Reliable extraction of clinical-trial evidence with provenance, verification, and targeted review

Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Reliable extraction of trial characteristics and outcomes is essential for systematic reviews, yet incorrect values can alter evidence synthesis and are difficult to detect without clear source evidence. We evaluate EviSearch, a clinical evidence-curation system that links extracted values to source passages, independently verifies cited evidence, and prioritizes higher-risk cells for review. Using ten trial publications, a 133-column schema, reference annotations, and consistent scoring across model families, we assess answer quality, inter-reading agreement, evidence reproduction, targeted review, and evidence modality. In the Qwen evaluation, EviSearch achieved 91.70% accuracy, cited 100.0% of admitted values, and reproduced 98.9% independently. Reviewing 15.6% of cells increased accuracy to 95.22%.

Article activity feed

  1. Reliable extraction of clinical-trial evidence with provenance, verification, and targeted review https://www.biorxiv.org/content/10.64898/2026.10.01.756091v1

  2. Reliable extraction of clinical-trial evidence with provenance, verification, and targeted review https://www.biorxiv.org/content/10.64898/2026.10.01.756091v1