SentryLine: Evidence-Grounded Question Answering over Evolving Documents in Oncology Care

2026-09-08

AI Summary

研究解决了肿瘤学中快速变化的证据基础吸收问题，通过SENTYLINE系统使用无向量层次RAG管道检索指南段落并提供特定角色答案及验证报告。

Abstract

Oncology care operates at constant pressure of absorbing rapidly evolving evidence base in biomedicine. The American Society of Clinical Oncology (ASCO) addresses this through living guidelines, but the format introduces a new burden: any recommendation can change at any point, across multiple versioned documents. We present SENTRYLINE, a living guideline-aware clinical question answering system. SENTRYLINE retrieves guideline passages through a vectorless hierarchical RAG pipeline and returns a role-specific answer with inline citations, factual and temporal verification reports, and drift detection notes that surface when a guideline has been updated. We construct ASCOBENCH, a benchmark of 405 three-turn conversations across four question categories with gold answers from expert annotators(clinicians), and use test set to evaluate SENTRYLINE against five baselines under an LLM-as-judge framework. Experiments across three generation backbones show consistent improvements over four retrieval baselines and ASCO's guideline assistant, with particularly strong gains on Reasoning and Role-Specific questions where multi-hop synthesis and register adaptation are required

Problem

• Oncology Care
• Evolving Evidence Base
• Living Guidelines
• Dynamic Updates
• Clinical Recommendations

Innovation

• hierarchical RAG pipeline
• living guidelines
• temporal verification
• drift detection
• multi-hop synthesis