Modeling Disease Progression with Diffusion-Based Generative Models

Date

2025-8-29

Author

KURT, Meryem Mine

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

13278
views

0
downloads

Cite This

Disease progression modeling in medical imaging presents significant challenges due to the scarcity of longitudinal data and the inherent class imbalance in medical datasets. This thesis introduces a novel conditional diffusion framework for synthesizing realistic disease progression sequences from cross-sectional data, with a focus on ulcerative colitis endoscopic imaging. The proposed approach employs specialized ordinal class embeddings that capture the progressive nature of disease severity, enabling the generation of smooth transitions between discrete Mayo Endoscopic Score levels. Two embedding strategies are developed: a Basic Ordinal Embedder using linear interpolation between severity classes, and an Additive Ordinal Embedder that explicitly models the cumulative nature of pathological features. The framework is built upon Stable Diffusion v1.4 with custom modifications for medical imaging applications, incorporating advanced training techniques including Exponential Moving Average, Min-SNR-γ weighting, and class-balanced sampling. The methodology is evaluated using the LIMUC dataset through comprehensive quantitative metrics. By transforming classification-based datasets into continuous progression models, this framework enables fine-grained control over disease severity and the realistic synthesis of intermediate disease stages. This work addresses the critical limitation of longitudinal data scarcity in medical research and provides a foundation for improved clinical training, treatment planning, and enhanced understanding of progressive medical conditions.

Subject Keywords

medical image synthesis, diffusion models, ulcerative colitis, disease progression modeling

URI

https://hdl.handle.net/11511/115556

Collections

Graduate School of Informatics, Thesis