Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

RASALoRE: Region Aware Spatial Attention with Location-based Random Embeddings for Weakly Supervised Anomaly Detection in Brain MRI Scans

Bheeshm Sharma, Karthikeyan Jaganathan, Balamurugan Palaniappan

TL;DR

This work proposes RASALoRE: Region Aware Spatial Attention with Location-based Random Embeddings, a novel two-stage WSAD framework that achieves state-of-the-art anomaly detection performance, significantly outperforming existing WSAD methods while utilizing less than 8 million parameters.

Abstract

Weakly Supervised Anomaly detection (WSAD) in brain MRI scans is an important challenge useful to obtain quick and accurate detection of brain anomalies when precise pixel-level anomaly annotations are unavailable and only weak labels (e.g., slice-level) are available. In this work, we propose RASALoRE: Region Aware Spatial Attention with Location-based Random Embeddings, a novel two-stage WSAD framework. In the first stage, we introduce a Discriminative Dual Prompt Tuning (DDPT) mechanism that generates high-quality pseudo weak masks based on slice-level labels, serving as coarse localization cues. In the second stage, we propose a segmentation network with a region-aware spatial attention mechanism that relies on fixed location-based random embeddings. This design enables the model to effectively focus on anomalous regions. Our approach achieves state-of-the-art anomaly detection performance, significantly outperforming existing WSAD methods while utilizing less than 8 million parameters. Extensive evaluations on the BraTS20, BraTS21, BraTS23, and MSD datasets demonstrate a substantial performance improvement coupled with a significant reduction in computational complexity. Code is available at: https://github.com/BheeshmSharma/RASALoRE-BMVC-2025/.

RASALoRE: Region Aware Spatial Attention with Location-based Random Embeddings for Weakly Supervised Anomaly Detection in Brain MRI Scans

TL;DR

This work proposes RASALoRE: Region Aware Spatial Attention with Location-based Random Embeddings, a novel two-stage WSAD framework that achieves state-of-the-art anomaly detection performance, significantly outperforming existing WSAD methods while utilizing less than 8 million parameters.

Abstract

Weakly Supervised Anomaly detection (WSAD) in brain MRI scans is an important challenge useful to obtain quick and accurate detection of brain anomalies when precise pixel-level anomaly annotations are unavailable and only weak labels (e.g., slice-level) are available. In this work, we propose RASALoRE: Region Aware Spatial Attention with Location-based Random Embeddings, a novel two-stage WSAD framework. In the first stage, we introduce a Discriminative Dual Prompt Tuning (DDPT) mechanism that generates high-quality pseudo weak masks based on slice-level labels, serving as coarse localization cues. In the second stage, we propose a segmentation network with a region-aware spatial attention mechanism that relies on fixed location-based random embeddings. This design enables the model to effectively focus on anomalous regions. Our approach achieves state-of-the-art anomaly detection performance, significantly outperforming existing WSAD methods while utilizing less than 8 million parameters. Extensive evaluations on the BraTS20, BraTS21, BraTS23, and MSD datasets demonstrate a substantial performance improvement coupled with a significant reduction in computational complexity. Code is available at: https://github.com/BheeshmSharma/RASALoRE-BMVC-2025/.

Paper Structure

This paper contains 9 sections, 1 equation, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Proposed Methodology of RASALoRE
  3. Discriminative Dual Prompt Tuning (DDPT)
  4. Region Aware Spatial Attention with Location-based Random Embeddings (RASALoRE)
  5. Multimodality RASALoRE
  6. Related Works
  7. Experiments and Results
  8. Conclusion
  9. Acknowledgment

Figures (4)

  • Figure 1: Overview of Discriminative Dual Prompt Tuning (DDPT)
  • Figure 2: Overview of RASALoRE Architecture
  • Figure 3: (a) Left: Candidate prompt point locations (in blue) overlaid as grid on input image, center: point activation mask (red denoting active and blue denoting inactive points) overlaid on input image, right: weak anomaly mask corresponding to input image. (b) Refiner Module.
  • Figure 4: Qualitative Comparison of Predicted Anomaly Mask from Different Methods. Abbreviations: M+DDPT = MedSAM+DDPT(box), M2+DDPT = MedSAM2+DDPT(box) and R.-MedSAM = RASALoRE without MedSAM.