AI-Enabled Clinical Data Management
Street Art & Urban Creativity · 30.08.2024 · 10 (2) · 112-126
DOI: 10.66838/sauc.6558 ↗Kurz gefasst
Fachartikel in der Zeitschrift Street Art & Urban Creativity, online erschienen am 30.08.2024. Verfasst von Keerthi Manne. Der Volltext ist frei zugänglich (Open Access, CC BY-ND 4.0).
Abstract (Originalsprache)
Clinical data management (CDM) ensures that trial data are accurate, complete, and traceable, yet much of it still depends on manual cleaning: programmed edit checks, source data verification, query tracking, hand coding of medical terms, and a laborious path to database lock. Revised good clinical practice guidance and new regulatory principles for artificial intelligence (AI) encourage risk-proportionate, technology-enabled quality management. This paper proposes a framework for intelligent data operations that combines a maturity model, an architecture for AI-assisted detection, prioritization, and coding, a CRISP-DM-based lifecycle, and a credibility-based governance process, and it tests three components in a simulation. The simulation uses a synthetic trial of about 1,500 participants at 40 sites with about 84,000 data fields, 2% of which were corrupted, and a synthetic adverse-event dictionary; it demonstrates the method and does not report real trial data...
Zitieren
APA
Manne, K. (2024). AI-Enabled Clinical Data Management. Street Art & Urban Creativity, 10(2), 112-126. https://doi.org/10.66838/sauc.6558
Vancouver
Manne K. AI-Enabled Clinical Data Management. Street Art & Urban Creativity. 2024;10(2):112-126. doi:10.66838/sauc.6558
BibTeX
@article{Manne2024,
author = {Manne, Keerthi},
title = {AI-Enabled Clinical Data Management},
journal = {Street Art & Urban Creativity},
year = {2024},
volume = {10},
number = {2},
pages = {112--126},
doi = {10.66838/sauc.6558}
}
EUVIMED