Unsupervised anomaly detection using Bayesian flow networks: application to brain FDG PET in the context of Alzheimer's disease
This work addresses the problem of diagnosing Alzheimer's disease through neuroimaging for medical practitioners, representing an incremental advancement by applying a novel generative model to a specific domain.
The paper tackled unsupervised anomaly detection in brain FDG PET images for Alzheimer's disease by introducing AnoBFN, an extension of Bayesian flow networks, which outperformed state-of-the-art methods like beta-VAE, f-AnoGAN, and AnoDDPM in detecting anomalies and reducing false positive rates.
Unsupervised anomaly detection (UAD) plays a crucial role in neuroimaging for identifying deviations from healthy subject data and thus facilitating the diagnosis of neurological disorders. In this work, we focus on Bayesian flow networks (BFNs), a novel class of generative models, which have not yet been applied to medical imaging or anomaly detection. BFNs combine the strength of diffusion frameworks and Bayesian inference. We introduce AnoBFN, an extension of BFNs for UAD, designed to: i) perform conditional image generation under high levels of spatially correlated noise, and ii) preserve subject specificity by incorporating a recursive feedback from the input image throughout the generative process. We evaluate AnoBFN on the challenging task of Alzheimer's disease-related anomaly detection in FDG PET images. Our approach outperforms other state-of-the-art methods based on VAEs (beta-VAE), GANs (f-AnoGAN), and diffusion models (AnoDDPM), demonstrating its effectiveness at detecting anomalies while reducing false positive rates.