Semi-Supervised Anomaly Detection in Brain MRI Using a Domain-Agnostic Deep Reinforcement Learning Approach

2025-08-01

AI Summary

Brain MRI anomaly detection faces challenges including severe label scarcity, large-scale data, class imbalance, and overfitting. Method: This paper proposes a domain-agnostic semi-supervised deep reinforcement learning (DRL) framework—the first to integrate DRL into medical image anomaly detection—jointly optimizing feature representation learning and decision policies without requiring extensive annotations. Preprocessing includes intensity normalization, skull-stripping, and co-registration, supporting multi-modal T1- and T2-weighted MRI inputs. Contribution/Results: The method achieves 88.7% (pixel-level) and 96.7% (image-level) AUROC on brain MRI datasets, outperforming state-of-the-art approaches. Cross-domain evaluation on the MVTec AD industrial dataset yields 99.8% AUROC, demonstrating exceptional generalizability. Its core contribution is establishing the first transferable, annotation-efficient, and highly robust semi-supervised anomaly detection paradigm for medical imaging.

Technology Category

• Machine Learning: Semi-Supervised Learning
• Computer Vision: Adversarial Attacks & Robustness
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

To develop a domain-agnostic, semi-supervised anomaly detection framework that integrates deep reinforcement learning (DRL) to address challenges such as large-scale data, overfitting, and class imbalance, focusing on brain MRI volumes. This retrospective study used publicly available brain MRI datasets collected between 2005 and 2021. The IXI dataset provided 581 T1-weighted and 578 T2-weighted MRI volumes (from healthy subjects) for training, while the BraTS 2021 dataset provided 251 volumes for validation and 1000 for testing (unhealthy subjects with Glioblastomas). Preprocessing included normalization, skull-stripping, and co-registering to a uniform voxel size. Experiments were conducted on both T1- and T2-weighted modalities. Additional experiments and ablation analyses were also carried out on the industrial datasets. The proposed method integrates DRL with feature representations to handle label scarcity, large-scale data and overfitting. Statistical analysis was based on several detection and segmentation metrics including AUROC and Dice score. The proposed method achieved an AUROC of 88.7% (pixel-level) and 96.7% (image-level) on brain MRI datasets, outperforming State-of-The-Art (SOTA) methods. On industrial surface datasets, the model also showed competitive performance (AUROC = 99.8% pixel-level, 99.3% image-level) on MVTec AD dataset, indicating strong cross-domain generalization. Studies on anomaly sample size showed a monotonic increase in AUROC as more anomalies were seen, without evidence of overfitting or additional computational cost. The domain-agnostic semi-supervised approach using DRL shows significant promise for MRI anomaly detection, achieving strong performance on both medical and industrial datasets. Its robustness, generalizability and efficiency highlight its potential for real-world clinical applications.

Problem

• Detecting anomalies in brain MRI using semi-supervised DRL
• Addressing data scarcity and overfitting in medical imaging
• Enhancing cross-domain generalization for anomaly detection

Innovation

• Domain-agnostic DRL for anomaly detection
• Semi-supervised learning with limited labels
• Robust cross-domain generalization performance