Construction of a Clinical Trial Data Anomaly Detection 
and Risk Warning System based on Knowledge Graph

Yachen Wang

doi:10.70711/frim.v3i6.6661

Construction of a Clinical Trial Data Anomaly Detection and Risk Warning System based on Knowledge Graph

Yachen Wang

Abstract

With the surge in clinical trial data volumes, traditional biostatistical methods face challenges in handling data heterogeneity, highquality control, and real-time processing. This paper proposes an AI-driven data governance framework based on knowledge graphs, integrating domain knowledge modeling, semantic reasoning, and intelligent algorithms to achieve efficient integration and anomaly detection of
multi-source heterogeneous clinical data. Experiments show that this method improves anomaly recognition accuracy by 23.6% and reduces
Type I error rate by 41.2%. The study validates the potential of this framework in enhancing the intelligence level of clinical data governance
and ensuring the quality of new drug development data, promoting the upgrade of data management models.

Keywords

Knowledge graph; Clinical trial; Anomaly detection; Artificial intelligence; Data governance

Full Text:

PDF

Included Database

References

[1] Brbi? M, Yasunaga M, Agarwal P, Leskovec J.Predicting drug outcome of population via clinical knowledge graph.medRxiv

2024.03.06.24303800.

[2] Shi X, Du J. Constructing a finer-grained representation of clinical trial results from ClinicalTrials.gov.Scientific Data 2024;11(1):41.

[3] Zhang Y, Sui X, Pan F, [3] Zhang Y, Sui X, Pan F, et al. A comprehensive large-scale biomedical knowledge graph for AI-powered datadriven biomedical research.Nat Mach Intell 2025.

[4] Lin H, Wu X, Huang T. An Artificial Intelligence and Blockchain technology-based data management framework for multicenter randomized controlled trials.Sci Bull 2025;70(2):1-15.

[5] Bodenreider O. The Unified Medical Language System (UMLS): integrating biomedical terminology. Nucleic Acids Res 2021;49(D1):

D1232-D1239.

[6] Leskovec J, Sosi? R.Gremlin traversal machine: A distributed framework for fast graph query processing. Proc VLDB Endow

2016;9(12):1057-1068.

[7] ClinicalTrials.gov U.S. National Library of Medicine. https://clinicaltrials.gov

[8] OMOP Common Data Model Observational Health Data Sciences and Informatics (OHDSI).https://ohdsi.org

DOI: http://dx.doi.org/10.70711/frim.v3i6.6661

Refbacks

There are currently no refbacks.