Smart Data Quality

Smart Data Quality (SDQ) leverages advanced AI models to streamline the data review processes in clinical trials, enabling study teams to efficiently manage the vast volumes and diverse types of data encountered. By automating these processes, SDQ significantly reduces the time required to generate queries from 30 minutes to just 3 minutes per query, enhancing the speed and accuracy of data handling.

SDQ offers several key benefits, including the automation of data review processes which eliminates the need for manual reviews, thereby reducing errors and minimizing delays in trials. It accelerates the time to database lock by maintaining clean data as it is collected, and it automatically identifies data discrepancies to generate queries swiftly. The system is scalable across various portfolios, supported by a cloud-based architecture on AWS, and has been proven effective in large, global trials.

Traditional manual data review processes typically require 47 days and three resources, taking 27 minutes to review and write a query, resulting in 2,615 queries and 1,177 hours of work. In contrast, SDQ's automated process reduces this to 16 days with just one resource, taking only 3 minutes to review and approve AI-generated queries, maintaining the same number of queries but reducing the workload to 130 hours.

Features

• AI-assisted data reviews: Advanced AI models automatically detect data discrepancies and generate predefined query text, which would usually require manual intervention.
• Interactive review listings: Users can manually review data listings and perform advanced data reviews in one location, with the ability to create custom listings using generative AI and assign tasks to team members and vendors.
• Integrated rule builder: A self-service rule builder allows users to code data quality checks directly within SDQ, which can be reused across studies and work alongside AI-driven checks.
• Data quality Co-Pilot: Users can describe desired data quality checks, and SDQ will automatically write and test the code using generative AI trained on historical data.
• Discrepancy management: Track, review, and manage all queries in one location, regardless of their source.
• Blinded/Unblinded integration: Automatically captures data masking configurations to ensure secure and precise data management.
• Clean patient tracker: Monitors data-cleaning status and missing data at the patient level, with configurable and reusable milestone readiness.
• Query anomaly detection: Utilizes generative AI to uncover unknown data patterns and anomalies that might otherwise be missed.

SDQ is designed to accelerate clinical trials while ensuring high-quality, clean data, making it an essential component for modern data management in life sciences.