Standards-Driven and AI-Driven Automation in Clinical Data Flows
Publicerad
Författare
Typ
Examensarbete för masterexamen
Master's Thesis
Master's Thesis
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
This study explores how standards-driven and AI-driven automation can streamline
the clinical data flow at AstraZeneca (AZ) and improve interoperability within
the healthcare industry. As clinical data volumes and complexity increase, efficient
management of data is critical for accelerating drug development while ensuring
robust clinical evidence. The study maps AZ’s current clinical data flow
and identifies opportunities for automation and standardization. A qualitative research
method was applied, using semi-structured interviews with AZ employees
and an internal documentation review. Process mapping and thematic analysis
were conducted to generate a Data Flow Diagram, extract key insights, and deliver
recommendations. Findings show that the current data flow involves several manual
steps across data collection, management, statistical analysis, and regulatory
submission. Key areas for improvement include stricter metadata requirements, a
centralized data flow, and detailed workflow documentation to support automation.
The study also highlights the potential of machine-readable specifications and executable
metadata, aligned with the Clinical Data Interchange Standards Consortium
360Implementation (CDISC 360i) program, to enable platform-based, streamlined
data flows. Prioritizing critical-to-quality data in study protocols and shifting to
structured, data-centric regulatory submissions are recommended. Several AI opportunities
were identified, including digital twins for virtual control arms, natural
language processing for regulatory text generation, and agentic AI for end-to-end
automation. These require strong quality control, human oversight, explainability,
and regulatory acceptance. The study concludes that by proactively adopting
these recommendations, AZ can improve internal efficiency, support a connected and
data-driven healthcare ecosystem, and help lead the transformation toward faster,
higher-quality drug development and timely patient access to innovative treatments.
Beskrivning
Ämne/nyckelord
Clinical Data Flow, Standards-Driven Automation, AI-Driven Automation, Interoperability, Regulatory Submission, CDISC Standards, HL7 FHIR, Digital Health, Clinical Trials, Pharmaceutical Data Management
