CVApr 24, 2024

AnoFPDM: Anomaly Segmentation with Forward Process of Diffusion Models for Brain MRI

arXiv:2404.15683v42 citationsh-index: 13
Originality Highly original
AI Analysis

This addresses the cost and effort of pixel-level labeling in medical imaging for anomaly detection, representing a significant improvement over existing weakly-supervised approaches.

The paper tackles the problem of anomaly segmentation in brain MRI by introducing AnoFPDM, a fully weakly-supervised framework that eliminates reliance on pixel-level labels for hyperparameter tuning, and it outperforms recent state-of-the-art weakly-supervised methods.

Weakly-supervised diffusion models (DMs) in anomaly segmentation, leveraging image-level labels, have attracted significant attention for their superior performance compared to unsupervised methods. It eliminates the need for pixel-level labels in training, offering a more cost-effective alternative to supervised methods. However, existing methods are not fully weakly-supervised because they heavily rely on costly pixel-level labels for hyperparameter tuning in inference. To tackle this challenge, we introduce Anomaly Segmentation with Forward Process of Diffusion Models (AnoFPDM), a fully weakly-supervised framework that operates without the need of pixel-level labels. Leveraging the unguided forward process as a reference for the guided forward process, we select hyperparameters such as the noise scale, the threshold for segmentation and the guidance strength. We aggregate anomaly maps from guided forward process, enhancing the signal strength of anomalous regions. Remarkably, our proposed method outperforms recent state-of-the-art weakly-supervised approaches, even without utilizing pixel-level labels.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes