Search:
Match:
1 results

PathoSyn: AI for MRI Image Synthesis

Published:Dec 29, 2025 01:13
1 min read
ArXiv

Analysis

This paper introduces PathoSyn, a novel generative framework for synthesizing MRI images, specifically focusing on pathological features. The core innovation lies in disentangling the synthesis process into anatomical reconstruction and deviation modeling, addressing limitations of existing methods that often lead to feature entanglement and structural artifacts. The use of a Deviation-Space Diffusion Model and a seam-aware fusion strategy are key to generating high-fidelity, patient-specific synthetic datasets. This has significant implications for developing robust diagnostic algorithms, modeling disease progression, and benchmarking clinical decision-support systems, especially in scenarios with limited data.
Reference

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.