The authors propose an ordinal latent diffusion model that, for the first time, incorporates ordinal constraints within a diffusion framework. Instead of discrete class labels, the model uses a scalar representing disease severity to explicitly capture the continuous pathological progression of DR. Through scalar-based conditioning and interpolation mechanisms, the method generates more realistic retinal images on the EyePACS dataset, achieving substantially lower FID scores across all four DR stages and improving the quadratic weighted kappa coefficient from 0.79 to 0.87. Interpolation experiments further demonstrate the model’s ability to faithfully represent the continuous spectrum of disease severity.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

It has been suggested that generative image models such as diffusion models can improve performance on clinically relevant tasks by offering deep learning models supplementary training data. However, most conditional diffusion models treat disease stages as independent classes, ignoring the continuous nature of disease progression. This mismatch is problematic in medical imaging because continuous pathological processes are typically only observed through coarse, discrete but ordered labels as in ophthalmology for diabetic retinopathy (DR). We propose an ordinal latent diffusion model for generating color fundus images that explicitly incorporates the ordered structure of DR severity into the generation process. Instead of categorical conditioning, we used a scalar disease representation, enabling a smooth transition between adjacent stages. We evaluated our approach using visual realism metrics and classification-based clinical consistency analysis on the EyePACS dataset. Compared to a standard conditional diffusion model, our model reduced the Fr\'echet inception distance for four of the five DR stages and increased the quadratic weighted $\kappa$ from 0.79 to 0.87. Furthermore, interpolation experiments showed that the model captured a continuous spectrum of disease progression learned from ordered, coarse class labels.

Problem

• ordinal diffusion models
• diabetic retinopathy
• disease progression
• medical image generation
• ordered labels

Innovation

• ordinal diffusion model
• color fundus images
• disease progression modeling
• scalar disease representation
• diabetic retinopathy