Research

Paper

AI LLM February 25, 2026

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

Authors

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

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.

Metadata

arXiv ID: 2602.21855
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-02-25
Fetched: 2026-02-26 05:00

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.21855v1</id>\n    <title>Understanding Annotation Error Propagation and Learning an Adaptive Policy for Expert Intervention in Barrett's Video Segmentation</title>\n    <updated>2026-02-25T12:30:54Z</updated>\n    <link href='https://arxiv.org/abs/2602.21855v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.21855v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>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.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-02-25T12:30:54Z</published>\n    <arxiv:comment>Accepted at IEEE ISBI 2026</arxiv:comment>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Lokesha Rasanjalee</name>\n    </author>\n    <author>\n      <name>Jin Lin Tan</name>\n    </author>\n    <author>\n      <name>Dileepa Pitawela</name>\n    </author>\n    <author>\n      <name>Rajvinder Singh</name>\n    </author>\n    <author>\n      <name>Hsiang-Ting Chen</name>\n    </author>\n  </entry>"
}