Hierarchical Diffusion Framework for Pseudo-Healthy Brain MRI Inpainting with Enhanced 3D Consistency

2025-07-23

AI Summary

To address the volume discontinuity in 2D slice-based models and the excessive data requirements of 3D models for pseudo-healthy brain MRI restoration, this paper proposes a two-stage vertical-view hierarchical diffusion framework. First, axial and coronal 2D diffusion models are independently trained to capture planar anatomical details; then, adaptive spatial resampling enforces cross-plane consistency and enables hierarchical 3D reconstruction. By avoiding full 3D modeling, the method circumvents reliance on large-scale annotated datasets while preserving high in-plane fidelity and significantly improving volumetric coherence. Experiments demonstrate superior realism and anatomical integrity compared to existing state-of-the-art methods, particularly in low-data medical imaging scenarios. The source code is publicly available.

Technology Category

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

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Pseudo-healthy image inpainting is an essential preprocessing step for analyzing pathological brain MRI scans. Most current inpainting methods favor slice-wise 2D models for their high in-plane fidelity, but their independence across slices produces discontinuities in the volume. Fully 3D models alleviate this issue, but their high model capacity demands extensive training data for reliable, high-fidelity synthesis -- often impractical in medical settings. We address these limitations with a hierarchical diffusion framework by replacing direct 3D modeling with two perpendicular coarse-to-fine 2D stages. An axial diffusion model first yields a coarse, globally consistent inpainting; a coronal diffusion model then refines anatomical details. By combining perpendicular spatial views with adaptive resampling, our method balances data efficiency and volumetric consistency. Our experiments show our approach outperforms state-of-the-art baselines in both realism and volumetric consistency, making it a promising solution for pseudo-healthy image inpainting. Code is available at https://github.com/dou0000/3dMRI-Consistent-Inpaint.

Problem

• Improves 3D consistency in pseudo-healthy brain MRI inpainting
• Balances data efficiency and volumetric synthesis fidelity
• Overcomes slice-wise discontinuity in 2D inpainting methods

Innovation

• Hierarchical diffusion framework for 3D MRI inpainting
• Two perpendicular coarse-to-fine 2D diffusion stages
• Combines axial and coronal views for volumetric consistency