Conditional diffusion models for guided anomaly detection in brain images using fluid-driven anomaly randomization

2025-06-11

AI Summary

Existing reconstruction-based unsupervised anomaly detection methods—e.g., diffusion models—rely on the strong assumption that anomalies are inherently unreconstructable; however, this assumption frequently fails in brain MRI, where healthy tissue reconstruction is imprecise and pathological regions are inadequately suppressed. To address this, we propose a weakly supervised conditional diffusion model trained exclusively on healthy images. Our key innovation is a fluid-dynamics-driven anomaly randomization mechanism that synthesizes anatomically consistent pseudo-pathological images. Leveraging these, we introduce the first pseudo-pathology-guided conditional diffusion framework, enabling precise reconstruction of healthy anatomy and accurate anomaly localization. The method integrates fluid-dynamic modeling, pathology segmentation map augmentation, and weakly supervised learning. Evaluated on ATLAS and multiple synthetic benchmarks, it significantly outperforms VAEs and both unconditional and conditional latent diffusion models, and surpasses paired supervised inpainting methods—requiring matched healthy/pathological image pairs—on most metrics.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Supervised machine learning has enabled accurate pathology detection in brain MRI, but requires training data from diseased subjects that may not be readily available in some scenarios, for example, in the case of rare diseases. Reconstruction-based unsupervised anomaly detection, in particular using diffusion models, has gained popularity in the medical field as it allows for training on healthy images alone, eliminating the need for large disease-specific cohorts. These methods assume that a model trained on normal data cannot accurately represent or reconstruct anomalies. However, this assumption often fails with models failing to reconstruct healthy tissue or accurately reconstruct abnormal regions i.e., failing to remove anomalies. In this work, we introduce a novel conditional diffusion model framework for anomaly detection and healthy image reconstruction in brain MRI. Our weakly supervised approach integrates synthetically generated pseudo-pathology images into the modeling process to better guide the reconstruction of healthy images. To generate these pseudo-pathologies, we apply fluid-driven anomaly randomization to augment real pathology segmentation maps from an auxiliary dataset, ensuring that the synthetic anomalies are both realistic and anatomically coherent. We evaluate our model's ability to detect pathology, using both synthetic anomaly datasets and real pathology from the ATLAS dataset. In our extensive experiments, our model: (i) consistently outperforms variational autoencoders, and conditional and unconditional latent diffusion; and (ii) surpasses on most datasets, the performance of supervised inpainting methods with access to paired diseased/healthy images.

Problem

• Detects brain anomalies without diseased training data
• Improves healthy image reconstruction using synthetic pseudo-pathologies
• Outperforms existing unsupervised and supervised anomaly detection methods

Innovation

• Conditional diffusion model for anomaly detection
• Fluid-driven anomaly randomization technique
• Weakly supervised synthetic pseudo-pathology integration