OpenAlex · Aktualisierung stündlich · Letzte Aktualisierung: 24.05.2026, 17:07

Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.

Analysis of article screening and data extraction performance by an AI systematic literature review platform

2025·0 Zitationen·Frontiers in Artificial IntelligenceOpen Access
Volltext beim Verlag öffnen

0

Zitationen

7

Autoren

2025

Jahr

Abstract

Background: Systematic literature reviews (SLRs) are critical to health research and decision-making but are often time- and labor-intensive. Artificial intelligence (AI) tools like large language models (LLMs) provide a promising way to automate these processes. Methods: We conducted a systematic literature review on the cost-effectiveness of adult pneumococcal vaccination and prospectively assessed the performance of our AI-assisted review platform, Intelligent Systematic Literature Review (ISLaR) 2.0, compared to expert researchers. Results: ISLaR demonstrated high accuracy (0.87 full-text screening; 0.86 data extraction), precision (0.88; 0.86), and sensitivity (0.91; 0.98) in article screening and data extraction tasks, but lower specificity (0.79; 0.42), especially when extracting data from tables. The platform reduced abstract and full-text screening time by over 90% compared to human reviewers. Conclusion: The platform has strong potential to reduce reviewer workload but requires further development.

Ähnliche Arbeiten