P243: Transforming Medical Review in Clinical Trials Through AI-Enabled Data Workflow Automation
Poster Presenter
Amit Patel
CEO
Octozi United States
Objectives
To evaluate how AI-enabled automation of clinical data workflows can improve the efficiency, consistency, and timeliness of medical review by reducing manual data preparation and enabling earlier identification of safety signals and data quality issues in clinical trials.
Method
A pilot implementation was conducted using Octozi, an AI-enabled clinical data workflow platform. The system ingested trial data from EDC and related sources, automatically generated review listings and signals, and supported medical monitors in continuous patient-level review across the study.
Results
Medical review workflows in many clinical trials rely on periodic manual preparation of patient listings, spreadsheets, and data summaries generated from EDC systems. These workflows require significant effort from data management and clinical operations teams and often introduce delays between data entry and clinical review.
This study evaluates the implementation of Octozi, an AI-enabled clinical data workflow platform designed to automate several steps in the medical review process. The platform continuously ingests incoming clinical data, generates standardized medical review listings, and applies configurable rules to identify predefined clinical signals and data anomalies. These automated outputs are designed to support medical monitors in reviewing patient-level data more efficiently and consistently.
The ongoing pilot focuses on assessing the operational impact of automation across several components of the medical review workflow, including time required to prepare review materials, frequency of medical review cycles, and the ability to surface predefined safety and data quality signals for clinical evaluation.
Quantitative results from the pilot are currently pending and will evaluate metrics such as reduction in manual data preparation time, improvements in timeliness of medical review, and consistency of signal identification across patient records. Preliminary qualitative feedback from study teams will also be assessed to understand how automation affects clinical workflows and reviewer experience.
Final results are expected to provide insight into how automated data workflows may support more scalable and proactive medical review processes within existing clinical trial infrastructure.
Conclusion
Automation of clinical data workflows has the potential to significantly improve the efficiency and scalability of medical review in clinical trials. By automating data ingestion, listing generation, and initial signal detection, platforms such as Octozi aim to reduce the operational burden associated with manual preparation of review materials and enable more continuous clinical oversight of trial data.
Rather than replacing the role of medical monitors, automated workflows are intended to augment clinical teams by handling repetitive data processing tasks and surfacing relevant signals for clinical interpretation. This approach may allow medical experts to focus more on clinical evaluation while maintaining visibility across large and complex datasets.
The ongoing pilot will provide quantitative and qualitative insights into the operational impact of automated medical review workflows. Findings are expected to inform best practices for integrating automation into existing clinical data ecosystems and to highlight areas where AI-enabled workflows may help improve efficiency, signal detection, and overall study oversight.
As clinical trials continue to increase in complexity and data volume, scalable approaches to medical review will become increasingly important. Results from this study may help demonstrate how automation can support more timely and consistent review of patient-level data while maintaining appropriate clinical oversight.
