OpenAlex · Aktualisierung stündlich · Letzte Aktualisierung: 19.03.2026, 22:01

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

The performance boundaries of knowledge prompts: A study on injection strategies and scale effects for ICD coding tasks in large language models

2026·0 Zitationen·Digital HealthOpen Access
Volltext beim Verlag öffnen

0

Zitationen

9

Autoren

2026

Jahr

Abstract

An integrated framework combining contextual prompting with DeepSeek-V3 substantially improves automated ICD coding accuracy and efficiency, demonstrating strong potential for clinical application.

Ähnliche Arbeiten

Autoren

Institutionen

Themen

Machine Learning in HealthcareMedical Coding and Health InformationArtificial Intelligence in Healthcare and Education
Volltext beim Verlag öffnen