Bayesian method improves brain scan anomaly detection without labels
Unsupervised Brain Anomaly Detection as a Bayesian Inverse Problem with Diffusion Prior
Artificial IntelligenceComputer Vision and Pattern Recognition
Summary
Detecting unusual areas in brain scans without detailed labels is challenging. The authors treat the problem like solving a puzzle backwards, using a special statistical method called Bayesian inference combined with diffusion models. They create a way to estimate both a cleaned-up healthy image and the spots with anomalies together, improving detection results. They tested this approach on brain PET and MRI scans and found better localization of anomalies than existing methods.
What this means in practice
- •For medical imaging teams: Enhance detection and localization of brain anomalies in PET and MRI scans without needing annotated data.
- •For healthcare ai developers: Develop advanced diagnostic tools that infer abnormal brain regions jointly with healthy anatomy using Bayesian diffusion models.
Authors
Hugues Roy, Reuben Dorent, Ninon Burgos
Abstract
Unsupervised anomaly detection (UAD) aims to localize abnormal regions in medical scans without pixel-level annotations. A typical strategy seeks to reconstruct a pseudo-healthy image that preserves subject-specific anatomy. Recently, diffusion models have been proposed to perform UAD. However, these methods rely on heuristic noise schedules or synthetic corruptions to balance subject-specificity and anomaly removal. In this work, we propose an alternative formulation of UAD as a Bayesian inverse problem under a diffusion prior. First, we introduce a latent spatial anomaly mask that models pixel-wise consistency between a test image and its latent corresponding pseudo-healthy image. Then, we propose an approximation of the unknown generation process that links healthy anatomy, anomalies, and the observed image, enabling a well-defined likelihood within the Bayesian framework. Building on recent advances in diffusion-based inverse problem methods, we jointly infer the pseudo-healthy image and the anomaly mask via annealed posterior sampling. We evaluate our approach on FDG PET (ADNI) and FLAIR MRI (BraTS 2021), demonstrating improved anomaly localization performance compared to other diffusion-based approaches and validating the contribution of our introduced model. Our code is available at https://github.com/HuguesRoy/UAD_DAPS.