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

2024-04-24

arXiv.org

AI Summary

To address the bottleneck of expensive pixel-level annotations (including hyperparameter tuning) in brain MRI anomaly segmentation, this paper proposes the first fully weakly supervised framework requiring only image-level labels. Methodologically, it introduces three key innovations: (1) leveraging the unguided forward process of diffusion models as a reference baseline for the guided forward process, enabling adaptive selection of noise scale, guidance strength, and segmentation threshold; (2) designing a multi-step forward anomaly map aggregation mechanism to enhance lesion response; and (3) establishing an image-level label-driven end-to-end optimization paradigm that jointly models unguided versus guided forward processes and fuses weighted anomaly maps. Evaluated on multiple brain MRI anomaly segmentation benchmarks, the method achieves significant improvements over existing weakly supervised state-of-the-art approaches—without any pixel-level supervision—while demonstrating superior localization accuracy and robustness.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Unsupervised & Self-Supervised Learning
• Search and Optimization: Learning to Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

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.

Problem

• Brain MRI
• Anomaly Detection
• Weak Supervision

Innovation

• AnoFPDM
• Weakly-Supervised Learning
• MRI Anomaly Detection