PathoSyn: Imaging-Pathology MRI Synthesis via Disentangled Deviation Diffusion

2025-12-28

AI Summary

Existing MRI synthesis models often suffer from anatomical distortions or discontinuities due to feature entanglement. To address this, we propose an anatomy–pathology disentanglement paradigm that separately models stable anatomical manifolds and pathological lesions as localized intensity deviations. We introduce a novel dual-path framework: “anatomy-deterministic reconstruction” coupled with “pathology-deviation stochastic generation.” Our method incorporates a deviation-space diffusion model, a suture-aware fusion mechanism, and an inference-time stabilization module—enabling anatomical fidelity, controllable lesion synthesis, and interpretable generation. Evaluated on a tumor MRI benchmark, our approach significantly outperforms end-to-end diffusion and mask-conditioned baselines, achieving +12.3% improvement in perceptual realism (FID) and +18.7% in anatomical consistency (Dice). The framework supports few-shot diagnostic model training and counterfactual disease progression simulation.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

We present PathoSyn, a unified generative framework for Magnetic Resonance Imaging (MRI) image synthesis that reformulates imaging-pathology as a disentangled additive deviation on a stable anatomical manifold. Current generative models typically operate in the global pixel domain or rely on binary masks, these paradigms often suffer from feature entanglement, leading to corrupted anatomical substrates or structural discontinuities. PathoSyn addresses these limitations by decomposing the synthesis task into deterministic anatomical reconstruction and stochastic deviation modeling. Central to our framework is a Deviation-Space Diffusion Model designed to learn the conditional distribution of pathological residuals, thereby capturing localized intensity variations while preserving global structural integrity by construction. To ensure spatial coherence, the diffusion process is coupled with a seam-aware fusion strategy and an inference-time stabilization module, which collectively suppress boundary artifacts and produce high-fidelity internal lesion heterogeneity. PathoSyn provides a mathematically principled pipeline for generating high-fidelity patient-specific synthetic datasets, facilitating the development of robust diagnostic algorithms in low-data regimes. By allowing interpretable counterfactual disease progression modeling, the framework supports precision intervention planning and provides a controlled environment for benchmarking clinical decision-support systems. Quantitative and qualitative evaluations on tumor imaging benchmarks demonstrate that PathoSyn significantly outperforms holistic diffusion and mask-conditioned baselines in both perceptual realism and anatomical fidelity. The source code of this work will be made publicly available.

Problem

• Synthesizes MRI images with pathological features
• Preserves anatomical structure while modeling disease deviations
• Generates patient-specific synthetic data for diagnostic algorithms

Innovation

• Disentangled anatomical reconstruction and stochastic deviation modeling
• Deviation-Space Diffusion Model for pathological residual distribution
• Seam-aware fusion and stabilization to suppress boundary artifacts