Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Understanding Annotation Error Propagation and Learning an Adaptive Policy for Expert Intervention in Barrett's Video Segmentation

Lokesha Rasanjalee, Jin Lin Tan, Dileepa Pitawela, Rajvinder Singh, Hsiang-Ting Chen

TL;DR

This work systematically study how annotation errors propagate across different prompt types, namely masks, boxes, and points, and proposes Learning-to-Re-Prompt (L2RP), a cost-aware framework that learns when and where to seek expert input.

Abstract

Accurate annotation of endoscopic videos is essential yet time-consuming, particularly for challenging datasets such as dysplasia in Barrett's esophagus, where the affected regions are irregular and lack clear boundaries. Semi-automatic tools like Segment Anything Model 2 (SAM2) can ease this process by propagating annotations across frames, but small errors often accumulate and reduce accuracy, requiring expert review and correction. To address this, we systematically study how annotation errors propagate across different prompt types, namely masks, boxes, and points, and propose Learning-to-Re-Prompt (L2RP), a cost-aware framework that learns when and where to seek expert input. By tuning a human-cost parameter, our method balances annotation effort and segmentation accuracy. Experiments on a private Barrett's dysplasia dataset and the public SUN-SEG benchmark demonstrate improved temporal consistency and superior performance over baseline strategies.

Understanding Annotation Error Propagation and Learning an Adaptive Policy for Expert Intervention in Barrett's Video Segmentation

TL;DR

This work systematically study how annotation errors propagate across different prompt types, namely masks, boxes, and points, and proposes Learning-to-Re-Prompt (L2RP), a cost-aware framework that learns when and where to seek expert input.

Abstract

Accurate annotation of endoscopic videos is essential yet time-consuming, particularly for challenging datasets such as dysplasia in Barrett's esophagus, where the affected regions are irregular and lack clear boundaries. Semi-automatic tools like Segment Anything Model 2 (SAM2) can ease this process by propagating annotations across frames, but small errors often accumulate and reduce accuracy, requiring expert review and correction. To address this, we systematically study how annotation errors propagate across different prompt types, namely masks, boxes, and points, and propose Learning-to-Re-Prompt (L2RP), a cost-aware framework that learns when and where to seek expert input. By tuning a human-cost parameter, our method balances annotation effort and segmentation accuracy. Experiments on a private Barrett's dysplasia dataset and the public SUN-SEG benchmark demonstrate improved temporal consistency and superior performance over baseline strategies.
Paper Structure (13 sections, 3 equations, 2 figures, 1 table)

This paper contains 13 sections, 3 equations, 2 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Method
  4. Preliminaries
  5. Learning-to-Re-Prompt
  6. Experiment Setting
  7. Results and Discussion
  8. Difference in Prompt Types and Error Propagation
  9. Efficient Expert Intervention through L2RP
  10. Clinical Interpretation and Practical Use of $\lambda_{\text{corr}}$
  11. Conclusion
  12. Compliance with Ethical Standards
  13. Acknowledgments

Figures (2)

  • Figure 1: Example Barrett’s video showing mask propagation from the initial prompt and after an L2RP-suggested correction, with improved annotation quality.
  • Figure 2: Mean Dice Loss averaged across all videos at each frame, showing how segmentation error changes over time for different prompt types. Smoothed (20-frame window). Shaded regions indicate 95% confidence intervals.